Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Organizational injustice and its resistance using voice and silence Harlos, Karen P. 1997

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1998-271544.pdf [ 11.18MB ]
Metadata
JSON: 831-1.0088786.json
JSON-LD: 831-1.0088786-ld.json
RDF/XML (Pretty): 831-1.0088786-rdf.xml
RDF/JSON: 831-1.0088786-rdf.json
Turtle: 831-1.0088786-turtle.txt
N-Triples: 831-1.0088786-rdf-ntriples.txt
Original Record: 831-1.0088786-source.json
Full Text
831-1.0088786-fulltext.txt
Citation
831-1.0088786.ris

Full Text

ORGANIZATIONAL INJUSTICE A N D ITS RESISTANCE USING VOICE A N D SILENCE by . K A R E N P. HARLOS B. A. , University of British Columbia, 1982 M . A. , University of British Columbia, 1993 A THESIS SUBMITTED IN PARTIAL F U L F I L L M E N T OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE F A C U L T Y OF G R A D U A T E STUDIES (Faculty of Commerce and Business Administration) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH C O L U M B I A December 1997 © Karen P. Harlos, 1997 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of £X>\A\A!c2C& The University of British Columbia Vancouver, Canada Date DE-6 (2788) ABSTRACT This research is based on the premise that employees respond to dissatisfaction in general and organizational injustice in particular in two primary ways: by speaking up and/or by staying silent. This qualitative, theory-generating study examines the phenomenon of organizational injustice (including its antecedents and consequences) and employees' responses toward three research goals: 1) greater understanding of organizational injustice; 2) greater conceptual consensus through concept development of voice and silence; 3) a process model of organizational injustice, voice and silence. Also, new knowledge about voice and silence is linked to organizational practice by examining the availability of various voice systems and perceptions of their efficacy. The research design is influenced by several organizational research streams, as well as grounded theory and clinical methods. Thirty-two employees, each representing different organizations and occupying both managerial/professional positions and clerical/line positions participated in semi-structured, open-ended interviews in which they described 33 cases of workplace injustice. The interview design includes two methods: 1) a retrospective critical incident technique to discuss a workplace experience which participants defined as unjust; and 2) a projective exercise in which participants were asked to imagine that they could speak with impunity to the person(s) involved or responsible for their perceived injustice. Interview cases were supplemented by 30 archival cases of employees' voicing of discontent through a government-sponsored voice system. Significant results concerning the phenomenon of organizational injustice included the introduction of a four-category typology which departs from traditional classifications with its inclusion of interactional injustice (interpersonal mistreatment by a boss) as a distinct category, the systematic delineation and description of interactional injustice according to eight emergent behavioural dimensions, the identification of organizational antecedents to workplace injustice according to four emergent groupings (i.e., structural, procedural, cultural and global) and the identification of individual- and organizational-level consequences. In addition, the concepts of voice and silence emerged as forms of resistance to organizational injustice. Voice was found to encompass two distinct but related constructs: formal and informal voice. Specific strategies by which participants resisted injustice were identified for voice (formal and informal) and silence. A process model of voice and silence in organizational injustice was also introduced. i i i T A B L E O F C O N T E N T S Page A B S T R A C T i i T A B L E OF CONTENTS iv LIST OF TABLES vi LIST OF FIGURES vii A C K N O W L E D G M E N T S viii CHAPTER I: INTRODUCTION A N D LITERATURE REVIEW A. Background: What to Study, Why and How 1 B. Relevant Research Streams 9 CHAPTER II: R E S E A R C H METHODS A. Overview 47 B . Theories of Method 49 C. Procedure 53 D. Data Analysis 67 CHAPTER III: INDIVIDUAL-LEVEL EXPERIENCES OF ORGANIZATIONAL INJUSTICE A. Overview 73 B. A Typology of Organizational Injustice 75 C. Properties of Organizational Injustice 99 D. Antecedents to and Consequences of Organizational Injustice 110 Chapter Summary 114 CHAPTER IV: ORGANIZATIONAL-LEVEL ASPECTS OF ORGANIZATIONAL INJUSTICE A. Organizational Antecedents to and Consequences of Workplace Injustice 115 B. Characteristic Patterns: Images and Themes 127 Chapter Summary 139 CHAPTER V: RESISTING ORGANIZATIONAL INJUSTICE WITH VOICE A N D SILENCE A. Identifying Organizational Injustice 142 B. Voice and Silence as Strategies to Resist Organizational Injustice 146 C. Projective Exercise Results 163 D. Voice Systems In/Action 166 iv Chapter Summary 172 CHAPTER VI: DISCUSSION A N D CONCLUSIONS A. Overview of Significant Findings 173 B. The Phenomenon of Organizational Injustice 175 C. Conceptualizations of Voice and Silence 192 D. A Process Model of Voice, Silence and Organizational Injustice 202 E. Conclusions 204 REFERENCES 208 APPENDIX A: Interview Protocol 221 APPENDIX B: Recruitment Letter To Potential Participants 224 APPENDIX C: Copy Of Newspaper Article For Recruitment 225 APPENDIX D: List Of Available Voice Systems 226 APPENDIX E: Consent Form 227 APPENDIX F: Demographic Information 228 APPENDIX G: Participant Pseudonyms And Job Titles 229 APPENDIX H: Incidence Of Categories Of Organizational Injustice 230 APPENDIX I: Training Manual 231 APPENDIX J: Inter-Rater Reliability 241 v L I S T O F T A B L E S Page Table 1. Summaries of Relevant Research Streams 10 Table 2. Position Titles of Research Participants by Industry, Hierarchical Level and Gender 61 Table 3. A Typology of Organizational Injustice 77 Table 4. Behavioural Dimensions of Interactional Injustice 86 Table 5. Consequences of Organizational Injustice for Individuals 112 Table 6. Organizational Antecedents of Organizational Injustice 116 Table 7. Consequences of Organizational Injustice for Organizations 128 Table 8. Typology of Voice and Silence as Resistance Strategies 148 v i LIST OF FIGURES Page Figure 1. The Voice-Silence Continuum of Strategies to Resist Organizational Injustice 198 Figure 2. A Process Model of Voice, Silence and Organizational Injustice 203 vii ACKNOWLEDGMENTS I would like to thank my committee members — Dr. Craig Pinder, Mr. David McPhillips and Dr. Elvi Whittaker — for their generous contributions to the successful completion of my dissertation. As my advisor, Craig deserves particular thanks for the intellectual guidance and emotional support he provided with apparent ease throughout my doctoral education, from facilitating my crossing from Psychology to Commerce to preparing me for a larger one to New Zealand. His beliefs and behaviour personify McGregor's elusive Theory Y of motivation: that people basically like hard work and responsibility, that they will achieve their greatest potential if provided with a rewarding work environment, and most fundamentally, that students are people too. David's legal expertise, combined with a direct and open personal style, made asking questions and seeking advice a pleasure while Elvi's gentle guidance into the land of qualitative methodology made the journey enjoyable (dare I say fun!) and fruitful. Financial assistance was provided by the Social Sciences and Humanities Research Council and, in various ways and times, by the U B C Faculty of Commerce. At one Faculty-funded conference in 1993, a serendipitous meeting with Robert Folger steered me to Ronald Cohen, whom he called "the conscience of organizational justice." Ronald's provocative ideas about voice, silence and fairness were seminal to my thinking. Two years later, a brief but equally fortunate meeting with Bob Quinn gave me the necessary encouragement to keep looking in the dark corners. The Faculty also provided support for my participation as a finalist in the 1996 INFORMS College of Organization Science Best Dissertation Proposal Competition in Atlanta, Georgia. In our Faculty, I am also grateful to Dr. Tom Knight, past Director of The Centre for Labour and Management Studies, for providing me with the opportunity, encouragement and resources to learn about organizational (injustice. Thanks also to now-Northwesterner Dr. J. Keith Murnighan for helpful suggestions following my proposal. Also, assistance from Pat Shanahan, Associate Director of the Master's Program, was invaluable and Gary Schwartz of Technological Services was an equipment wizard. Thank you to these individuals and institutions for their significant financial and intellectual support. Chris Gadsby of the British Columbia Institute of Technology, Part-Time Studies in Administrative Management Diploma Program, kindly gave me access to several classes for a recruitment pitch; instructors Rhonda Margolis, Starr Owen and Dawne Bringeland generously made class time available for me. Gudrun Wills, a reporter with the Vancouver Courier, concisely captured my research purpose and approach in an entertaining and informative way; her article resulted in many calls and a significant number of interviews. I am deeply indebted to the participants of this study who so freely gave their time and of themselves toward a better understanding of organizational injustice. Confidentiality concerns preclude me from naming them, but their contributions were critical not only to the completion of this research but also to its legitimacy and credibility. My unwavering gratitude goes to my parents Roland and Shary for nurturing my sense of curiousity about the world and what has been written about it. My mother's love of reading, including Saturday morning trips to the Burnaby Public Library, my father's love of debate, and the encyclopedia set in the family room — chewed by children and dogs alike — were formative influences on what Toril Moi calls 'the making of an intellectual woman.' Thank you is so very inadequate to acknowledge the impact of my parents' resolute faith in V l l l my ability, their pride in my accomplishments, unending encouragement and their absolute acceptance throughout my life. Lastly, a lifetime of gratitude to Cy-Thea for demonstrating creative ways to show support, which included enduring interest in my thinking, incisive editing of my writing, sometimes saintly patience with the rainbow of emotions that accompany the dissertation process and calming comfort for my spirit. I look forward to reciprocating these gifts in kind over the years. ix CHAPTER I INTRODUCTION AND LITERATURE REVIEW A. Background: What to Study, Why and How In the first decade since its introduction by Hirschman (1970), the three-choice framework of responses to dissatisfaction (exit-voice-loyalty) generated modest, intermittent interest. The framework was grounded largely in economics and focused on organizational customers rather than organizational members. In recent years however, this work has enjoyed increased and enduring attention in the organizational behaviour literature as researchers have focused on organizational members and expanded the number of response options (Farrell, 1983; Robinson, 1993; Rusbult, Zemboldt & Gunn, 1982), explored its predictive potential (e.g., Withey & Cooper, 1989) and applied the framework, in whole or part, to organizational functioning (Freeman & Medoff, 1984; Lewin & Boroff, 1994). In particular, the application of Hirschman's concept of voice to organizational justice theory has generated much research interest. Voice — broadly defined as a constellation of active change efforts — is now seen as a critical element of procedural fairness and, more broadly, of organizational justice (Sheppard, Lewicki & Minton, 1992). Perceptions of procedural and organizational justice, in turn, lead to increased legitimacy being accorded to organizations by members and their greater compliance with authority (Tyler, 1990). Research also suggests that the use of voice systems (e.g., suggestion box, grievance procedure, open door policy) results in greater employee satisfaction (Gorden, Infante & Graham, 1988), improved loyalty and commitment to the organization (Sheppard et al., 1992) and reduced turnover (Lind & Tyler, 1988; Tyler, 1989). 1 Boundaries of Current Knowledge Despite the large literature concerning justice at work, little is known about organizational injustice. Few studies, for example, are conducted in organizational settings where unfairness, exploitation or abuse are either idiosyncratic or the norm. Moreover, several serious problems affect Hirschman's original framework and subsequent models developed by later researchers. Of particular concern is the neglect of employees who stay in the organization and remain silent. Ironically, as Kolarska and Aldrich (1980) observe, silence is probably the most common but least understood response expressed by dissatisfied employees. Extant models incorporating voice and silence report weak construct validity of relevant measures; existing constructs likely fail to capture the conceptual complexity of voice and silence (Withey & Cooper, 1989). In turn, it is probable that conceptual limitations partially account for the difficulty interpreting and predicting voice and silence. In addition, despite their integral connection, few theoretical or empirical efforts link voice as used in Hirschman's (1970) model on the one hand, to voice as used in organizational justice research on the other (Lewin & Boroff, 1994). Moreover, organizational justice research on voice focuses on its role in the positive effects of procedural fairness but neglects its role in procedural tmfairness (Greenberg, 1990b). Furthermore, it is not widely-recognized that both voice and silence may indicate either justice or injustice (Cohen, 1990). It is both evident and curious that in organizational (injustice research many questions remain both unasked and unanswered about voice and silence, thereby limiting theory development and practical contributions. At present, little is known about the antecedents, processes and consequences of voice and silence as responses to perceived injustice. In addition, the relationship between 2 employees' perceptions of voice systems and their actual use has been neglected by organizational justice theories which assume that procedures to review and redress employees' discontent benefit both the individual and organization (Lewin & Boroff, 1994). In the present research, I explore the consequences associated with the use of voice systems for employees and for the organization, as well as perceptions of voice system efficacy. Additional questions addressed by this research include: who uses voice systems and why? What are the experiences of people who express injustice-induced discontent through voice systems? Some employees forgo the use of voice systems, despite their availability, and remain silent. Who stays quiet, why and with what results? In exploring these questions, this research extends Staw (1984) by addressing how organizational injustice can be used for individual and organizational improvement. By studying a context where deep feelings of injustice can emerge, this research counters the prevailing focus on fairness and thus is well-positioned to respond to Martin's (1993) implicit call to explore possible boundary conditions of organizational injustice (points at which no organizational practice or procedure would reduce discontent). In sum, this research extends the boundaries of current knowledge of organizational injustice and responses to it by studying employees' deeply-felt experiences of injustice, by developing relevant concepts and by considering how injustice can be used for individual and organizational improvement. Some Alternate Assumptions Throughout the relevant literature(s), several substantive assumptions remain untested. For example, organizational justice theory and practice implicitly assume that all employees enjoy equal access to and facility with voice systems. This untested assumption of legitimacy (i.e., that employees are believed to perceive organizational policies and/or 3 practices as right and just, regardless of differential benefits or relevance; Martin, 1993)1, supports an implicit and untested conclusion that those who decline to express their discontent have no complaint. In this way, silence is interpreted as universal and consensual endorsement (Cohen, 1990) despite the fact that assumptions about a homogeneous workforce are now obsolete (e.g., lies & Aulick, 1991; Nkomo, 1992; Kitzinger, 1991). In this research, I reject traditional, untested assumptions of the unidimensionality of voice and silence and explore, for the first time, their potential multidimensionality through interviews with employees who report dissatisfaction in the form of perceived injustice at work. Also, contrary to previous efforts, this research rejects the implicit assumption that voice and silence are static endpoints of dissatisfaction. Instead, the present study considers temporal and situational aspects of responses to dissatisfaction by exploring the antecedents and consequences of organizational injustice and responses to it, namely voice and silence. Further Rationale Several additional factors encourage a better understanding of the phenomenon of organizational injustice and employees' reactions to perceived mistreatment at work. Significance. One widespread (if implicit) assumption in organizational research is that dissatisfaction in general and injustice in particular abound. Thus, given its presumed ubiquity, perceived injustice may be expected to have profound significance for all levels of organizational functioning and membership. Recent findings support such speculation, with reported reactions to perceived injustice ranging from feelings of anger and resentment (Folger, 1987, 1993) to hostile or destructive actions against the self (e.g., drug dependency; 1 This definition of legitimacy from Martin (1993) is more rigorous than the traditional one which simply asks whether organizational policies or practices are uniform and widely-applied. 4 Martin, 1993), others (e.g., organizationally-targeted insider murders; Allen & Lucerno, 1996) or corporate property (e.g., sabotage; Jermier, 1988; LaNuez & Jermier, 1994). i Costs. Given the association between perceived injustice and aggression, the over 1200 acts of workplace violence reported to the Workers' Compensation Board (WCB) of British Columbia (BC) in 1994 are noteworthy. Together, BC employers and the W C B paid j over $5.5 million in claims to cover the wages, medical and support costs of the 57,000 days of work lost to violence reported in that year alone. Moreover, North American trends to include stress as a compensable claim for workers' compensation make it an emerging but significant | source of financial liability (Wayne Corneil, Health and Welfare Canada, personal communication, November 25, 1991; Ralph Barrows, B C Ministry of Labour, personal communication, May 6, 1997). While specific costs from workplace injustice have yet to be i calculated,!research indicates that perceptions of managerial unfairness are an emerging i source of stress-related disability claims (Bies, 1987). The present study continues and expands previous research by exploring injustice-induced reductions in productivity, increased turnover, greater resistance to organizational initiatives and inefficient use of organizational resources as a preliminary step to future assessments of the financial i implications of perceived injustice. Human dignity issues. While dissatisfaction with our work, colleagues or pay may be unfortunate features of our organizational experiences, freedom from more severe or extreme mistreatment or unfairness from individuals, policies or procedures constitutes a basic right 2 These figures represent only post-violence costs — in many cases, the period before violence erupts contains hidden costs created by the increasing tensions and stresses in workplace relationships which can result in greater absenteeism, for example. , I At the timelof writing, the question of stress as a compensable claim is under review by a B.C. Royal Commission. 5 i i enshrined in Canada, in provincial and federal human rights legislation and fair wage legislation, for example. In this research I take as self-evident that employees' entitlement to a workplace free from acts of injustice is a basic humanitarian as well as legal issue. Thus, the spiritual (i.e., of the human spirit) significance of workplace injustice is a factor I also observe amid necessary economic and, to a lesser degree, legal considerations. Research Purpose and Design The limits to current knowledge (outlined more fully in Section B) suggest that the following three research goals will foster new knowledge: 1) description of the phenomenon of organizational injustice; 2) conceptual consensus through concept development of voice and silence; 3) the introduction of a process model of organizational injustice, voice and silence. In pursuing these goals, this research is also guided by the following research questions: 1. How do employees describe and interpret organizational injustice and their responses to it, particularly voice and silence? 2. What are the antecedents, processes and consequences of organizational injustice and of voice and silence as responses? 3. How do contextual conditions (i.e., characteristics of the situation and organization) influence employees' experiences of and responses to organizational injustice? Qualitative Inquiry: Why and How Previous research relating voice and silence to dissatisfaction and justice at work has primarily relied on quantitative approaches to predict responses to dissatisfaction and to identify determinants of justice, for example. Despite its volume, the predictability of voice and silence remains problematic and determinants of injustice remain largely unknown. 6 Together, these limitations and the research goals and questions identified above point to the use of qualitative inquiry to generate new knowledge of the phenomenon of organizational injustice and the role of voice and silence in responding to it. Qualitative methods are well-suited to reveal possible multidimensionality of concepts, to uncover tacit processes in phenomena (Morse, 1994) and to identify relevant variables (Marshall & Rossman, 1995). Specifically, because this research is concerned with understanding meanings or c u l t u r a l s e n s e m a k i n g within a context of organizational injustice by those who have experienced it, the specific form of qualitative inquiry I use is ethnography. The research approach followed here is consistent with the first two of eight steps toward concept development in organizational behaviour outlined by Greenberg (1992): 1) consideration is given to ways the concepts of interest are explained in the literature; 2) their consensual meaning is empirically established through interviews with lay persons. Consequently, this research systematically examines through qualitative inquiry the phenomenon of organizational injustice and employees' responses to it as processes. Thus, this study takes its place in an emerging organizational research tradition which relies on qualitative methods exclusively (e.g., Brown & Eisenhardt, 1997; Gersick, 1988; Kanter, 1977) or in part (e.g., Elsbach, 1994). More broadly, this research is an example of naturalistic inquiry which typically relies on field rather than experimental settings, purposive rather than random sampling and an emergent research design and inductive data analysis rather than a predetermined design and deductive analysis (Lincoln & Guba, 1985). Specifically, I use grounded theory (Glaser & Strauss, 1967; Strauss, 1987) to generate knowledge about organizational injustice, to refine the concepts of voice and silence and to construct a process model of organizational injustice. 7 Thus, unlike previous research, this approach allows the contribution of naturalistic, qualitative inquiry to the study of employees' experiences of and responses to workplace injustice to be assessed. Summary of Contributions This research adds to the accumulated literature in several ways. First, because no single stream of theoretical or empirical research provides an adequate frame for the present work, it links several historically disparate literatures for the first time. Secondly, emergent findings are integrated with both popular and academic literatures to address the current paucity of systematic descriptive and analytical accounts of and frameworks for organizational injustice. Third, concept development of voice and silence contributes to future efforts to improve the construct validity of measures of voice and silence and predictability of relevant variables. Fourth, I introduce a process model of organizational injustice which includes voice and silence as resistance strategies and which considers their respective roles in resolving, maintaining and exacerbating injustice. Lastly, this research examines perceptions of the efficacy of various voice systems and explores which two systems have the greatest potential to encourage employees' expression of discontent. Overview of the Document The next section of this chapter provides a review of relevant research streams. In Chapter 2,1 outline the research methods used to pursue the goals and questions identified earlier. Findings are reported in the next three chapters: in Chapter 3,1 describe and examine individual-level aspects of the phenomenon of workplace injustice, noting commonalities and unique features; in Chapter 4,1 examine organizational-level aspects of organizational injustice; in Chapter 5,1 analyze participants' responses to organizational injustice and 8 develop the concepts of voice and silence. I also report in Chapter 5 participants' assessments of organizational voice systems. In the sixth and final chapter, I discuss emergent findings in relation to extant theory and consider their implications. I also introduce a process model by which organizational injustice is resisted using voice and silence. Consistent with previous research (e.g., Sheppard et al., 1992) and the language of laypersons interviewed in this study, I use the terms 'injustice' and 'unfairness' interchangeably, as I do justice' and 'fairness.' As well, the terms organizational injustice and workplace injustice are considered synonymous. Moreover, I recognize that issues of injustice (and justice) are inherently perceptual and thus subjective. Accordingly, the focus of this study is on participants' current perceptions of injustice rather than the actuality of past events or their legal interpretation. Given this assumption and focus, the adjective 'perceived' is sometimes omitted as a descriptor for 'injustice' or justice' to enhance the readability of this document. B. Relevant Research Streams Several research streams offer direct or indirect knowledge of organizational injustice and employees' responses to workplace unfairness. One stream of research focuses broadly on justice and injustice in organizations while a second examines Hirschman's (1970) responses to dissatisfaction, focusing on their measurement and prediction. Additional literatures relevant to the present research (although to a lesser degree) include the whistle-blowing and organizational power literatures. Table 1 summarizes the research streams guiding this study, including relevant assumptions, findings and future lines of inquiry in each of these streams. 9 c o s o Z 3 ;=? "ca c _o C3 N '8 ca 6 0 <~ .a, s i o ° 3 oo tU C T3 C C3 ca C O C Ci oo o <u o o s o o o c oo tu C 3 " 2 o 3 tu , .§ § * a 2 oo ii « .s .a •§ _ c u oo op 00 •° o c 5 ca O <U "3 o 2 "S ° > 3 c n o b M 7 3 S u O co \ 3 C^ .s tu o c o "o M T3 c C3 >-» oo • M D 0 0 , 2 G on S T3 tu > X> o <D ^> o ca C CT ID oo C O u T 3 C C CU M & o a. tu > '5 o M u ^ o a o ia C M ID 6 0 cA c *<s o i o T 3 c cu 0 0 _e OO '> -> o o J 3 c ca ca o s C M ID CD O c tu CU > X> C O oo W O0 tU ID > ~ o a . o ca e WO a CU 0 0 tU 5 c u o M tu o > cu s • i : CD -a, s s s 2 c ^ "3 ""- c I ° £ « c 2 cu 1 oo C 3 T3 1) ON OO OS I " S3 — o ° > U On tu o c C M O oo C 3 o •a ^ (U tu J= o £ > tu 3 a o S 6 o o cd c/l tu o .6 S3 « c o 3 O" — U ccJ o i ^ O C o a S 9 t« Vi < > cu 05 T3 tU o tu c tu oo C O o c tu M oo >^-§ = & S 6 2 - <t> CT , O 3 (U 5 8 is o 03 M oo tu E S tu • M OO O0 O0 tU oo o <u y CJ C3 C cc! T 3 C aj S. tU O C o o 13 a o OD c tu E 'S 3 ca 00 T3 tU O o o •5 M > . M tu o c tu ca o T3 O CU a , x u tU > ca cu . 2 ° c 2 & ^ « '-3 tu ca c o a cu 6 o c tU * J J 3 C 2 OM i3 c t3 <u 00 c •-- ca c 01 o o > JS o =3 00 .S! J 3 x : 0 00 CU l—i tt cu c ca > u u o CU H co 3 "a c Ci s C3 ID 6 0 cu ca tt N ca OS 00 r, ^ a \ u ON - H j -OO" ^ ^ p q o 2 0 0 6 0 <^ rr, X ) ^ 6 0 <U ? M <U o > O X • M 1) •a x ; c M 0 '5 tt > 01 — tu g M O O l CN % <* i J x ; _o 00 o „ tu o i5 CJ g.c2 ca | 00 W T3 o <u o <^ ^ « ON „ 00 e o\ ca — E M X! U O tt j« o •3 o ffi U I, - a a o tt 01 <u -3 ca ID !D ID 3 >, • -^ >-> tt o E "ca <D O E: ID o c 00 3 c ca X ) O 01 X 3 S CO 0 00 ca o ii tt o tu ON 0O ON =5 -ca tu 73 S «« cac o es N 'E C3 6 0 „ O T 3 tu •3 C S " 00 ca O tu 00 ON Sl <u . 2 6 0 S 10 © c o Z 60 3 C o CJ § o — C LO "O CJ c o 8/3 cj > 3 * c '-2 .a CJ 2 cj G & .2 CJ cd cj N o C 00 . J o .5 3 LO rv LO CJ cj S- 2 Cd c M C3 CJ LO O C "3 5 u o : t i > a , cs e -g two -3 a E <u CJ & o u a, — ° T3 C O •o o >> 3 a CJ O t o C a E c CS > 05 o c CJ CJ c C3 > JO CJ I H CJ c o c T3 "2 C c cd cd O « 3 c 3 '33 CJ u O 3 > CJ CJ CJ c CJ CO cd CJ O u o H C 0.1 cd c O © C s CJ r v ^o rr-h U C U 2 ° L(_ | LO c o CJ CJ cj ~ oo w 5 cj r ? - O oo <£ 2 " CJ CJ -a c cd 6 ^ S B o ON ON cd r, & C o v. ' -a <J f-. LO O R c ^ oo 3 g ^ Taken together, this research stimulates critical questions rather than provides clear answers to the phenomenon of organizational injustice and voice and silence among employees who feel unjustly treated. To provide a comprehensive theoretical framework for this research, I integrate these literatures to shape a research focus, to develop initial research questions and to introduce a set of propositions, including a priori constructs,4 to guide data collection and preliminary analysis. Contrary to confirmatory research protocol but consistent with the exploratory tradition, I have formed neither hypotheses nor theories (e.g., Eisenhardt, 1989; Glaser & Strauss, 1967; Strauss, 1987). In sum, the breadth of disciplines and knowledge drawn on for this research provides a rich foundation for the synthesis needed both to build organizational theory and to inform its practice. Knowledge of Injustice through the Organizational Justice Literature Defined broadly, justice is said to exist when people receive that to which they believe they (and others) are entitled or they deserve, whether those receipts are seen as benefits or burdens (Buchanan & Mathieu, 1986; Cohen, 1991). In contrast, injustice involves a violation of a moral contract for goods, services, opportunities or treatment (Fine, 1983). Historically, justice theory has developed outside of organizational theory. Thus, a large volume of literature explores justice at the level of society (e.g., Rawls, 1971) and its institutions, such as the judiciary (i.e., courts) and law enforcement (i.e., police) while a much smaller volume of research directly and explicitly links itself to organizations and an even smaller volume directly examines workplace injustice. In this review, I limit the organizational justice literature to that which is most relevant to perceived unfairness at work. 4 Eisenhardt (1989) argues for the inclusion of a priori constructs in qualitative organizational research because their subsequent emergence in data analysis provide strong, triangulated measures on which to ground and provide support for emerging theory. 12 Component justice literatures briefly considered here include distributive (Homans, 1961) and procedural justice theories (Thibault & Walker, 1975). These are similar in their mutual concern with the fairness of allocations and resolutions but different in their targets of concern. That is, while distributive justice focuses on the fairness of outcomes, procedural justice focuses on the fairness of procedures used to achieve those outcomes. In recent years, organizational justice has emerged with its exclusive focus on variables and issues concerning fairness in organizations, particularly issues of procedural fairness (Greenberg, 1987, 1990b). Applications to or implications for injustice at work are often implicit and indirect, but significant nonetheless. Distributive Justice Approaches Equity research. Stouffer, Suchman, DeVinney, Star and Williams' (1949) studies on soldiers' promotion satisfaction introduced issues of justice and injustice in organizational research from which relative deprivation (Runciman, 1966) and equity theories (Adams, 1963; 1965) evolved. Both theories suggest a similar individual-level process by which people can feel unjustly treated in relation to others. Specifically, individuals compare their ratio of outcomes (e.g., pay, benefits, job satisfaction) over inputs (e.g., ability, education, performance) to the ratio of others, called referent comparisons. Equal ratios are believed to produce perceptions of equity (hence justice) while unequal ratios produce perceptions of inequity (injustice or deprivation). In particular, people with lower equity ratios than others are likely to feel angry while people with higher ratios are likely to feel guilty. Equity theory in particular has been widely researched in and applied to organizations (see Mowday, 1987). Equity theory holds that when people perceive injustice, they try to restore perceptions of justice by changing their beliefs or by taking direct action. (For fuller 13 descriptions and reviews, see Berkowitz & Walster, 1976 and Walster, Walster & Berscheid, 1978.) However, recent research suggests that people are not equally motivated to change inequity-induced injustice perceptions: the significance of injustice perceptions appears to vary across people due to an intrapersonal moderator, the equity sensitivity construct, which determines personal tolerances for injustice (Huseman, Hatfield & Miles, 1987). Its validity has been challenged because of the possibility that situational factors such as varying norms of justice represent confounding factors (Greenberg, 1990b). To date, equity sensitivity remains relatively untested except for two empirical studies which provide support for the concept (Miles, Hatfield & Huseman, 1989, 1994). Together, this research is noteworthy for its attention to justice and relative inattention to injustice concerns in organizational settings. Although some researchers conclude that the large literature supports the assumptions and propositions of equity theory (e.g., Cropanzano & Randall, 1993; Mowday, 1987), others question the construct validity of typical experimental injustice manipulations (Lawler, 1968; Schwab, 1980) and the neglect of controversial and emotional issues which reduce the likelihood of contradictory findings (Martin, 1993). For example, equity theory has stimulated a substantial volume of research which largely involve students in experimental laboratory settings using pay as the outcome.5 Accordingly, the limited realism from experimental settings and the disregard for more intrinsic outcomes are examples of what some researchers consider serious shortcomings which preclude assessment of the merits of equity theory (e.g., Pinder, 1998) and its relevance to injustice at work. 5 Notable exceptions are equity studies involving job title status (Greenberg & Ornstein, 1983), workplace status (Greenberg, 1988) and satisfaction with work environment (Greenberg, 1989). 14 In summary, equity theory research offers insight, albeit often indirect, to the phenomenon of workplace injustice and employees' responses to it. Moreover, the focus on non-controversial and non-emotional issues creates what Martin (1993) argues is a tendency to legitimate justice norms by restricting opportunities for the study of delegitimation (i.e., when organizational policies or practices are not consensually understood as right or just). Although this research does not explicitly address the roles of voice and silence, its effects on these concepts is significant, nonetheless, by maintaining a focus on non-emotional issues which continues to this day and which supports the continued neglect of voice and silence as dissent or delegitimation. Procedural Justice Approaches Seminal work. Research on procedural justice began with Thibault and Walker's (1975) theory which suggests that the perceived fairness of decision-making procedures influences people's reactions to third-party allocation and dispute-resolution decisions more than the fairness of actual decisions reached. Specifically, the main procedural factor influencing people's views about fairness is the type of control: process or decision. Process control refers to participants' sense of control over the presentation of evidence concerning decisions whereas decision control refers to participants' direct control over actual decisions. To a large extent, subsequent research has examined how people evaluate the fairness of procedures. Subsequent research. For example, Leventhal (1980) proposes six criteria which influence perceptions of procedural justice:6 representativeness, whether there is 6 As noted earlier, most reviews place Leventhal's justice judgment model within the distributive justice tradition because of its focus on the fairness of outcome distributions. However, I position Leventhal's justice criteria in the procedural justice tradition because of my belief that their purpose is to articulate elements used in the process of evaluating the fairness of outcome-distribution procedures. 15 representation of the views of all interested parties; consistency, whether there is consistency of actions and rules across parties and in other disputes; bias suppression, whether the decision maker is impartial; accuracy, whether the information used is accurate; challenge, whether decisions made are correctable; and ethicality, whether decision makers are ethical. In forming judgments of procedural fairness, the relative priority or salience afforded these criteria is assumed to change in different situations. To date, research on Leventhal's criteria has been limited to consistency (e.g., Greenberg, 1987; Sheppard et al., 1992; Tyler, 1990) with results supporting its importance to procedural justice (Greenberg, 1990b). However, the relevance and validity of these criteria to injustice judgments in work settings are not well known. Similarly, Lind and Tyler's (1988) group-value model of procedural justice which proposes that people's reactions to third-party allocation decisions and dispute resolution are affected by their concern for long-term social relationships with authorities and institutions has questionable validity because of limited research attention in general (Greenberg, 1990b) and to organizations in particular. In contrast, there is strong support for Bies and Moag's (1986) concept of interactional justice whereby the interpersonal manner of authority figures during the implementation of procedures significantly influences others' procedural justice perceptions (Greenberg, 1990b, 1993; Tyler & Bies, 1990). Interactional justice includes attributes such as truthfulness, respect and appropriate justification. Thus, if authority figures display honesty, civility and professional decorum and show consideration for employees' viewpoints, suppress personal biases, and provide explanations, then both interactions and procedures will likely appear fair to employees (Folger & Bies, 1989). 16 However, consistent with the majority of research reviewed here, the relative inattention of this research to contexts of injustice limits its generalizability to perceived unfairness at work. One exception is Bies' (1987) study of how harmdoers (i.e., perpetrators of perceived injustice) can manage others' injustice perceptions by providing a social account (i.e., an explanation that includes a justification or apology) to aggrieved parties. Nevertheless, its focus on authority figures means that, whether managerial or non-managerial employees, subordinates' experiences of or responses to interactional injustice remain largely unexplored. Justice theorists typically subsume interactional justice as a feature of distributive and/or procedural justice. For example, interactional justice has been incorporated into referent cognitions theory (RCT) which combines distributive and procedural concerns in organizations to explain the escalation of resentment into violence as reactions to mistreatment at work (Folger, 1993, 1995). Despite little empirical validation, preliminary findings support the promise of RCT as a valuable theory to understand responses to injustice in organizations. There have been two basic ways of studying procedural justice empirically. One uses legal settings involving trials (LaTour, 1978) and compliance with the courts and police (Tyler, 1990; see Lind & Tyler, 1988 for a review). A second and more recent approach examines procedural justice in organizations through its role in evaluations of supervisors (Tyler, 1991) and employees (Folger, 1986), reactions to drug testing programs (Konovsky & Cropanzano, 1991) and smoking bans (Greenberg, 1994), selection practices (Singer, 1990), pay raise decisions (Folger & Konovsky, 1989), pay freeze decisions (Schaubroeck, May & Brown, 1994) and reactions to layoffs (reviewed by Brockner & 17 Greenberg, 1989). This approach in which elements from either distributive or procedural justice are associated with organizational variables is commonly called organizational justice research (Greenberg, 1990b). Organizational Justice Approaches Most organizational justice research uses procedural justice as a dependent variable. Among the fewer studies which use procedural justice as an independent variable, its impact on job satisfaction, organizational commitment, trust in management, intentions to sue and/or quit and performance (Konovsky & Cropanzano, 1991; Tyler, 1991) have been studied. As well, many studies report that distributive and procedural justice perceptions appear to interact.7 For example, findings from Greenberg's (1986) open-ended interviews of managers' experiences with fair and unfair performance appraisals indicated that both distributive and procedural concerns affected their perceptions of fairness and unfairness. Similarly, both distributive and procedural factors were cited among the 16 justice principles identified from managers' experiences of fair and unfair treatment. Other studies report that procedural fairness is less important when personal outcomes are seen as positive (Greenberg, 1987) and that people tend to disregard procedural irregularities if they believe that their outcomes are not affected (Folger, Rosenfield & Robinson, 1983). Finally, Sheppard et al. (1992) combine two principles of justice (balance and correctness) with three levels of justice (outcomes or decisions, procedures or process and systems or the larger environment) which are believed to determine justice and injustice perceptions in their model of organizational justice. 7 Many studies conducted outside of organizations find main effects for procedural justice (see Lind & Tyler, 1988). 18 / Limitations. In these studies, procedural justice is typically measured by 1- or 2-item scales which ask how fair were the procedures used and how fairly were you treated with a 4- or 5-point frequency-based category scale of 'not at all' to 'very.' Similarly, the sole question commonly used to measure distributive justice is how fair was the outcome you received? However, 1- or 2-item scales yield reliabilities and validities which are tenuous at best and unacceptable at worst. Single-item scales, for example, preclude assessments of internal consistency, necessary for establishing construct validity (Schwab, 1980). Low scores on these scales are assumed to indicate perceptions of unfairness; the generalizability of findings to organizational injustice from measures with such limited reliability and construct validity is questionable. Furthermore, while theoretical research supports the multidimensionality of distributive and procedural justice, empirical research implicitly assumes, through its measurement practices, their unidimensionality. Although some research has generated multidimensional and standardized organizational justice scales (e.g., Lewis, 1990; Roehl, 1988), these remain relatively unused. Fundamentally, this research is marked by its inattention to injustice in organizational settings. Arguably, indirect inferences could be drawn for organizational contexts where unfairness, exploitation or abuse are either idiosyncratic or the norm, although such inferences necessarily must be speculative. However, even the few studies of workplace injustice tend to be conducted among 'benign' organizational contexts where managers seek fairness and employees have power (e.g., Rutte & Messick, 1995) or they focus on perpetrators (i.e., managers as harmdoer) rather than victims or recipients (i.e., subordinates) (e.g., Bies, 1987). 19 In summary, little is known of employees' experiences of and reactions to perceived injustice at work. Little research has been conducted to alter an earlier conclusion that "our present understanding of procedural justice in organizations is highly limited and skewed" (Greenberg, 1990b: 420) because of our limited knowledge of how people respond to organizational injustices. In addition, most organizational justice research fails to validate assumptions of legitimacy by failing to ask whether outcomes or processes are widely-held as right or just across organizational members. In part to address these limitations, in this research I continue earlier qualitative research of injustice perceptions in organizations (e.g., Greenberg, 1986a; Sheppard & Lewicki, 1987) to more closely examine processes involved in employees' experiences of workplace injustice. In the next section, extant research relevant to employees' responses to organizational injustice is reviewed. Voice and Silence in the Organizational Justice and Injustice Insofar as attention to voice is increasing, gaining an appreciation for current and future directions in research on voice and silence in organizational (injustice requires that consideration be given to their historical development.8 Voice Voice is widely accepted as critical to fairness perceptions. Folger (1977) first extended Hirschman's (1970) concept of voice (i.e., the capacity to be heard) to organizational justice research by comparing fairness in voice to mute (i.e., no opportunity for voice) conditions. In this and subsequent research, a consistent finding is that people perceive procedures allowing voice as fairer than mute procedures (Folger et al., 1979; Unlike Greenberg's comprehensive reviews (1987; 1990b), this review explicitly considers voice and silence. 20 Greenberg & Folger, 1983), even when a decision is unfavorable to them (LaTour, 1978). Also, decision-makers' voice by providing a justification for an unfavorable decision has been found to have an independent effect on fairness perceptions (Bies & Shapiro, 1988). Two perspectives attempt to explain how and why voice influences justice and, to a lesser degree, injustice perceptions, namely the value-expressive and instrumental perspectives. Two perspectives of voice. The value-expressive view proposes that voice influences perceptions of fairness by allowing employees to be heard and to have their views considered, regardless of the outcome (Sheppard et al., 1992; Tyler, 1987). Within this view, some suggest that simply voicing dissatisfaction is not enough to produce perceptions of justice; people must also believe that their viewpoints are considered by the listener (Cohen, 1989; Shapiro, 1993). In contrast, the instrumental perspective holds that voice influences justice perceptions by creating the belief that outcomes will be more just from having influenced the decision-maker (Leventhal, 1980; Thibault & Walker, 1975). Research findings suggest support for both the value-expressive and instrumental perspectives (Cohen, 1989; Lind & Tyler, 1988). Impact of voice systems. Research indicates that voice systems lead to reduced turnover and greater loyalty and commitment to the organization (Sheppard et al., 1992; Tyler, 1989), although how voice systems positively affect organizations is largely unknown. However, Sheppard et al. (1992) propose that five core attributes determine whether voice systems are perceived as effective (i.e., encourage access to and satisfactorily responding to employees' discontent): elegance (efficiency), accessibility, responsiveness, correctness, and non-punitiveness. To date, no known research has substantiated the role of these attributes in the functioning of voice systems. 21 Consequences of voice: Fair process and frustration effects. Consistent with the predominant focus in organizational justice research on the positive effects of fairness, most research on voice has explored the fair process effect which refers to the pattern of increased satisfaction when people are given a voice in decisions (Folger et a l , 1979). Examples include more positive evaluations of decision-makers and institutions, decreased disruptive behaviour and increased compliance, participation and performance following voicing (Cohen, 1989; Lind & Tyler, 1988). However, much of this research fails to include non-users of voice systems. Thus, because it is not known if employees who use voice systems are representative of their respective workforces and because silence may indicate dissent, the generalizability of these traditional favorable results to employees, both within the target organization and to other organizations, is questionable. Similarly, many studies of the fair process effect use experimental settings designed to model employer-employee relations: a subject (typically an undergraduate student) plays the role of an employee who is paid an amount of money perceived as unjust for hypothetical work. As a result, the generalizability of findings to the workplace is questionable. Moreover, Cohen (1985) argues that voice research often overlooks the possibility that interests of employees and employers may be conflictual and can result in inaccurate assumptions that fair process effects are beneficial for all. In contrast, the frustration effect describes employees' increased dissatisfaction or frustration due to raised expectations from giving people voice (Folger et al., 1979). In the few studies which consider the potential for voice to engender negative reactions, users of organizationally-sponsored child care centers (Kossek & Nichol, 1992) and subordinates 22 completing upward performance assessment (Westerman & Rosse, 1996) demonstrated the frustration effect by reporting greater dissatisfaction and frustration after voicing. Access to voice systems. Voice research implicitly assumes that all employees enjoy equal access to and facility with voice channels despite societal evidence that access to voice channels is not equally distributed across populations. For example, because of gender-based social and financial inequities, women's access to voice systems is believed to be restricted (Cohen, 1990). In organizations, parallel inequities likely exist, thereby challenging the assumption of equal access. In addition, voice system theory and practice typically fail to consider the diverse needs of a diverse workforce; it is unlikely that all employees have identical needs for protection and safety in order to use voice systems. In particular, vulnerable or less advantaged employees (e.g., hourly workers, victims of employment discrimination) (Davidson & Earnshaw, 1991; Martin, 1993) may have greater needs for anonymity and/or confidentiality compared to more advantaged employees. Assumptions. One explicit but untested assumption underlying organizational justice research in general and voice research in particular is that employees who believe in the possibility of improvement will stay and those who do not will leave (Hirschman, 1970). However, an equally reasonable but historically neglected possibility is that dissatisfied employees may stay and remain silent with a less-than-hopeful belief about the potential for improvement. Although not desirable, staying with a negative or cynical attitude may be common, particularly for less advantaged employees whose employment alternatives are restricted (e.g., those affected by gender-, race-, or class-based economic inequities). Despite 23 its common-sense appeal, the link between organizational cynicism and voice has not been widely studied. However, indirect evidence of this link from research on Hirschman's (1970) responses to dissatisfaction is noteworthy. Withey and Cooper (1989) hypothesized that optimism about change would be associated with voicing behaviour. However, support was found in one of two samples studied in their research. It is also noted that low optimism, measured by their scale which assessed leader-member relations and receptiveness to change, is not equivalent to organizational cynicism (cf. Reichers, Wanous & Austin, 1997; Wanous, Reichers & Austin, 1994). Further investigation of the link between cynicism, voice and silence is supported by Withey and Cooper's observation that "several employees...described themselves as quite unhappy with their jobs, yet they didn't voice their dissatisfaction because they didn't think any good would come of it" (1989: 535). Another implicit, related assumption in voice research is that injustice will be objected to. However, this assumption is challenged by findings which suggest that disadvantaged people are rarely able to act as their own advocates (Crosby, 1984) and that, despite unfair treatment, individuals are often hesitant to take action on their own behalf (Martin, 1986). Curiously, voice research fails to test this assumption by its historical neglect of disadvantaged employees and by its failure to evaluate voice systems through tracking usage by individuals and sub-groups (e.g., gender, managerial level) to assess representativeness or through measuring voice system satisfaction. Summary. Despite growing interest, little is known about efficacy or legitimacy of voice systems, particularly in the context of organizational injustice. A common, implicit assumption is that voice mechanisms are afforded equal legitimacy by all organizational 24 members, regardless of differential benefits or relevance. This assumption, in turn, supports the untested but common premise that those who decline to voice have no complaint. Thus, silence is implicitly defined as the absence of voice and signifies endorsement (Cohen, 1990). Silence Cohen's (1990) work represents a rare, significant effort to develop the concept of silence within the context of workplace injustice. Specifically, he considers factors that create silences among those unjustly treated or those observing such treatment and proposes several types of silence: as endorsement, as objection, as a result of lack of information, as a result of absence of voice opportunities, as a result of believing that voice is a sham (i.e., fabricated or illusory); and/or as a result of the dangers (real or imagined) of voicing. These views directly challenge the enduring interpretation in the organizational literature which views silence as an absence of voice indicating consent which requires no further study. This historic bias, Cohen argues, allows decision-makers to use silence as evidence of endorsement to substantiate their claims of justice, even when justice is fabricated (i.e., constructed only to appear just; Greenberg, 1990a). Cohen's (1990) work notwithstanding, little attention has been given to silence as a response to organizational injustice or its consequences. Martin (1993), however, does speculate that physical and mental illness, drug and alcohol dependency may be symptomatic of chronic, unexpressed injustice or dissatisfaction. Accordingly, this research explores, as an a priori line of inquiry, alternate meanings of silence (e.g., as objection) in work settings perceived as unjust and the consequences of silence for the individual and the organization. 25 Implications for the Present Study This literature points to the unexplored possibility that voice systems may increase silence by means of the frustration effect. For example, if dissatisfied employees believe that the organization is guided more by "looking fair" rather than "being fair" (Greenberg, 1990a), their trust in the organizational commitment to justice may be low such that they remain silent, thereby reducing the efficacy of voice systems through reduced representativeness (Leventhal, 1980). Accordingly, both the fair process and frustration effects and their impact on voice and silence are studied in this research, which include the study of possible links between voice system expectancies and the fair process and frustration effects by asking participants about their expectations of and assessments of voice systems. Furthermore, I include dissatisfied employees who have remained silent, despite voice system availability, and I include members of disadvantaged groups (e.g., blue-collar workers, women, victims of discrimination), in part to respond to Martin's (1993) call for empirical testing of legitimacy assumptions. The Exit, Voice and Loyalty Literature: Responses to Dissatisfaction Origins. Hirschman (1970) introduced exit and voice as behavioural responses to customers' perceptions of organizational decline and, to a lesser degree, organizational members' dissatisfaction. His attention to the latter serves as a theoretical foundation for this research. Specifically, he considered exit (i.e., voluntarily leaving an organization) as an economic response while he viewed voice — "any attempt to change, rather than escape from, an objectionable state of affairs" (1970: 30) — as a political response. Hirschman compared the conditions in which voice and exit were chosen as mutually-exclusive responses and the conditions in which these options acted conjointly. 26 Of particular interest to this research is Hirschman's conceptualization of loyalty as an attitudinal construct which moderates between exit and voice in responding to dissatisfaction. Although most loyal employees are thought more likely to stay and voice their complaints or suggestions rather than exit, some loyal employees are believed to stay and "suffer in silence, confident that things will soon get better" (1970: 38). In effect, Hirschman subsumed silence under loyalty and, in so doing, inadvertently created confusion between these concepts and, more fundamentally, initiated an inattention to silence which is maintained to this day.9 Accordingly, acknowledging silence as a distinct and significant response to dissatisfaction and reducing conceptual confusion in both silence and voice form a primary focus of this research. There have been two ways of continuing Hirschman's work theoretically and empirically. One has been to retain his original focus on exit and voice, exemplified by studies in the labour market and unionized work settings (e.g., Freeman & Medoff, 1984). The second, less-developed approach, expands Hirschman's responses to dissatisfaction to include neglect (e.g., Farrell, 1983; Withey & Cooper, 1989). As a result, the most commonly considered set of responses to dissatisfaction comprises exit, voice, loyalty and neglect and is commonly abbreviated to E V L N . Generally, E V L N research treats responses to dissatisfaction as dependent variables and focuses on their measurement and prediction. As well, it is emphasized that the standard research context of E V L N responses is dissatisfaction in general rather than injustice in particular. 9 A notable exception is Robinson's (1993, 1994) explicit inclusion of silence as a response to dissatisfaction. 27 The Measurement of Voice and Loyalty in E V L N Research Of E V L N responses, voice and, to a lesser degree, loyalty (read silence) form the conceptual and empirical foundation of this research. Therefore, this review focuses on findings concerning these two concepts. Insofar as the present work departs from traditional research into dissatisfaction responses, gaining an appreciation for its points of departure requires that consideration be given to the historical development of voice and loyalty (/silence). Voice. The consensual definition of voice across studies is that it reflects active change efforts such as filing a grievance or making a suggestion (e.g., Olson-Buchanan, 1996; Rusbult et al., 1988). Despite relatively consistent agreement on its definition and description, there are conflicting findings about the predictability of voice. Of the studies which predict voice, Saunders, Sheppard, Knight and Roth (1992) report that employee perceptions of how their supervisors will manage voice predict whether employees will speak up about their work-related concerns to supervisors. Also, Farrell and Rusbult (1992) identify job satisfaction, quality of job alternatives and degree of investments in the job as significant predictors of responses to dissatisfaction, including voice. However, many studies fail to explain voice (e.g., Leek & Saunders, 1992; Robinson, 1994). Withey and Cooper (1989) report that, among E V L N responses, results explaining voice are weakest. These conflicting findings suggest the need for research to develop the concept of voice and to explore its expression by employees in response to dissatisfaction at work. The most likely explanation for its contested predictability is the weak construct validity of voice measures, indicated, in part, by low reliability indices. To date, most empirical studies have assumed that voice is a unidimensional construct. However, low 28 indices of internal consistency (a = .41 to .48) suggest that voice is multidimensional in nature (Meyer, Allen & Smith, 1993; Rusbult et al., 1988; Withey & Cooper, 1989). Moreover, efforts to assess the construct validity of E V L N measures appear inadequate, at best, or absent, at worst. Withey and Cooper (1989), for example, attempt to provide evidence for the construct validity of their voice scale by asking supervisors to rate the extent of their subordinates' voicing activities in the previous six months. Only 14% of employees were identified by their supervisors as active voicers. This unexpectedly low base rate precluded the development of a continuous scale to measure employees' voicing behaviour. As a result, a dichotomous ("yes" or "no") scale of voicing activity was completed by supervisors; it was then correlated with a 3-item self-report measure of voice. The low variance in ratings likely accounted, in large part, for the reported low (albeit significant) correlation of .18, suggesting that supervisory ratings are not an adequate measure to assess the construct validity of voice. One obvious limitation to this approach is that it fails to capture variations in both the type and degree of activity in employees who voice. Consequently, the base rate for voice is likely underestimated by excluding voice efforts made outside the employee-supervisor dyad. Given that voice systems typically provide confidentiality (and, in some cases, anonymity) to encourage participation, supervisors are often unaware of the different types of voicing activities their subordinates engage in: for example, downward (i.e., to employees' subordinates), lateral (i.e., to co-workers) and peripheral (i.e., to ombudspeople). More fundamentally, this approach relies on a restricted and inadequate conceptualization of voice. Despite its limitations, Withey and Cooper's (1989) research does address the construct validity of E V L N measures, unlike most previous studies. Previous research 29 typically asks respondents to complete scales of the frequency or extent to which they engaged in (or are likely to engage in) responses using 5 or 7-point frequency-based category scales with options from "I have never considered this action" to "I have engaged in this action frequently." The number of items used in voice scales ranges from 2 (Meyer et al., 1993) to 10 (Robinson, 1993, 1994). Some of the larger scales, however, include items which do not appear relevant to voice. For example, one 5-item scale to measure voice includes "voluntarily wearing clothing that bears your organization's symbol or insignia" and "saying good things about your job even when others criticize it" (Leek & Saunders, 1992). The modest internal consistency of this scale (a = .72) may reflect the inclusion of items seemingly unrelated to change efforts. Yet, improved reliability does not necessarily yield improved predictability. Robinson (1994), for example, reports a promising reliability coefficient of .91 for the 10-item voice scale.10 However, none of the hypotheses involving voice received support, leading Robinson to conclude that, contrary to its strong evidence of internal consistency, voice may be a multidimensional construct. In addition to conceptual and methodological inadequacies in the construction and measurement of voice, many research designs introduce confounds in dissatisfaction perceptions which may partially account for the contested predictability of voice. Specifically, many E V L N studies fail to restrict samples to those who perceive (or have perceived) dissatisfaction, thereby introducing possible confounds (Bemmels, 1997; Boroff, 1989). Furthermore, most studies implicitly assume that dissatisfaction perceptions are equal 1 0 Robinson does not identify the type of reliability index used; I have assumed that it is an alpha reliability coefficient. 30 in intensity and meaning for all employees. Most studies, for example, assume that all participants have experienced some form of dissatisfaction and that neither excludes those who have not perceived dissatisfaction nor distinguishes the degree of dissatisfaction. As Bemmels (1997) speculates, if implicit assumptions that all participants have experienced dissatisfaction and that dissatisfaction perceptions are equally meaningful are invalid, then base rates of dissatisfaction responses will be underestimated and correlations with relevant variables may be both underestimated and unstable. Contrary to common research practice, the accurate prediction of voice requires the confirmation of dissatisfaction; in the absence of dissatisfaction, employees will be unmotivated to express voice. Accordingly, in this research I attempt to reduce such confounds by asking participants to describe a workplace situation they experienced as unjust, their response(s) to the situation and the meaning and consequence(s) associated with both the situation and their response. Another common but implicit assumption is that voice is the preferred response to dissatisfaction (Bemmels, 1997). Each of the four E V L N responses has been categorized on the basis of the level of employees' involvement (i.e., active vs. passive) and effects of responses (i.e., destructive vs. constructive) (Farrell, 1983; Rusbult & Lowery, 1985). However, Gorden (1988) challenges the assumption that voice is favorable by applying this four-cell typology to differentiate among active, destructive forms (e.g., verbal aggression) and passive, constructive forms of voice (e.g., attentive listening, unobtrusive compliance). Despite its theoretical contribution, Gorden's model furthers conceptual confusion by including "calculative silence" and "psychic withdrawal" as examples of passive, destructive voice. In other research, Parker (1993) distinguishes voice from the concept of reformist 31 dissent which she introduces as a continuum of behaviours by which loyal employees try to improve the organization through sanctioned means. In summary, conflicting findings for the prediction of voice suggest the need for a more definitive investigation of the concept of voice and of its consequences. Recent research suggests that voice is multidimensional, reflecting several aspects of initiating change. Other findings suggest that different types of voice may exist (Sheppard et al., 1992), perhaps with different antecedents (Robinson, 1994). Loyalty. As noted, since Hirschman (1970) first equated loyalty with silence, ambiguity and controversy continue to surround the meanings ascribed to and the applications of loyalty (Barry, 1974; Leek & Saunders, 1992). Subsequent research continues to equate silence with loyalty, at times explicitly but usually implicitly. Its relevance to this research stems, in part, from its traditional connection with silence. In E V L N research, loyalty has alternately (and sometimes simultaneously) been conceptualized as an attitude or a behaviour. For instance, Hirschman (1970) saw loyalty as an attitudinal construct defined loosely as a feeling of attachment to the organization which moderates the choice between exit and voice. But he also described and measured loyalty as employees' behaviour supportive of the organization. Later research has relied on three main approaches to study loyalty. In one approach, research efforts focus exclusively on loyalty as an attitude (e.g., Mayes & Ganster, 1988). In contrast, other research defines loyalty as a passive but constructive behaviour measured by scales comprised of items such as being quietly supportive or waiting for the problem to disappear (Farrell, 1983; Farrell & Rusbult, 1992; Rusbult et al., 1988). However, these behavioural scales are sometimes confounded by attitudinal measures of loyalty, seen in 32 items measuring the belief that one's job and/or organization is good and that superiors know what they are doing. In the third approach, still others study loyalty as an attitude and behaviour simultaneously. Leek and Saunders (1992), for example, re-label behavioural loyalty as patience and develop an attitudinal measure of loyalty which reflects feelings of fidelity, trust and allegiance to the organization and includes items from Mowday, Steers and Porter's (1979) Organizational Commitment Questionnaire (OCQ). As expected, the reported correlation between these variables is positive and significant, although modest (r = .19). As hypothesized, attitudinal loyalty was found to promote voice and patience (read silence). However, debate continues about the effect of loyalty on voice. Evidence for both conflicting positions — that loyalty either activates or suppresses voice — appear equally strong (for a review see Graham & Keeley, 1992). Several efforts to reconcile the contradictory findings for loyalty have been made. Graham and Keeley (1992), for example, cite inconsistencies in the type and meaning of loyalty used among studies. They find evidence for three equally valid types of loyalty: unconscious, passive and reformist each with different antecedents and consequences. More fundamentally, Withey and Cooper (1989) argue that conceptual inadequacies account for contradictory findings. On the basis of a significant, negative validity coefficient (r = -.23) reported between behavioural loyalty (measured by Farrell's scale; 1983) and attitudinal loyalty (measured by the OCQ; Mowday et al., 1979), they conclude that "something is seriously amiss with the concept of loyalty" (1989: 537). These problematic findings may result from treating organizational commitment and loyalty as identical and thus interchangeable constructs. While close conceptually, these 33 constructs are not invariably equivalent, especially given the multiple conceptualizations of loyalty as an attitude, behaviour or combination of the two. Similarly, organizational commitment is alternately conceptualized as unidimensional (e.g., affective attachment to the organization; Mowday et al., 1979) and multidimensional (e.g., affective, continuance and normative commitment or attachment; Allen & Meyer, 1996). In addition, the construct validity of loyalty measures used in E V L N research is often not well-established. For example, Lewin and Boroff s (1994) five-item loyalty scale asks how employees would respond to an ethical dilemma and whether they would rely on their union or a lawyer to launch workplace complaints. In this case the correspondence between the concept of loyalty and its measure seems questionable. Methodological problems also appear to compound these conceptual inadequacies. In particular, the inconsistent use of identical items across different scales is evident in several studies. For example, the item "I wear clothing that bears the organization's symbol or insignia" appears in Leek and Saunders' (1992) voice scale but in Rusbult et al.'s (1988) scale measuring loyalty. Silence. As noted earlier, many E V L N studies continue to use loyalty and silence interchangeably and to ignore silence, despite its hypothesized rank as the second-most common response to dissatisfaction (Kolarska & Aldrich, 1980) and its emerging position with voice as one of two primary responses of dissatisfied employees (Withey & Cooper, 1989). Despite this and other support for incorporating silence in the dissatisfaction-response framework (e.g., Spencer, 1986), few efforts to explicitly identify, describe and predict silence exist. A notable exception is Robinson's (1993, 1994) inclusion of silence as a response to dissatisfaction. However, Robinson's silence scale is primarily comprised of 34 items used by other researchers to measure loyalty (i.e., Leek & Saunders, 1992; Rusbult et al., 1988; Withey & Cooper, 1989). Not surprisingly, Robinson reports findings which provide weak support for silence as a predictable response; only two of eight hypotheses concerning silence were confirmed. Boroff s (1989) conceptualization of silence is even more limited as an implicit exclusion of voice. That is, employees who experienced unfair treatment but did not file a formal complaint were considered, by default, to be silent. Thus, silence as endorsement is implicitly assumed. Lastly, Parker (1997) incorporates silence, not as a distinct response, but as a response conjoint with exit to yield principled turnover (that which arises from perceived injustice). In short, the E V L N literature offers little insight, conceptually and empirically, into silence as a response to dissatisfaction. Accordingly, the conceptual development called for in the present research relies on other literatures to formulate preliminary questions about silence. Relationships between voice and loyalty. Findings indicate that responses to dissatisfaction are often interrelated. Specifically, responses may co-occur, evidenced by employees who express voice as they exit, and may occur sequentially, seen among employees who voice prior to exit (Boroff, 1989; Keeley & Graham, 1991). Despite theoretical support, no known research has explored the sequential nature of responses to dissatisfaction. Moreover, the reliance on a static model to explain what likely is a dynamic process is an under-recognized limitation of E V L N research (cf. Farrell & Rusbult, 1992). While multi-response combinations are possible (Bemmels, 1997), few are theorized or tested given the traditional inattention to voice and silence as responses to a dynamic process of dissatisfaction. To address this significant limitation, the present study explores the 35 processual nature of responses to dissatisfaction by considering the temporal dimensions of voice and silence. Commonly Studied Predictors of Voice and Silence Along with conceptual inadequacies in outcome variables (or their operationalization), conceptual limitations with predictor variables (or their operationalization) explain, at least in part, the consistently low correlations reported between predictors (e.g., job satisfaction) and outcomes (i.e., E V L N responses). The following review differentiates among the most commonly studied independent variables used to predict voice and silence based on their level of analysis: individual, group or organizational.11 Individual. The most widely-studied of the three categories, this set of predictor variables refer to personal characteristics such as locus of control (Withey & Cooper, 1989), belief in the possibility of improvement (Kolarska & Aldrich, 1980; Withey & Cooper, 1989), prior job satisfaction (Lewin & Boroff, 1994; Rosse & Hulin, 1985; Rusbult et al., 1988; Withey & Cooper, 1989), and loyalty and commitment (Hirschman, 1970; Withey & Cooper, 1989). Across studies the most consistent relationship reported is that between prior job satisfaction and voice, suggesting that people who remember being happier are more likely to make active change efforts. However, contrary to predictions, Saunders et al. (1992) did not find a significant relationship between satisfaction and employee likelihood to voice. Most studies also report correlations between prior satisfaction and loyalty (read silence) as a response to dissatisfaction in the expected positive direction. A notable exception is a 1 1 Bemmels (1997) places EVLN predictors into four categories: individual characteristics, constraints, opportunities and social influences. The arrangement I suggest here is a broader aggregation of the predictor variables and is intended to draw attention to their implications for organizational functioning. 36 negative correlation between satisfaction and loyalty (Withey & Cooper, 1989). Because we would expect a negative correlation between prior satisfaction and silence, this finding provides indirect support for the misidentification of loyalty as silence; people who remember being unhappy or dissatisfied likely forgo active change efforts. Many other contradictory and unexpected findings are reported across E V L N studies, often indicating only partial support for hypotheses. For example, Saunders et al.'s (1992) expected relationship between commitment and employee likelihood to voice was not observed. Also, Lewin and Boroff (1994) report an unexpected negative correlation between loyalty as a predictor and voice as an outcome; they attribute this to the misidentification of silence as loyalty. In other research, voice, as expected, is associated with an internal locus of control, optimism about improvement, higher commitment and lower voice costs (estimates of the time, effort and consequences of expressing dissatisfaction) while loyalty appears related to external locus of control (Withey & Cooper, 1989). However, contrary to expectations, loyalty is also related to pessimism about improvement and low commitment. Again, these unexpected findings provide partial support for the claim of conceptual misidentification as they are more consistent with expectations about silence than loyalty. In any case, Withey and Cooper acknowledge the unreliability of findings given their failure to replicate in a second sample. One possible explanation for contradictory findings involving commitment as a predictor variable may lie in the historic reliance in E V L N research on the OCQ (Mowday et al., 1979) to measure this construct. Designed to measure only affective attachment to organizations, its unidimensional focus may not fully capture the concept of commitment and, as a result, correlations with other measures may be biased toward zero. Recent research 37 supports a multidimensional conceptualization of commitment as comprising three components: affective commitment refers to emotional attachment to the organization; continuance commitment refers to recognition of the costs associated with leaving the organization; normative commitment concerns commitment based on a sense of obligation to the organization (Allen & Meyer, 1996; Meyer et al., 1993). If voice and silence are also multidimensional, a more complex commitment construct involving different forms may be needed to fully explain its relationship to voice and silence. Group. This category of independent variables refers to employees' social environment and includes employees' perceptions of supervisors' ability to manage voice (Saunders et al., 1992), of the supervisory relationship, and of social norms (social rewards and punishments from superiors and co-workers for expressing voice or remaining silent; Robinson, 1994). Robinson found that silence is more likely when employees do not enjoy a mutually supportive relationship with their respective supervisors. As well, silence is more likely when co-workers respond to their own dissatisfaction with silence. However, no hypotheses for voice received support. Other findings report that supervisors who are perceived as responsive and approachable are more likely to have employees who voice (Saunders et al., 1992). Together, these results suggest that employees' social environment is an important consideration in understanding voice and silence in the context of organizational injustice. This line of inquiry is furthered in the present study by examining associations between the contextual conditions (e.g., social, organizational) and workplace injustice. Organizational. Of the three categories, variables referring to organizational policies, practices and/or structural features have received the least attention in explaining voice and silence. To date, voice system availability is the only variable which appears related to 38 dissatisfaction responses, although findings are conflictual. Specifically, Graham (1986) reports that when employees are aware of voice channels, they are less likely to remain silent and more likely to express voice . However, Robinson (1994) found that voice system availability influenced neither voice nor silence. However, she did find that the availability of voice systems is positively related to destructive behaviours and hypothesized that the presence of voice systems increases employees' expectations that their discontent will be resolved — when their expectations are unmet, they damage organizational property, for example. While research on voice system availability considers the number of systems, it does not explicitly consider their perceived quality or effectiveness. To date, no known research has examined employees' perceptions of voice system quality and their impact on dissatisfaction responses, whether voice or silence. Summary. Despite the considerable attention given to these individual-level variables, their relationships to voice and silence remain poorly understood. The consistently low correlations may be a consequence of not studying salient concepts (e.g., self-efficacy or employees' power; Robinson, 1994) or using non-equivalent concepts interchangeably (e.g., using organizational commitment to measure loyalty). At the group and organizational levels, few studies have attempted to explain voice and silence; further research at these levels of analysis holds great promise for understanding the responses of dissatisfied employees. Consequently, the present research explores the relevance of group and organizational-level variables to voice and silence. In addition, this research explores both the multidimensionality of voice and silence and attempts to identify salient concepts through in-depth interviews. 39 Consequences of Voice and Silence As previously indicated, most E V L N research implicitly investigates dissatisfaction as a static phenomenon and views voice and silence as endpoints of dissatisfaction. Some studies have examined the consequences to employees from voicing discontent, suggesting that punishment or retaliation in the form of lower promotion rates, lower performance ratings or dismissal is common (see Feuille & Delaney, 1992 for a review). However, Olson-Buchanan (1996) reports that such punishment may be due, at least in part, to actual decreases in objective performance among those who voiced. These studies notwithstanding, the historic inattention to consequences, in turn, allows many untested assumptions concerning the effects of voice and silence to remain unsubstantiated. Summary of E V L N Literature A close inspection of E V L N research reveals several implicit and untested assumptions (e.g., that voice and silence are unidimensional) and substantial evidence of conceptual and methodological inconsistencies in the conceptualizations and measurement of voice, loyalty and silence which likely account, at least in part, for their contested predictability. In short, to address the limitations identified here, research which inductively develops rather than predicts the concepts of voice and silence is required to better understand responses to dissatisfaction and, by extension, organizational injustice. Qualitative methods are well suited to reveal possible multidimensionality of concepts, to uncover tacit processes (Morse, 1994) and to identify relevant variables. 40 Voice and Silence in Other Literatures As noted, the E V L N and organizational justice literatures offer some direct insight into voice but little direct insight into the incidence of or factors associated with silence. Accordingly, I draw on an even wider body of knowledge to stimulate questions and explore possible constructs related to the process and consequences of silence and, to a lesser degree, voice in response to perceived injustice in organizations. The Whistle-blowing Literature Whistle-blowers — employees who disclose illegal, immoral or illegitimate organizational practices to persons who can effect action (Near & Miceli, 1987) — represent a distinct type of voicer whose sense of injustice concerns externally-focused violations (e.g., misuse of organizational resources by superiors). In contrast, the present research focuses on voicers whose sense of injustice is internally-focused (e.g., personal mistreatment by a superior). However, in whistle-blowing research the study of inactive observers — employees who observe wrongdoing but stay silent (Miceli & Near, 1984; Near, Dworkin & Miceli, 1993) — provides support (albeit indirect) for the involvement of silence in perceived injustice. Specifically, research on whistle-blowers consistently identifies retaliation (e.g., demotion, expulsion) and intimidation (e.g., nullification and defamation) as serious and enduring threats against those who speak out (Miceli & Near, 1985, 1989; Near & Miceli, 1987). As well, U.S. federal laws designed to protect federal employees from retaliation are widely seen as ineffective (Near et al., 1993). In particular, the perceived threat to career advancement appears to separate whistle-blowers from inactive observers. For instance, inactive observers appear to be aware of probable retribution and so maintain their silence, in 41 part, to protect themselves and their jobs (Near et al., 1993). However, nothing is known about the costs (emotional or physical) of inactive observers' silence. In addition, inactive observers appear to know less about channels for whistle-blowing and view themselves as poorer performers than whistle-blowers (ibid.). This research notwithstanding, findings concerning inactive observers are rare, in part because the data used in whistle-blowing research is largely archival (drawn from the U.S. federal government). As well, given the sensitivity of the topic, access to inactive observers through field studies of governmental organizations is highly restricted (Janet Near, personal communication, August 6, 1995). Nonetheless, inactive observers constitute the only silent employees to have received explicit consideration in the organizational literature and so findings about them provide a valuable stimulus for questions in the present study. One resulting line of inquiry links the whistle-blowing and power literatures: given that whistle-blowers feel they are better performers than inactive observers (Near et al., 1993), do they also feel more powerful or effective? Conversely, do inactive observers feel powerlessness or ineffective? Related research suggests that whistle-blowing is political behaviour, specifically political resistance (Rothschild & Miethe, 1994) because the essential purpose of whistle-blowing is to effect change by resisting misconduct and unjust reprisals. Of particular interest is the comparability of these findings to those of the present study. Additional perspectives on power and politics influence this research. Organizational Power and Politics Little attention has been given to the relationship between power, politics and voice and no known research has yet explicitly considered the relation between power, politics and 42 silence in organizations. One contribution of this research is its study of the intersections between these traditionally independent areas. Power dependence and politics in organizations. Of the many perspectives on organizational power and politics (Harlos, 1995; Pfeffer, 1981), power dependence theory influences the present research most heavily. This theory is rooted in exchange-based models of reciprocity between people using power as a commodity with costs and rewards which affect the type and frequency of exchanges between agents and targets (Homans, 1961). Emerson (1962) proposed that social relationships are based on mutual dependencies and that power is inversely proportional to the degree of dependency; the more that A is dependent on B, the more power B has over A. In addition, Blau (1964) maintained that asymmetrical dependence is a precondition to power in social relationships. In the present study, I view organizational hierarchies as systems of workplace interpersonal relations which differentiate among individuals on the basis of their authority, status, role and which prescribe interactions (Scott, 1975). In particular, I assume that the power associated with varying hierarchical levels influences how employees experience injustice and how they choose to respond. Indirect support for this position stems from research which connects hierarchies with resources. Specifically, an important source of asymmetrical power relations of hierarchies is the differential access to information, opportunities and resources among members of various organizational groups (Kanter, 1977; Maneiro, 1986). In this study, I also attempt to assess the validity of Hirschman's (1970) claim that voice is a political response to dissatisfaction and I explore the role of politics in voice and silence. Politics is conceptualized here as activities intended to acquire, develop and use 43 power to obtain one's preferred outcome when there is uncertainty or disagreement about choices (Pfeffer, 1981). Dependency in sociocultural power relations. In addition, Scott's (1990) study of voice and silence in sociocultural power relations influences this research. He argues that the exercise of power creates public and private transcripts for both dominant and subordinate groups. Public transcripts are verbal and non-verbal behaviours expressed in the other's presence while private transcripts are those behaviours excluded from the other. The discrepancy between subordinates' private and public transcripts is believed to indicate their perceived powerlessness and dependency. Like Rothschild and Miethe (1994), Scott adopts a political perspective in the analysis of voice, suggesting that the form that voice takes varies according to the capacity of powerholders to punish open resistance. Thus, voice as resistance to domination is often disguised and muted for safety's sake. Moreover, contrary to the conceptualization of silence as endorsement in other related literatures, Scott's (1990) work provides theoretical support for the conceptualization of silence as objection. Thus, it is possible that the dissatisfied employee whose public transcript indicates smiling submission but whose private transcript reveals fantasies of aggression against a supervisor perceived as tyrannical feels very powerless. In my research, I examine the gap between employees' public and private transcripts, its relation to silence and powerlessness and its consequences by means of a projective exercise (further described in Chapter 2). General Summary and Propositions As previously noted, the breadth of disciplines and knowledge drawn on for this study provides a rich foundation for the synthesis needed both to build organizational theory and to 44 inform its practice. The literature review points to three main lines of inquiry, summarized here as follows: 1. Descriptions of Organizational Injustice a) exploration of employees' experiences of organizational injustice b) role of contextual conditions (e.g., characteristics of work settings) in injustice experiences 2. Responses to Organizational Injustice a) concept development of voice and silence as distinct and significant responses to dissatisfaction in general and injustice in particular b) voice and silence as dynamic processes in response to workplace injustice 3. Experiences and Assessments of Voice Systems a) examination of employees' experiences with and beliefs about voice systems, including reasons for not using them despite their availability. b) exploration of employees' voice system expectancies of and subsequent satisfaction with systems to study possible fair process and frustration effects. c) exploration of the role of power and politics in relation to voice and silence. Preliminary Conceptualizations of Voice and Silence I used a priori, working definitions of voice as speaking up and of silence as staying quiet in response to organizational injustice. These are seen as preliminary definitions intended initially to guide research and subsequently to be refined by its findings. Propositions Taken together, these lines of inquiry and the research goals and questions identified earlier suggest the following preliminary propositions to guide data collection and analysis: Proposition 1: Voice and silence as responses to organizational injustice are multidimensional concepts and enjoy multiple meanings within and across organizational members. Proposition 2: Voice and silence represent dynamic processes in response to organizational injustice. 45 Proposition 3: Attributes of effective voice systems are both particular and universal; characteristics of individuals and groups of organizational members influence efficacy. Proposition 4: Voice is associated with feelings of power whereas silence is associated with powerlessness. Proposition 5: For members who choose silence, the gap between their public and private transcripts will be greater than for those who choose voice. Proposition 6: Both voice and silence may result in either (or both) fair process or frustration effects. These propositions are used to explore relationships which may emerge from the data and to facilitate the interplay between induction and deduction, proposing and checking, that is characteristic of data analysis using grounded theory (Strauss & Corbin, 1990). In the next chapter I outline the specific research methods used in this study. 46 CHAPTER II RESEARCH METHODS A. Overview Research strategy As Chapter 1 indicates, both the phenomenon of organizational injustice and responses to it remain poorly understood. It appears, at present, that the research goals of description and explanation are more appropriate and compelling than goals of prediction and control to better understand the central concepts and processes associated with experiences of and responses to organizational injustice. Qualitative inquiry is well-suited to address the general goals of description and explanation and to pursue the specific research goals and questions outlined in Chapter 1. Unlike quantitative approaches, qualitative research designs are typically inductive; they are intended not to prove or test existing theory but to allow theory to emerge once data are collected (Denzin & Lincoln, 1994; Patton, 1990). The present study is an example of naturalistic inquiry with its emergent research design, purposive sampling to examine generic processes (Miles & Huberman, 1994) and use of non-experimental research settings. Specifically, this research employs a mixed sampling strategy (e.g., purposive, snowball) using multiple data sources to provide needed flexibility and to support the triangulation or convergence of findings (Eisenhardt, 1989; Jick, 1979). It should be noted, however, that purposive sampling does not permit generalizations to other settings or populations; instead, generalizations are to existing or new theories of target process(es) and thus are essentially analytic (Miles & Huberman, 1994). In this study, I use grounded theory 47 to analyze data from two sources: a) interviews with employees who have experienced organizational injustice; and b) archival cases of employees who voiced discontent through a government-sponsored voice system. Thus, this research is an empirical inquiry into a contemporary phenomenon within its real-life context with unclear boundaries between the phenomenon and its context and with multiple sources of evidence; as such, this research is also a collection of case studies (Yin, 1984). Criteria for Research Soundness As a naturalistic inquiry, this study was designed to meet qualitative standards for research soundness. Although there is no universally-accepted set of standards, in this research I attempted to meet the following four widely-used criteria, roughly parallel to traditional quantitative research criteria of reliability and validity: 1) credibility - assurance that the subject or phenomenon is accurately described; 2) transferability - generalizability to other settings (or, in this case, theories); 3) dependability - an accurate accounting for changing conditions in the phenomenon of study and/or social world; 4) confirmability -whether findings can be confirmed by another (Lincoln & Guba, 1985). To enhance the credibility and confirmability of my findings, I tape-recorded all interviews (McCracken, 1988) and supplemented them by field notes of participants' nonverbal behaviour and, at times, of my own emotional reactions following interviews (Kirk & Miller, 1986). Confirmability was also augmented by using the paper or audit trail strategy to document the transformation of raw data into findings by including quotations and excerpts, for example, to allow readers to assess the adequacy of connections between the data and subsequent interpretations (Lincoln & Guba, 1985; Marshall & Rossman, 1995). Individual researcher bias, which threatens credibility and confirmability, was addressed by 48 determining the inter-rater reliability of the emergent coding protocol for classifying types of organizational injustice using two independent raters. Moreover, the criterion of dependability was supported by the diversity of participants' experiences of injustice and their respective organizational settings. Qualitative research standards also recognize the role of the researcher's skills on issues of reliability and validity. As Miles and Huberman (1994) argue, researchers who are familiar with the phenomenon under study, who have strong conceptual interests, a multidisciplinary background and good investigative skills to draw out participants enhance the investigator-as-instrument process (Guba & Lincoln, 1981) by which the researcher, through his or her experience, imagination and intellect, assists in the collection and analysis of data. In several ways, I believe that I satisfy these indicators (discussed further in subsequent sections of this chapter). B. Theories of Method Paradigm of Choice: Postpositivism Of the paradigms which guide disciplined inquiry, this research relies on a postpostivist perspective (Guba, 1990) which can be summarized as follows: 1) ontologically, it proposes a critical realism - that is, reality exists "out there" but can never be completely understood; 2) epistemologically, it takes a modified objectivist stance -objectivity is an ideal but can never be fully achieved because the inquirer's influence can never be completely separated from the inquiry; 3) methodologically, it relies on multiple methods and "...inquiry in more natural settings, using more qualitative methods, depending more on grounded theory and reintroducing discovery into the inquiry process" (Guba: 23). 49 Thus, unlike positivism's realist ontology that there is one "true" reality driven by immutable natural laws which can be discovered through objectivist epistemology to ultimately predict and control natural phenomena through experimental empiricism, postpositivism is, at least in part, inherently subjective and reflexive. It is also a somewhat controversial approach to understanding the nature of scientific knowledge. While some view it as synonymous with positivism (e.g., Guba, 1990), others view postpositivism as significantly different from its philosophical precursor, citing the diversity of emerging postpositivistic streams (e.g., Phillips, 1990). Regardless of the particular stream, the question of qualitative validity, by which the authority of the research is established on the basis of legitimized truth, is a central issue within postpositivism (Denzin, 1996). Although there is disagreement about the specific criteria by which postpositivistic research evaluates its claims to legitimation or validity, Hammersley (1992) cites four postpositivistic goals which were met, in varying degrees, by the present research: 1) theory generation; 2) empirically grounded and scientifically credible theory; 3) generalizability to other settings (in this case, other theories); and 4) consideration of the researcher's and the research strategy's influence on findings. In short, I am committed to produce research which provides 'valid' accounts, defined as plausible, credible and relevant (Hammersley, 1992). Because of the considerable diversity in the conceptual and methodological positions within the postpositivist perspective, I will further clarify my own position by adapting Guba (1990): ontologically, unlike the relativist who believes that all differing and contradictory views of the same reality are valid, as a critical realist I believe that there is one view of the same reality, even though I may never fully comprehend it. However, I do accept multiple 50 views of different realities (Phillips, 1990). Epistemologically, I am more subjectivist than objectivist; while I strive for careful, precise and unbiased scholarship, I recognize that I probably can never fully escape the influence of my own values on the inquiry process. When necessary, I have identified these values which then become part of the framework of and limitations to the present research. Lastly, methodologically I am motivated in this research to facilitate transformation (Guba, 1990) by exploring how injustice (as a specific form of dissatisfaction) can be used for individual and organizational improvement (Staw, 1984). Thus, this research combines two of three positions underlying qualitative research; it is both problem-focused in its effort to improve organizational practice and concept-driven in its desire to develop the concepts of voice and silence (Wolcott, 1992). More broadly, this research also integrates distinct and historically disparate literatures and theoretical perspectives to generate knowledge by combining, within a postpositivist paradigm, research streams of E V L N , organizational justice, whistle-blowing and organizational power and grounded theory. Ethnographic Approaches to Qualitative Research Qualitative inquiry comprises several distinct approaches each with differing philosophical perspectives and methodologies. These approaches include phenomenology, hermeneutics and ethnography (for detailed reviews see Atkinson, Delamont & Hammersley, 1988; Jacob, 1987). The ethnographic approach aims to interpret and explain specific aspects of the life of a particular group through anthropological description, naturalistic research and participant observation (Hammersley & Atkinson, 1995; Miles & Huberman, 1994). As adapted here, it is concerned with understanding the meaning of situations (i.e., 51 organizational injustice) held by relevant social actors (i.e., those who have experienced it) through the study of their cultural sensemaking within a particular cultural context (i.e., 'culture of injustice'). In the present study, the research goals and design preclude what are arguably more traditional methods of ethnographic inquiry, such as immersion in one culture through observation of and extended close connection with participants. That is, to generate theory a sampling strategy which captured a range and depth of organizational injustice experiences, rather than immersion in any one organization, was required. Grounded Theory In this study, the ethnographic research method I use is grounded theorizing (Glaser & Strauss, 1967) to investigate employees' experiences of and reactions to perceived unfair treatment at work. A grounded theory is one that is inductively derived from the study of the phenomenon it represents. It generates testable, relevant and valid (i.e., legitimate) theory using data to generate connections with emerging concepts and hypotheses (Strauss & Corbin, 1990) and is well-suited to the study of processual experiences (Morse, 1994) across disciplines, including organizational research (see Martin & Turner, 1986; Turner, 1983). Recent applications of grounded theory to organizational analysis include the study of punishment (Butterfield, Trevino & Ball, 1996), leadership (Hunt & Ropo, 1995), technological change (Prasad, 1993) and organizational caregiving (Kahn, 1993). Grounded theory uses the constant comparative method to systematically code data into as many themes and meaning categories as possible. Data analysis is an iterative process: as categories emerge, the researcher considers and re-considers their validity through connections to the data and theoretical implications. Strauss and Corbin (1990) describe it as repeated movement between inductive and deductive thinking, or a constant interplay 52 between proposing and checking. Thus, grounded theory results from patterns which emerge from this recursive process of analysis, comparison and interpretation. C. Procedure Data Collection Interview Cases Selection. Several populations whose members were likely to have experienced organizational injustice were targeted for recruitment. Selecting samples from populations whose members intensely manifest the phenomenon in question is a standard purposive sampling strategy intended to provide information-rich cases (Patton, 1990). In the present research, the selection of populations was also guided by theoretical sampling (Glaser & Strauss, 1967), increasingly used in inductive organizational research (e.g., Eisenhardt & Bourgeois, 1988; Gersick, 1988) to strategically select diverse samples and cases. Its goal is to maximize the number of emergent categories by selecting both similar and dissimilar cases to compare and contrast categories, to integrate categories and to delimit the scope of the emerging theory by finding cases where categories are not relevant or upheld. As well, in theoretical sampling the choices of participants and/or events are driven by conceptual questions; the primary concern is not with representativeness or generalizations of findings to other settings, but with the conditions under which the theory or construct(s) operates (Miles & Huberman, 1994). Here, theoretical sampling and, by extension, theory generation, are enhanced by the variability across both organizational settings and by the nature of participants' perceived injustice. In addition, studying experiences of both past and current unfair treatment by 53 including participants who are at differing points or moments in their processes of responding to organizational injustice supports theoretical sampling. Recruitment. The process of recruiting participants involved several stages during the fall of 1996 and the winter and spring of 1997. Prior to recruitment, the interview protocol was pilot tested with two participants in the summer of 1996, resulting in minor modifications to structure (i.e., order of questions) and style. The interview protocol is provided in Appendix A. 13 I initially intended to interview between 10 and 15 cases each of male and female participants drawn from the population of 73 part-time graduate business students at the University of British Columbia (UBC). Most of these students had five years' experience in the workforce and most were currently employed full-time. Because contact with potential participants was made away from their respective organizations, I could provide organizational anonymity, which I assumed would enhance the response rate. To encourage participation, I also secured donations from two local restaurants to allow participants the opportunity to win one of two gift certificates worth $75 each. In the first stage of recruitment, I contacted the part-time graduate business students through a combination of personal appeal (i.e., visiting classrooms), mail and telephone follow-up. Recruitment letters which outlined the purpose of the study, expected time commitment, confidentiality and participants' rights were sent to all students (see Appendix B), followed by a telephone call to confirm delivery, answer questions and determine interest I tended to interrupt pilot participants as they described their injustice experiences to keep them "on track" with my preferred order of questions. However, I found that this practice blocked the emergence of experiences by imposing an artificial and arbitrary linearity. I then adopted Mishler's (1986) advice and stopped interrupting. 1 3 A estimate which falls within accepted limits for multiple case studies (Eisenhardt, 1989). 54 in the study. Approximately 22% of the students could not be directly contacted. Of those contacted, a substantial number (58%) declined participation. Of those who declined, the overwhelming majority (86%) cited no experience with injustice at work; most were supportive of the research itself, some even apologetic about not being able to help and many wished me well. Of the remainder who declined participation, several cited lack of time while a few did not provide a reason. In total, 15 students agreed to participate. Because of scheduling conflicts, 11 interviews were completed, representing a response rate of 15%. In the second stage of recruitment, I visited several classes at the British Columbia Institute of Technology (BCIT) to draw from a population of 1,700 part-time graduate business students from the Administrative Management Diploma Program believed to be comparable to the U B C sample in terms of age and prior work experience. From these brief classroom solicitations, five interviews were conducted. At the same time, I also used a snowball sampling strategy to secure six interviews through references from U B C and BCIT students who believed them to represent information-rich potential participants (Miles & Huberman, 1994). Finally, 10 interviews were completed following a call for participants from the general adult population in an article describing the study published by a local community newspaper. This appeal to the broader community was designed, in part, to reduce the potential for managerial bias by responding to Martin's (1993) call for including members of economically and/or socially disadvantaged groups (e.g., blue-collar workers, women, victims of discrimination) in organizational research. A copy of this article is provided in Appendix C. 55 Participation was restricted (with one exception ) to those who reported direct experiences of injustice themselves. This restriction minimized (if not eliminated) the potential for invalid assumptions of participants' prior injustice and reduced possible confounding factors. Another restriction was the availability of at least one voice system in participants' respective organizations. A survey list of nine voice systems from Robinson's (1994) adaptation of Spencer's (1986) measure to assess the availability of voice systems was administered to participants (see Appendix D). In sum, these samples were compelling choices for the inductive, qualitative investigation of organizational injustice and related responses. Taken together, the final aggregate sample of participants was neither intended to be randomly selected nor expected to be representative of any one organization, group or population. Voluntary self-selection supported the assumption that participants would be candid and honest, perhaps more so than those who may have been coerced or pressured into participation. Moreover, because statistical inference was not a research goal, this non-random, non-representative sampling strategy was not considered methodologically problematic. Supplementary documents. In many instances, participants offered documents to confirm their experiences of injustice, including letters of appointment, notices of termination, internal memos, organizational charts, a newspaper article and, in one case, reference to a published book written by a CEO to describe 'how not to' restructure; the downsizing initiative reportedly devastated both those laid off (including the participant) and those who remained. In another case, a participant's fear of retribution (either from those The exception was one participant whose witnessing of and involvement in the injustice experienced by a fellow employee significantly affected her sense of security and morale in the workplace. 56 with whom he worked or others in the industry) if he were to redress the perceived injustice was later substantiated by a business periodical article which outlined a number of recent murders, suicides and at least one disappearance in the larger association to which his bosses belonged (Cannon, 1997). In several cases involving arbitration and/or litigation, participants made original, hard copy transcripts of hearings and/or case numbers available to me for further study. In summary, the supplementary documentation provided by participants to support the credibility and legitimacy of their experiences was substantial. Archival Cases Interview data were also supplemented by archival data; specifically, I conducted a quasi-random sampling of archival cases published from 1982 - 1996 in The Canadian Human Rights Reporter. This periodical reports cases of alleged violations of provincial and federal human rights legislation which required resolution by a tribunal. Cases published over the last 15 years were selected for inclusion. Across these cases, the dates of occurrence, of reporting and of the decision were noted. Selection. To select archival cases for analysis, I first identified three 5-year periods of publication dates to draw on cases from the last 15 years: 1980 - 1984; 1985 - 1990; 1991 -1996. Within each period, I then randomly chose one year (1982, 1987 and 1992)15 and selected 10 cases from each year for a total of 30 cases. For the 1982 cases, the complaints' dates of occurrence and reporting ranged from 1978 to 1981 while the date of the decision ranged from 1981 to 1982. For the 1987 cases, dates of occurrence and reporting ranged from 1979 to 1985 while the date of the decision ranged from 1984 to 1987. Finally, for the cases published in 1992, the dates of occurrence ranged from the mid-1970's to the late 1 5 While these target years appear systematically chosen, they were in fact selected blindly and independently. 57 1980's (with most occurring in 1988/89), the dates of reporting ranged from 1982 to 1989 (with most in 1988) while all decisions were handed down in 1991. Within these groupings, I chose cases which were representative across both experiences of injustice and geographical locations. For example, to further supplement theoretical variation, I selected archival cases for geographical representation by drawing on those from eastern, central and western Canada. This is consistent with the geographical representation in interview cases, although to a lesser degree (described later). Scope of data. Although the published cases varied in the nature of the reported injustice and the detail with which cases were described, most summaries provided a review of evidence and relevant law, as well as the adjudicator's decision and rationale. While the archival data did not reflect the richness and depth of interview data, it did offer unique breadth (both by topic and geographical location of the case), thus enhancing theoretical sampling. In sum, because the archival data offered limited emotional and psychological information from complainants' experiences with and responses to workplace injustice, interview data were relied on more heavily throughout this chapter to illustrate findings and insights. This limitation notwithstanding, it is to be understood that the inclusion of archival data provided a strong, triangulated measure against which to ground and test emerging findings from the interview data. Exclusions. I excluded cases from Quebec because the majority were published in French and I am not bilingual. As well, I excluded cases which did not represent issues of organizational functioning and which did not involve an extant employment relationship. 58 Thus, cases concerning refusals of employment, housing issues, access to public services and jurisdictional questions were not included. Characteristics of the Interview Sample Individual-Level Characteristics Gender and age. Thirty-two participants from both managerial/professional (salaried) and clerical/line (waged) positions, each representing different organizations, described 33 cases of injustice in research interviews.16 The sample comprised 19 women and 13 men ranging in age at the time of the interview from 23 to 71 with an average age of approximately 37 years. The participants cited experiences of perceived unfair treatment which occurred from 24 years previously to current situations to which they returned immediately following the interview. The average age of participants at the time of the perceived injustice was approximately 33 years and 7 months. Date of occurrence. On average, these experiences were reported to have taken place 3 years and 5 months in the past. Across the aggregate sample, participants described experiences of injustice which occurred at varying times throughout their tenure at the organization. On the whole, organizational tenure was slightly over 3 years. For seven participants, perceptions of injustice began immediately upon joining their respective organizations. For the remainder, experiences of unfair treatment were reported to begin slightly after 2 years' employment. Employment status. Approximately eight participants remained employed with their organizations, seven were dismissed or laid-off, two were on medical leave and 15 had quit. 1 6 One participant reported two cases of injustice which occurred in two different organizations. 1 7 One participant had been injured on the job and the other was on long-term disability from job-related stress. 59 Of those who had quit, 11 cited the injustice as the primary reason for leaving; for four others, it was acknowledged as a secondary but contributing factor in the decision to quit. In one case, a participant's departure was unrelated; his two-year contract with the organization expired. Additional demographic data. Participants also represented all levels of educational standing ranging from Grade 10 equivalency (one) to a doctorate (one) at the time of the interview. For most participants, the highest education level attained at the time of the interview was a graduate business or professional degree (14). Nine had an undergraduate degree, three a college diploma, three a high school diploma. Ethnically, the sample was relatively diverse: 24 Caucasian, six Asian, one East Indian and one Middle Eastern participant (split proportionally across genders) volunteered. Regionally, the majority of cases involved organizations located in the city of Vancouver. However, 7 of 33 cases were set in organizations outside Vancouver: three were based in the interior of British Columbia while four occurred outside the province (two in Alberta and two in Ontario). Group and Organizational Level Characteristics Table 2 lists the range of industries, organizational levels and positions represented in this research. Across positions, approximately half involved supervision of others' work while the other half reported no supervisory responsibility. Similarly, about half the positions were team-based while the rest reflected largely independent work. Seven of the organizations described were unionized and of the participants' positions described therein, three were unionized. In three cases, union drives within participants' organizations were occurring at the time of the perceived injustice (in two of these three cases, participants were actively involved in the union drives). 60 TABLE 2 POSITION TITLES OF RESEARCH PARTICIPANTS BY INDUSTRY,* HIERARCHICAL L E V E L AND GENDER** Industry Professional/Managerial Clerical/Line Male Female Male Female Forestry Engineer (1) Manufacturing Technical Writer (1) Computer Programmer (2) Sales Agent (1) Construction Engineer (1) Office Assistant (1) Wholesale Trade District Sales Manager (1) Merchandiser/ Sales Agent (1) Finance Sales Assistant (1) Real Estate Operator General Manager (1) Business Service Architect (1) Contract Administrator (1) Engineer (2) Clerk (1) Counsellor (1) Government Service Health Officer (1) Education Service Assistant Curator (1) Manager (1) Health and Social Service Executive Director (1) Administrator (1) Field Executive (1) Project Coordinator (1) Social Worker (1) Clerk (1) Counsellor (1) Accommodation Tour Director (1) Other Engineer (1) Records Analyst (1) Manager (2) Registration Officer (1) Determined using the Standard Industrial Classification, Statistics Canada, Ministry of Supply and Services Canada, 1980. The number of female participants totals 20 (rather than 19) because one participant reported two cases of perceived injustice in two different positions at different organizations. 61 Organizations ranged in size from small ones with two employees in a private non-profit society, for example, to large ones with over 5,000 employees in a hospital. Fourteen organizations had between 1 and 100 employees, 10 organizations had from 101 to 500 employees while at least six organizations employed over 500 employees.18 Five organizations were based in the public sector; 6 of the remaining 28 organizations were non-profit, private-sector based organizations. Instrumentation Interviews Rationale. The interview was selected as the primary data collection method for several reasons. First, interviews allow large amounts of expansive and contextual data to be gathered quickly (Marshall & Rossman, 1995). Second, with assurances of personal confidentiality and organizational anonymity, I expected that potential participants would be less resistant to a personal interview than to a more impersonal paper-and-pencil survey. Also, my training and experience in both therapeutic and research interviews — ranging from mother-child observations in homes to assessment of psychopaths in a medium security prison — have given me a familiarity and expertise with this methodology and its emotional demands for both the researcher and participant. In summary, both the sensitivity of this research and the need for theory generation (as opposed to theory verification) support the use of interview data analyzed by grounded theorizing. Design. The design of the semi-structured, open-ended interviews included two methods: 1) a retrospective critical incident technique which asked participants to describe their responses to organizational injustice, including their motivations, feelings and related Data on organizational size was not available for 3 of 33 organizations. 62 behaviours, and 2) a projective exercise based on Scott's (1990) concept of the private transcript which asked participants to imagine that they could now speak with impunity to the person(s) involved or responsible for their perceived injustice. Interview protocol. Interviews took place at a mutually negotiated time and location and ranged from 45 to 180 minutes in length (averaging approximately 90 minutes) and spanned five phases. In the first phase of the interview, I reviewed the purpose of the study and contents of the Consent Form (provided in Appendix E) and I obtained background information about participants' organizations (e.g., size, type of industry) and positions (e.g., reporting structure, team-based vs. independent work). In the second phase, I simply asked participants to tell me what happened by describing the incident or situation they considered unjust. Depending on the information provided, probes outlined in the interview protocol were used to identify and/or clarify how participants responded, why they chose certain responses and with what results. As well, questions about the workplace culture and norms were asked (e.g., "describe what it was like to work there"; "what's the worst story of injustice that you heard"). Unlike standard survey interviews, this semi-structured format allowed experiences or stories to emerge by permitting participants to speak in their own voices, to control the reported order and focus of their responses and to extend their responses at will (Mishler, 1986). (The interview protocol is provided in Appendix A.) The third phase of the interview was a projective exercise which asked participants to imagine that they could now speak with impunity to the person(s) responsible for or involved in the injustice and to ask or say whatever they wished. The fourth phase of the interview concerned the availability and efficacy of organizational voice systems; of those systems 63 present in their organization, I asked participants to describe the most effective one(s) and to consider how those systems identified as ineffective could be improved. To close the interview I obtained demographic information (outlined in Appendix F) and asked if there was anything I didn't ask that I should have. In most cases, participants responded that they could think of nothing more to say.19 In this fifth and final phase, I was very attentive to participants' emotional state. Many had experienced and several had expressed a range of deep feelings during the interview — anger, sadness, outrage, grief, shame and hate. I asked all participants how the experience of the interview itself was and how they felt at its close. Where necessary and appropriate, additional time was spent de-briefing. Methodological Limitations Self-report data. Of greatest significance to this research are possible limitations from the use of self-report methods. Despite their prevalence, self-report data are widely criticized for their questionable accuracy and validity (e.g., Nisbett & Wilson, 1977; Podsakoff & Organ, 1986). Fundamentally, it is impossible to know with absolute certainty whether participants are saying what they really think or if self-report data are completely believable. However, this study views response bias as less an issue to be eliminated or reduced than an issue to be acknowledged and incorporated into the investigative process in a meaningful and credible way (see Averill, 1985; Caproni, 1991). Thus, social desirability and underlying social norms merit investigation as influential aspects of the phenomenon in question. Other research suggests that the accuracy of self-reports is increased when 1 9 One participant suggested I ask about the use of intimidation tactics, possible ill-feelings toward the organization and whether she would return if changes were made. These were subsequently discussed in the final phase of the interview. 64 influential stimuli are salient and are plausible causes of the responses they produce (Nisbett & Wilson, 1977). Here, the accuracy of this self-report data is expected to be enhanced by the high salience and plausibility of the stimuli associated with the emotionally-charged nature of the research topic. These clarifications notwithstanding, the focus of this research is on current perceptions of injustice rather than on the actuality of past experiences. In this study, problems of recall and accuracy are secondary to the meaning(s) of participants' descriptions of perceived injustice and the questions they raise. Consistent with postpositivist assumptions and in the absence of willful deception by participants, the self-report data collected here is neither intended nor able to produce a complete version of the truth of processes involved in responding to perceived injustice at work. Instead, the knowledge generated from this data reflects what are inherently partial rather than complete versions of multiple realities of employees' experiences. Other modes of inquiry, different theoretical frames and alternate analytic procedures using this data are expected to reveal equally valid, if divergent, interpretations of the data (Coffey & Atkinson, 1996). Interview method. One weakness with the interview method employed in this study concerns the use of a single interviewer. The use of multiple interviewers in qualitative research is widely accepted as one way to control for individual bias in interview responses and interpretation (e.g., Miles & Huberman, 1994; Patton, 1990). However, limited resources preclude the use of multiple interviewers in this research. While the potential for single interviewer bias may not be completely eliminated, Marshall and Rossman (1995) outline several practices to minimize effects of interviewer bias, two of which were followed in the present study: engaging in a constant search for counter-examples or discrepant 65 instances and continuously examining the data for possible rival hypotheses. As previously noted, this study's assessment of inter-rater reliability of the emergent typology of organizational injustice was one systematic means by which interviewer/researcher bias could be estimated. When interviewer bias is controlled for (as far as possible), a potential advantage to the use of a single interviewer lies in the investigator-as-instrument process (Guba & Lincoln, 1981) defined earlier as ways that the researcher, through his or her experience, imagination and intellect, assists in the collection and analysis of data. McCracken (1988) describes it more fully as a matching process between ideas and actions from the researcher's own experience to those the participant has described. In this way, the researcher uses his or her subjective experiences to find parallels with emerging patterns in the data which then require substantiation and repeated confirmation before being accepted as evidence. In at least two important respects, I am likely to be a fairly close match to participants, thereby increasing the potential benefit from the investigator-as-instrument process. First, I am approximately the average age of participants. Second, I have six years' full-time work experience and over 10 years of part-time work experience in several capacities and in a wide variety of organizations. Within this work history, I have experienced several incidents of organizational injustice in response to which I have both spoken up and remained silent. Lastly, there is no evidence to suggest that my gender had a detrimental effect on either the investigator-as-instrument or the interview process itself (see Morton-Williams, 1993). While matched-gender interviews might increase a male interviewer's familiarity with male participants, female interviewers are reportedly more likely than males to be perceived 66 as friendly (Morton-Williams) and in fact may facilitate the interview process by increasing feelings of safety and comfort among all participants (Caproni, 1991). D. Data Analysis Overview Qualitative data analysis can be broadly understood as a three-part iterative process involving data reduction, data display and conclusion drawing/verification (Miles & Huberman, 1994). Data reduction refers to the process of selecting, simplifying and transforming transcribed or archival information and includes decisions about the portions of data to be coded and the patterns which best describe those portions. More informally, Wolcott notes "the critical task....is not to accumulate all the data you can, but to can (i.e., get rid of) most of the data you accumulate" (1990: 35). Wolcott's implicit directive, followed in this research, is to reduce data while satisfying qualitative research criteria for soundness and maintaining consistency with the ontological, epistemological and methodological assumptions of the research paradigm and its goals. Following data reduction, data displays involve decisions concerning the structure of data matrices, graphs or charts and which data should be entered into the display. From data displays, preliminary conclusions can be made, which may require modifying the data display. Unit and Levels of Analysis Unit. In this research, the primary unit of analysis was each of 33 cases of organizational injustice reported by participants. Of the cases reported, one was excluded: the case concerned perceived mistreatment by a person whose only connection to the participant's organization was co-leasing office space with the participant. While arguably no 67 less unjust, in the absence of either an employment relationship with the organization or a formal work relationship with the participant, this case was seen as largely personal rather than an organizational concern and was therefore excluded from the study. A secondary unit of analysis was each of the 30 archival cases of organizational injustice; these cases were secondary in that, while consequential, they figured less prominently in analyses. Levels. To pursue the goal of generating an integrative theoretical framework of variables involved in experiences of and responses to organizational injustice, a wide range of factors was considered. Accordingly, this research considered multiple sources of variables — psychological, social and organizational — across multiple levels of influence — individual, group and organizational — involved in employees' experiences of and reactions to perceived unfair treatment. It is a synthesis of micro and macro theory and research in the study of organizational processes which spans at least two levels of analysis, and as such this study represents 'meso' organizational research (House, Rousseau & Thomas-Hunt, 1995; Rousseau & House, 1994). Analytic Approach To prepare the interview data for analysis, the 32 tape-recorded interviews were first transcribed verbatim and then stored in computer files, resulting in over 1,400 pages of transcripts. Each case was assigned a unique case number. Typically, data concerning a given question or issue was not contained in one place either within or across interviews, and so locating relevant portions of data was required. Thus, after initially selecting interview extracts and storing them in computer files, working directly from hard-copy transcripts proved a systematic way to conduct preliminary analyses. 68 Consistent with the inductive, emergent approach required by grounded theory to analyze data, the specific analytic approach and techniques were unclear at the start of data analysis. However, one a priori approach which I attempted to follow was to roughly separate participants' experiences of injustice into broad categories of speaking up (i.e., voice) and staying silent with the intent of producing three descriptive profiles of responses — voice, silence and mixed — which were to be analyzed separately (within-profile) and then comparatively (across profiles). However, the clarity and consistency of participants' responses did not permit this analytic approach. Very few participants (two to three) clearly and consistently voiced their discontent or maintained their silence throughout their respective employment relationships. Instead, most participants engaged in a variety of responses which did not easily lend themselves to classification as speaking up or staying quiet. I then determined a new analytic approach by continuing to read and re-reread cases until patterns emerged which, along with the research questions and a priori propositions identified earlier, resulted in an overall analytic approach involving four iterations or phases. The first phase was oriented toward individual-level considerations of the phenomenon of organizational injustice, including antecedents and consequences. The second focused on organizational-level considerations of organizational injustice, also including antecedents and consequences. The third phase was intended to identify the specific techniques by which participants handled or responded to unfair treatment at work and to consider their relation to voice and silence. In the fourth and final phase, I sought to generate a theoretical framework within which to synthesize findings to understand, as a process, the experience of and responses to perceived injustice at work. 69 In addition, I used several specific techniques to analyze data as described below. In completing these phases, theoretical saturation (Glaser & Strauss, 1967) guided the determination of analytic completion: when new data no longer emerges from the data, when category development accounts for all elements and their variations and when relationships between categories are established and validated (theoretically), saturation is said to be reached and analysis is complete. Phase 1: Individual-Level Experiences of Organizational Injustice From multiple readings of transcripts, I noticed that there were both broad and sometimes specific similarities in the nature or type of injustice which participants described. For example, one set of cases was exclusively concerned with participants' complaints of interpersonal mistreatment by an authority figure. This mistreatment appeared significant in and of itself (i.e., it was often unrelated to the implementation of specific policies and procedures). Conversely, other cases revealed complaints of mistreatment which were exclusively concerned with procedural issues, while others concerned not getting certain rewards or being punished unfairly. In this way, a sorting of cases which emerged as typology of organizational injustice was developed through grounded theory (see Chapter 3). Specifically, to develop this and other classifications of the data, I relied on the technique of concept cards (Glaser & Strauss, 1967; Martin & Turner, 1986) by which core concepts are identified and categories generated. Using relevant interview portions, I began by noting what appeared to be salient concepts for a given question or topic in the margins of the transcript (e.g., intimidation, cynicism, fear, scrutiny, etc.) which I then transferred to separate sheets and ultimately to index cards as the concepts became more focused. Consistent with grounded theory protocol, identifying the salient concepts was an iterative 70 process: as new data came in, I added to or revised existing concept sheets and cards and created new ones. As well, each concept was intended to reflect one theme; multiple themes for a given concept indicated that additional, separate concepts were in order. As suggested by Corbin and Strauss (1990), the preliminary categorization of data was checked against existing ones to further refine analyses. Consequently, I regularly placed iterations of my classification scheme against the literature to broaden insights and theoretical connections until the final iteration which captured all cases was reached. This technique is one illustration of the constant comparative method of grounded theory (Strauss, 1987) by which themes and meaning categories are generated by systematic but discursive movement between the existing literature and the textual data. Thus, I identified patterns as concepts or categories which best summarized or captured segments. Analysis was by paragraph rather than word or line. This application of pattern coding (Miles & Huberman, 1994) was designed to identify emergent themes, constructs and explanations using both descriptive and inferential or explanatory concepts. Phases 2 and 3 To examine organizational-level aspects of organizational injustice (Phase 2), I was guided by an analytic approach adapted from Strauss and Corbin (1990) which relates the phenomenon in question (organizational injustice) to its causal conditions and associated consequences. This approach also calls for the identification of strategies by which participants handle the phenomenon; this formed the third phase of data analysis (specific procedures and methodology are further described in Chapter 5). More generally, it should be noted that analytic strategies of concept coding and segmenting data can reduce the overall sense of the account or the 'storied' quality of the data 71 (Coffey & Atkinson, 1996). However, given its research goals and questions, concept coding was more appropriate for this research than analytic approaches which rely exclusively on the storied quality of the text (e.g., narrative analysis). To some degree, theoretical 'losses' incurred by fragmenting or decontextualizing data were offset by including multiple levels of analysis and by considering contextual conditions and properties of the phenomenon. Patterns and images. Analyses of images also addresses concerns of decontextualization by allowing both the form and content of data to be explored (Coffey & Atkinson, 1996). In this research, consideration of images was also prompted by participants' frequent and vivid use of analogies, similes and other kinds of imagery to describe their experiences of and reactions to perceived injustice. Imagery can uncover themes that coding may not reveal and can convey key features of shared cultures (Coffey & Atkinson). Phase 4: Model Building In this final phase of data analysis, I synthesized findings from previous phases to produce a conceptual process model outlining organizational injustice and responses to it. From this description of the methodological choices and research design, including procedures and data analysis, I now turn to the results from the first phase of data analysis: an examination of individual-level aspects of organizational injustice. 72 CHAPTER III INDIVIDUAL-LEVEL EXPERIENCES OF ORGANIZATIONAL INJUSTICE A. Overview This chapter is the first of three which reports the results of this study. It focuses on individual-level issues associated with unjust treatment at work by examining participants' descriptions of their experiences of organizational injustice, which included relevant cognitive, affective and behavioural considerations. This chapter contains three main sections. In the first section, I introduce a typology to reflect the scope and complexity of both interview and archival cases of workplace injustice according to four emergent categories of injustice: interactional, distributive, procedural and systemic. I then discuss these categories in further detail by describing select issues which are illustrated by interview quotations from participants. Consistent with its prevalence among cases reported here, I explore interactional injustice first, and in some depth using a set of characteristic behavioural dimensions which I construct from participants' descriptions of interactional injustice. The second section describes three emergent properties of organizational injustice — impact, location, timing — and the use of threats across the four categories of perceived injustice. In the third and final section I describe individual-level antecedents and consequences of workplace injustice over time (immediate, intermediate and long-term) for participants. Throughout this chapter, I refer to each participant's experience with organizational injustice in the singular, although it may reflect several incidents. This is consistent with 73 Weick's (1979) understanding that we make sense of incidents over time by blending or merging them into one continuous stream of experience. Sources of Data As noted, findings reported in this chapter were drawn from a combination of interview cases (33) and archival cases (30). However, these two sources of data were not employed equally in the findings reported here. Because the interview data provided richer information than the archival data about participants' emotional and psychological states, this source figured more prominently in descriptions and analyses reported in this chapter. Nonetheless, the archival data was an important source of information which provided a strong, triangulated measure on which to ground and provide support for theory emerging from the interview data. Anonymity and Credibility To protect participants' anonymity I have used pseudonyms. Appendix G lists, by case number, the pseudonyms assigned to participants and their job titles. Although the inherent subjectivity of the interview data was not considered methodologically problematic (see Chapter 2), on several occasions participants attempted to support the credibility of their reported experiences in several ways. As noted in Chapter 2, for example, several participants offered additional documents (e.g., notices of appointment and termination, reports from arbitration hearings) to substantiate their experiences and provide richer, fuller data. As well, during the interviews many participants referred to unofficial documents such as personal journals, diaries or logs kept at the time as ongoing accounts of injustice; several had re-read their written records prior to my interview with them. 74 Use of Tables Following Miles and Huberman (1994), the use of tables in this chapter is intended to reduce the vast quantity of interview and archival information in order to shed light on the research questions asked. These tables were constructed largely for descriptive rather than explanatory purposes. That is, data was partitioned to examine what was there rather than to generate explanations about why certain events or processes occurred. It is also important to note that the written text is not intended to restate or review the data displayed in tables. Instead, consistent with Miles and Huberman, the text is used to present the meaning made from tables by noting patterns and themes, making contrasts and comparisons and drawing and verifying conclusions. Frequency Estimates Throughout the written text of this and subsequent chapters, findings are commonly reported using the descriptors many, several or some, and few. Because I relied on theoretical sampling to collect data and theoretical saturation to analyze it (see Chapter 2), precise quantitative estimates of frequencies of the descriptors were neither called for nor carried out. However, I recognize that approximate frequencies may indicate the relative importance of findings. Accordingly, in this research I denote the following ranges in the number of participants when using these descriptors: many - seven or more; several or some - between four and six;/<?w - between one and three. B . A Typology of Organizational Injustice General Framework Across cases (both interview and archival) there was great variability in the nature of the perceived injustice described. Because no existing theoretical framework captured the 75 scope of experiences reported here, I combined categories from existing classifications with ones emergent here to introduce a comprehensive set of categories which reflected variations in the experiences of injustice as described by participants. These categories are viewed as prototypes (Rosch, 1978) and thus are conceptually distinct but indiscrete categories with somewhat indistinct boundaries. The four categories are intended to be exhaustive of the injustices described across the interview and archival cases. The underlying philosophical approach to the generation of this typology is most closely aligned with nominalism, wherein categories are seen as socially-constructed, dynamic entities (Rich, 1992) which may change over time and across cultures. Within this approach, the procedure I used to classify experiences of injustice reflects a combined nominal-theoretical with nominal-empirical approach (Rich, 1992) in which data and emergent classifications were guided by and linked to organizational justice theory in an inductive, recursive process of data analysis and interpretation. Specifically, I identified four major categories of perceived injustice that are supported by a synthesis of academic (Bies & Moag, 1986; Sheppard et al., 1992) and, to a lesser degree, popular20 (Dumaine, 1993; Wright & Smye, 1996) literatures on organizational injustice. The four categories of perceived organizational injustice are as follows: interactional, distributive, procedural and systemic. These categories reflected single or repeated experiences that participants labeled and described as unfair. They are defined and described more fully in Table 3 which outlines, within each category, the broad issues involved and which provides examples of each issue. These issues are illustrative of rather Given the paucity of descriptive accounts of and frameworks for workplace injustice, this non-traditional source of data for academic research has been included for its rich, if preliminary, portraits of discontent due to perceived mistreatment by organizations. 76 U H GO P O H S3 O « O to O o o -o H ca W c o '3 ts a = a £ -"a oo ? -5 >^ 13 •a -3 R "S C L is-3 a s •S 3 a c . cu a -S .5 R 3 a ^ Oo 1 -s "5 o = R a cu S •s g a • I f a *o T •* J 0 0 R V 3 > CJ a ^ I § cu a * 13 I cu Cl, -a 3 s ^ a is 13 cu - 3 g * O c? 3 13 I ! C3 cu 3 ~ •S 2 to ^ 53 -13 53 13 53 -R C o a cu cu <i> -R 53 a 1? po S .2 op E ^-S 2 - 3 S3 -a J = .5 c ~ ^ -a o v. C ^ _r 3 co ca O Cu 3 X •c o 53 <U lU g 8 I & 8. a <u .3 * l h cu "53 ^ c .R -Si a £ R cu I a .1 cu R to "5 •a?-5 •I « § • i : "st -5= i2 r 3 « S. oo — ^ X ) 3 oo 9 M CJ ca c c o O in S2 o cu o. - 3 ' bo Cu CJ •c cj 00 ca ° Q —, >. S 3 ca jo o , bo c u a c o c 3 T. S E 8 "c CJ CJ CJ Cu ca cs CJ O CJ ca 3 ca C3 » o cu B o cu J S o s 5> ? s •5 £ s ~ 3 Jf S o -S F 1 1 R 3 O C cu §• •Si s a 1 ^ C S > o - cj -Cj a •S s <3 5" Cj C3 0 0 s •o e 0 0 S •S s C3 c 0 0 C) •S S = O cu cu ^3 -c a — *" -9 Cj 13 XI e S oo co Q -• X3 o c Is C3 Cl cu * ? •S >> R C3 cu 0 0 R 8 -S o o | oo C3 . « -R Si 3 O •a o •c CJ Cu 2 S to o 1 -rt o S ° 3 CT o "xs ~ V 13 « R oo y C R cu - ° 0 0 -a" -5 -C3 cu ». -c> a "I ^ 53 •is >^ 13 cu oo cj 0 0 R •S o & p « I R O cu ^ "I"« S R R « ftj N - ? > • - ; 13 ^ R .cu cu •o R « s - ^ i5-"S 13 »» R •s « a O °<> c, * cu -§ -^s "? R ^ J S -R c is I M Oo a e ju -cu cu 5 « -is •S "a S a W cu il ^ 3 ~ > i i : Si • 13 O 53 0 0 cu 1.1 S 3 !-§ « 53 S R R 3 Oo ^ 53 >-oo cu R O £^  53 O = . is .a cu O J 3 00 ^ T3 u* ca & CJ u* C CJ > CJ •a o 6 o Cu "S! CJ > o c o to 'N ca to 13 har C ferre har ferre o do rans due to ien sal o ca CJ cj to ca c i_ o cd CO CJ '3 00 •3 3 ca o Z t i . 3 •3 ^ > -C Oi Pu o E o E S to 3 5 to O a •a S CJ o c CJ .2 w S > -5 3 to g o g H ? 1 a o 8 | » P •r* 3 77 1 § £ CJ I" •5 1 « 5-Cj S 5 cu a s — "C3 o e ~ a -s 3 Oo o 'C ^ •S .S « S £ Si « £ s -a a •« a -o c -§ c O cu i t T3 £ B o c -o 0} 03 a. 5 05 00 Cj N? "S -5 ti .« ^-a ^ ^ •« ^ o ^ e •X3 ts -S" -o -a co —^ - "5 § c c cj c lissal day o loym mane G CO JC us emp CD ed di 364t us emp a. o pect c o ti , . pect o 3 o o pect c a ja CO 03 ive CU ive c >• C c ive o cd o 3 oo o T3 o .52 CU E -g t-c U '3 'o c <- .B 6 0 ° J= •S B * _ |1 s i 8 e.«=s a. « o S 3 S | c o. .2 E — -9 3 = "S 1 co <c. w" • .00 o 00 00 ~ •ft 3 s s 1-s 00 00 cu r-I 1 a ca 00 j 5 ~ oo 3.1 3 O 00 •C3 S a CJ a o g if JC c 2 E •9 -a :a E :a E o c cj O u, • 3 O "s :s o cx, u 3 U 5 o o -'3 B c c 03 O oo o ^ c o « ^ f" S i o -S "S • 3 oo « -S S3 c a 3 co c "** C3 "3 ~ cu 0 o" CU « R. cu •5 o a s i- "a 8 8 e ^ I -S S • 3 O -O •« . Ju e C3 „ cu S cu cu K C3 2 w IS •5 "3 o jo •a c o o CJ 2 £ S 3^ c B p 03 •Z B 60 Z. s °. 8.-a £ S t: .2 5" s -•c o o D C CJ C 3 O e s •2 E CO g N O S 3 o < . c -B £ a g " « -2 'S Jci » -3 O a N 3 S 3 X c — -a x c -a 'C o s a » : s s « 5 8 2 cj o -a 3 . 3 3 S =3 "3 o •X3 •5 "S °o Cj a c C3 cf « 3 ; S S cu *~ ^ Ju ^3 cS,T3 <4- .H s = g M _ c o c 3 *^ cj 03 <D "C fi •§ ° ca <u 3 ' 3 D X C >^  <U U O <" 6o s a, ccj 3 „ O D C -3 g O 03 O •n N ' 3 03 E O cj cj cj I H CJ 3 ~3 —, 3 03 O 3 ' 3 O 03 • 3 3 N E G -n a o 6 0 . 2 o Q "o" « £ , 00 ^3 IS a a o c s cu 3^ 5) * •£ s o cj ^ cu cu B 3 v. cu o IS O 3^ 2J ^12 - a cu cu <4J to Ci, e 1 G cu &• ~ s 3 "S ^ n «i O •~ "5 <J ^ O 3 c cu CJ - 5 ^ 3 l o ^ ^ 3 C3 \5 O w c ^: 2 es ** 3 o, s; -c .ts « cn ?s cu - a « ^ * -C O cu £ -4 "S .2 —< i-. O •St O 3 . O CO C _o 1 3 1 E o o S3 e a. M g E S -3 "S ^ S " 2 o S » .2 2 5 ? > co cj :> 3 S * a g oo 'a 78 than exhaustive of the issues described within each category. In Table 3, each issue is also illustrated by quotations from participants' descriptions of the essence of the injustice. For instance, distributive injustice concerned the perceived mwallocation of resources used as rewards or punishments and was illustrated by issues of pay inequity, non-promotion and unjust dismissals. In an example of pay inequity, Chien-chu was upset because he was frequently called upon to work extra hours without additional pay to 'cover for' the project leader who liked to sit in his office day after day without actually working. Moreover, the project leader's bills for hours 'worked' were routinely and summarily approved by the project leader's superior, although they were widely known to be fallacious. As the quotation in Table 3 indicates, the project leader's behaviour ultimately brought the project to a premature close because "the money ran out before we could get anything done." Chien-chu was then laid off. Each of the four categories of organizational injustice is seen as conceptually distinct, although not necessarily independent or unrelated. In this research, I focused more on describing and explaining categories of organizational injustice rather than on their interrelationships or the particular process(es) by which individuals decided if a situation was unfair. The Question of Intent Most instances of injustice were believed by participants to be premeditated, purposive actions or decisions (direct or indirect) seen as personally and/or professionally harmful. Whether involving managerial styles or organizational policies, participants believed that both the individual(s) acting unfairly and, to a lesser degree, the organization were well aware of the ongoing injustice and its impact. In one case, Joanne recounted: "I 79 know of three (ex-employees), still going through psychological help to get over the abuse (by the boss), who went into the office and spoke to (Director) and (Executive Director) and explained why they were leaving, what was happening...and asked that she (the boss) be replaced." At the time of the interview for this research, the boss remained employed at the organization in question and, according to Joanne, continued to adversely affect those whom she supervised. However, even when seen as unintentional, the impact of the injustice was no less devastating or traumatic. As Joe said of his boss, "He didn't want to hurt people but he was just so damn good at it. He does want the best for people - he just doesn't know how to achieve it." For the purposes of this research (consistent with the spirit of Canadian jurisprudence), injustice did need not be intentional to be considered valid. Regardless of whether individual behaviours or employment policies or practices (or their implementation) seemed neutral or were meant to be just, determinations of injustice were by their claim as such by participants. Thus, in this study participants' experiences of injustice were not required to satisfy legal standards to confirm their occurrence. However, participants frequently relied on and described the effects of injustice to make and/or substantiate their own determinations of whether injustice had occurred. Because participants' experiences of injustice were not assessed by legal standards, I have avoided using extant legal terminology to define and label emergent concepts. Instead, where possible I have relied on relevant concepts associated with existing organizational theory. 80 Interactional Injustice Def in i t i on In this research, the concept o f interactional injustice emerged f rom induct ive analyses o f participants' experiences o f unfairness at work . Ac ros s a l l cases ( interview and archival ) , a pattern o f mistreatment f rom interactions w i t h superiors formed a dist inct and significant source o f injustice perceptions. Spec i f ica l ly , I define perceived interact ional injustice as mistreatment result ing f rom interactions w i t h one or two authority f igures 2 1 w i t h w h o m a reporting relationship exists (see Table 3). Author i ty figures inc luded immedia te and senior bosses, both 'new hires' and 'o ld hands,' as w e l l as a contract consultant brought into the organizat ion for a specific project. In this research, I refer to the authority figure i n quest ion as 'the boss.' M a n y different instances o f what participants perceived as unjust treatment by bosses were described: ye l l i ng at employees unt i l they cr ied; boasting o f the abi l i ty to make employees quit; repeatedly slapping (privately and pub l ic ly ) a participant's buttocks; telephoning employees i n the middle o f the night to verify gossip about other employees; cont inual ly c r i t i c i z ing but rarely prais ing participants; " joking" w i t h participants us ing sexual and/or ethnic stereotypes. T y p i c a l l y , many such instances occurred over several months and were significant sources o f frustration, fear, depression and stress for participants and other employees. Taken together, I label i n Table 3 the under ly ing issue as 'managerial 2 1 In the majority of cases, interactional injustice involved a single authority figure. Typically, authority in the reporting relationship was explicitly indicated by hierarchical position — most interactions occurred between hierarchical superiors and subordinates. In one case, however, authority was implicit; although Faith viewed her offending colleague as occupying a lateral position, in practice he was responsible for her training, directed her duties, supervised her performance and was seen to have more knowledge and technical expertise than she. 81 maltreatment,' reflected by personal discrimination and/or person-to-person harassment (e.g., racial, religious, physical, sexual in nature or general bullying, coercive behaviour). This inductively-derived category of injustice is closely related to the extant concept of interactional justice, which refers to the fairness of interpersonal treatment received during the implementation of organizational procedures (Bies & Moag, 1986). However, unlike previous research, I focus on the unfairness of interpersonal treatment and suggest that interactional injustice constitutes, in itself, a category of workplace injustice as distinct from procedural-based injustices. It was observed in Chapter 1 that justice theorists typically subsume perceptions of interpersonal treatment under the broader rubric of procedural concerns. However, data from this study clearly indicated that when not implementing procedures, the behaviour of authority figures represented a unique and substantive component of injustice perceptions. Moreover, its conceptualization as a 'stand-alone' category of injustice was supported, in part, by its incidence in both absolute terms and relative to other categories of injustice (see Appendix H). As indicated, participants'judgments of injustice were made on the basis of several interactions with bosses whose maltreatment of participants (and others) appeared constant and enduring. Moreover, some bosses appeared to have an established history of mistreatment outside the organization. For example, Ashley's perceptions of her boss as domineering and intimidating were validated by a co-worker who went to high school with the boss's daughter and who "still remembered him (the boss) as a tyrant - everybody was scared to date his daughter." Similarly, Suzanne's boss (who had sexually harassed her) told her he had 'gotten into trouble' for touching his boss's breasts at the organization he was previously with. 82 Several bosses were believed to have learned how to mistreat others in the military. Joe's boss "who could (and would) yell longer and louder and harder than anyone he knew," traced the beginnings of his boss's intimidating style to military school: "he was a drill sergeant...then he learned engineering in the military. When he went out to manage construction crews in the bush, he yelled at them until they got things done." It was not clear why the boss did not recognize that this style was neither necessary nor effective in the real estate industry. In sum, participants used data from several sources in making judgments about their treatment by bosses. The frequency and consistency of mistreatment by bosses in other settings gave participants confidence that their own experience of interactional injustice was authentic. Interactional vs. systemic injustice. In small organizations or in larger, owner-operator businesses described in this study, the influence of authority figures throughout the organization was often significant. The pervasiveness of their influence sometimes blurred boundaries between interactional injustice and other categories, namely systemic injustice, defined as perceived unfairness from the larger organizational context within which interactions occur and allocation decisions are made and implemented. Arguably, in the smallest of organizations a single authority figure might be seen by others to represent the entire organization. In such cases, I coded interactional injustice as the singular injustice on the basis of participants' claims that the underlying, fundamental injustice was the interpersonal manner by which bosses implemented policies, practices or procedures. Interpersonal mistreatment in these situations was easily evident to participants and was readily traced to one or two authority figures. In contrast, cases of systemic injustice 83 typically involved widespread interpersonal mistreatment by authority figures throughout the organizations described. Participants tended to see such mistreatment as much less distressing than problems from the larger organizational context within which interactions, outcomes and procedures occurred. For example, B i l l was pleased to hear that the organization considered him 'associate material' and he was excited by the promise of associate-level benefits. He became hopeful and 'volunteered' even more unpaid overtime. However, the organization postponed his repeated requests for the performance review required to assess his suitability for promotion, and later repeatedly changed requirements to complete the review. As he said, "they danced around until....one vice-president said (to Bill) 'I have better things to do with my time.'" Despite his frustrating dealings with a controlling, intimidating director and a fearful, ineffectual vice-president, B i l l was distressed more by the organization's inconsistency than by his bosses' interpersonal manner. On the basis of this and similar evidence, this was coded as a singular case of systemic injustice, notwithstanding Bill 's frustration with the two authority figures who initiated or perpetuated the inconsistencies. More generally, other evidence included participants' language as one indicator of the appropriate classification of cases. Accordingly, repeated use of "they" rather than "he/she," for example, suggested systemic rather than interactional injustice. (Appendix I outlines in more detail the distinctions between categories and guidelines for classification.) Who Does It: Perpetrators of Interactional Injustice Both genders were seen to mistreat subordinates: 11 of 16 cases of interactional injustice involved male bosses while five involved female bosses. Eight cases concerned 84 same-gender interactions while eight cases reflected cross-gender interactions. Specifically, there were three cases of male boss/male participant; five cases of female boss/female participant; and eight cases of male boss/female participant. There were no cases of female boss/male participant. Of the 30 archival cases, nine were coded as interactional injustice; eight of these involved males bosses and female complainants. Only one case involved a male boss and male complainant. Dimensions of Interactional Injustice From the interview cases (and, to a lesser degree, the archival cases), I identified eight behavioural dimensions of interactional injustice. Together, these dimensions reflected the substantive behaviours by bosses which participants saw as unjust. The dimensions, defined and described in Table 4, are as follows: intimidation, abandonment, inconsistency, degradation, criticism, inaccessibility, surveillance and manipulation. Using a structure adapted from Kahn (1993), I list in Table 4 the key acts of each dimension and their immediate reactions, and illustrate dimensions with interview quotations. The degradation dimension, for example, is defined as verbal and non-verbal communication perceived by participants as hurtful or disrespectful, and included key acts such as publicly shaming employees, gossiping about or disparaging other employees, or personally and/or professionally attacking employees. Commonly-reported reactions to degradation included reduced self-confidence and increased frustration. Permutations and combinations. Several dimensions of interactional injustice were frequently interwoven throughout participants' descriptions of their interactions with bosses. Intimidation, for example, was often accompanied with or followed by criticism and/or degradation (these labels are largely self-explanatory; for additional detail see Table 4). 85 o oo cj c tu c .3 o .14 " CU CU T3 CU O > ca is cu c ca t. cu B CU O 3 T3 CU C O cj CJ CJ 3 TD C "O C ca Lu ca .«J ^ ft U cu co O CO o .S3 ™ r- CU *3 CJ • g CU 73 co ca c o o a ca "ca - O CJ cu CJ CO >> CJ O co •s. * s >> cu ti -^s <U ^ & c o O ~ ^ - ^ C O C u •a £ o. a CO 6 CJ cu C o *3 cu ^ 53 00 *H <u cu S => <-> * J fe, I % -5 •a ^ -fr •a 11 cu a S3 a a cu © 5 S, ^ a x •s 8- s <Tj cu oo to Q cu cu -s; s~ >~ *-CU ^ Oo su ^ to 3 a CJ s: cu Oo °0 3 JS . cu § 1 1-5 CU n 3 « § 00 "a •S k -cu 13 IS a c. 00 3 . 5 ^ 3 C t i cu C5 00 £ a c I?3 K : & 3 3 O C tS cu 3 00 •5 s 00 cu cu tS s: cs -s "3 o a -g o to fe cu S •a o 00 cu "a e a 15 3 o 00 •S a ob " . •S c ^ cu > a fe to a 3 o K e a ^ -e ^ 00 'S 3 to ^ as 1 * St cu to 3 CU 00 ca 3 00 c ca 3 o cu - C c a s a S .op -S a to -a a cu I4 00 3 O cu . S cu 13 3 S i S? a -§ 3 fr "§ a cu > a "§ Jo a; "a K to ts c « cu CJ cu •S ^ 3 00 , •S 1 s • B cu a is 31 CJ o ^ 1 B g OOcS t j C . CJ cu ca Q o * ca o CJ CL, o xi t l t i c 8 "§ &> ca c ca <u O) to a c C3 CU iu 00 > ^5 i i O c _ CJ CJ -o CJ rt o o e 00 o o Xi Cu CJ CU o "B. 6 CJ o -a c o O H FU CO a a C O <U M CU Xi 'S Z e £ e o •a cu >> o ca c "ca 00 c CU l a CJ u C/l ^ r3 cj ,ca cu t+H — . ^ 00 c CJ .2 c ett g o o •a If .2 ^ ca ca ^ V, O <u "+5 J J 'fc C " c« CO OH OH 3 o •S -5 <3 : " cv: ^ - H C U fe g ex. -= K 3 O w .a .<<, ts .is t 3 t? § £ 8 « 3 Cj Cu" 0 "is .y Oo cu > S u ts a J v. s •§ 3 o S £ -a a s -s a '5 -S £. O , - ~ cu cu £ - = -S K ^ -c s § > s c ^ ts 00 • cu CJ t^ §< g . B I S3 .2 & a OH 4 ^ 6 % u „ i i fc § I" § , S f a f . . S "co J2 S i> cu -s: ts e cs ts cu CS a cu 00 cu - C g •S to K •2 to cu 3 C^ £? a 'cu . 00 i f =4 ^ . § I o o ts cu 8 « c « s cu 3 ~ - J i *H. H >^ CU 00 s 3 § 3: <5J CU tS cfr cu 00 • o s cs -S3 cu Oo s .5 r1 K cu 3 5 S a ^ ^, > 7-5H S S W O H <U CJ cu tS 00 C a Q ^ to £> 00 00 « to W "cu • « 11 co o s ts 2! .5" Cj • - . 3 3 5 3 ts tS C CJ ^ > 5" K cu P > CJ cu •a -e 00 fe s i 5 IP S « a. -a CU CS 3 ts c gei am 3 red. s O Cj 03 a t o CJ J S -a ;xc 3 •id £• • o "60 FT1 •t3 > of an f-dou ideri) very • M 13 BO "cs" 3 CO •5 IS lopm co" IS CS lopm tres '/ting, kno Cj co tu C/l '/ting, ev hip. c 'C/r no real -o hip. o real CO 3 c3 M 3 e _o C/D =4 rea *.3 cct Cj rea C/3 rel; wrong n ~S :es tru gerial es fear /he rim wrong n d to fe edui lana real "All t 00 •S a arte s u "All t • co cu S -a 3 § •a -3 ^  co 3 j? ^ 00 § a -a . I * < e . oo •o a CU o c 03 CU OH 3 -a 73 3 ca u tt g> 6 "Si 3 3 M CU ,cu M C M tu C M «C 73 tu . 2 .22 o . E a o u r! « et S & S tU CU tD >. _ o " o . E tD ~ c o .2 3 CU C M r. CO 3 u .2 %° tu to 3 O • O CO P Jg 8 3 « 3 U i3 ea ° o M ._, o t5 co <U C3 r-^  73 to "3 tU « 00 . 2 5 <= •S 6 >. 6 rs o !S ° I & < OH u B cu > , <U TD 3 ^ . ~ cd M tU 3 •3 1 / 1 — ca t—i CU 5 a. <*5 U o3 tt ^cS J3 o 4-1 3 O O C \ B co _03 8 ° 73 > to cj CJ -a s cu -a s CJ cO tu -g a t^  tu CJ oo "13 s CJ 4) c j -5 C J c s Cj -a cu a -a •§ s 3 Cj -a 00 I > 5 O cy to e c j .5 Cj = o1 c -a 3 CJ tU 03 t l 03 tU u c O e 3^ ^ CJ 3 00 c a I f § | Co CJ 00 .e CJ •cs e tu ^ -a c j -a a . Oo 5 CJ tu S a, ^ o c * • cu s ^ cu cu C - cu t« V ^ O 7 ~ ! - £ >^ -a Cs -a a co a a, S 3 CS cj cs -a +H> tu CS _s; 3 J3 cU trt O M C O tU 03 0 CH cu J3 OH T3 Co ^ 2 CO" ctj O H 3 to la « -e .2 tu —3 1 § « t 73 O 03 tu >< 6 -3 O3 o3 CO CJ B •§ s a 5 % 3 U 6 < a ft" U \4 o •43 73 o3 "c3 3 o 'co C/3 CD to *H U O tu OH 0 O OH 73 6 s o a M tu H M - M ca o 3 on O tu co op OH S • - OH 53 .a cr _ .S ^ = 1 o ^J CO ca a, M 3 CA tu o 6 0 0 73 u g ° 5 3 OH ^ s - (U OJ 6 0 CU Q a s tu i-T to 3 cs tu cu w i •s ^  ~ a 2^ CO -a 3 a, to Ci Si a - s ; cj 3 to 73 CS - s ; •if s oo 3 • ~ § S 3 « J 3 0 •~ o tu * C ^ 3 3 S 00 C" 3 a § cu •S -a « 5 cu O 00 .a C J CS tu S 1? =4 a 00 cu CS - C s ** 1 1 2^ Ci < •S -a 73 -SI "cu - C .2? Ci - s ; i ^ ii a 3 a 3 I I co ^ , cu C 15 S s 3 a -a CJ cu Ci . cj S 0 0 ^ 73 "S tu a -a C J s a Cj a 3 a 73 3 Cj 00 a . in. tu a I tu 3 oo a o3 O f*5 tU CJ 3 cU 73 IS 3 O o cu CU >. OH 6 tU CJ 3 73 tU O . I H o ss -*—' 3 O ca s o 3 o ca tu tu o -U J3 CU s ca 3 " f 3 • 3 ca . g 6 • t « ° " OH co cu o3 > <U 73 - O tU r v O 3 >2 ca cj 6 « c2 u Ti 3 a 3 "co ra tU i J> on >. ° 8 CO ca "ca ' > 'C ca .5 tt > E tu J3 .5 6 +- jo >^  2 L3 O -3 j-j O . '55 u o 'S tt 2 c/l ca <u o M I—i — o . ca _ s .5 ° a M 60 U ca O , 60 3 O "3 >> ca « 3 ,ca oo CM c q 73 3 3 > tU o 11 r v- C M O W5 CU n=l o •<J JO >^  "BH <U 3 87 Cj a 73 °0 -13 3 3 S 8 -£ 3' * 2 Cj cj <U Oo ^ 3 to 73 cu a - f t O o .°o -c ^ a° to -5 1 « tu It 3 co & s 5 u cu" ^J1 a cs L. "5 c P to ^ 3 o O -a r-1 0 co ?*> -tjj § V. p JJ CS c a cv. .2? 1 a a ~Z cj .3 -a 00 cu > CJ *" Cj ' ~ C" CS -a CJ Co -a 03 tu CJ ob ^fe • S ^ -a tu -5 5 -5 a ^ 00 £ 5 fc a . ^ « a ft £• a cu -a -cs a cs a cu CJ cj a \ cs tu 00 3 . 2 "g 3 2 I * =U Co It > ~ « « J) Cj * 4 -a ^ CS .£ -a «> * S £ fe to CS 3 3 Cj CJ a cs cs i i cs -a 3 I * -2 3 cs •• 2. = -a •S s ^ 2. a £ co "43 CJ 3 *» •as I-fe s <o -~ 3 a -a cs -a C J 3 £ a co cu 3 § 21 00 cj ^ -a Cj tu 5 cu C J a a cj co cs =4 cu -a cs -a "cs tu a cs £? cs a. tU « -a cu co J * ^ S "§ -3 Jg £? C cu a 0 ..? 3 -S ^ a^  > *-•U CJ - . 3 3 cs Si r ^ s^ a a -a 1 = <•=, a t , cs cs -a -a *. a -I a a 8 <-> " J Ci CU « -a ^ 1 ^ •a cu 00 -a •5 § •5 i: 3 .a a- 43 3 cy to Oo o . 3 CO N 3 CU 6 o. _o > tu • T3 cfl VS « .ts . - xi g o . . . .5 -a c 00 . ~ IS C *-i t/3 O 13 is 3 S 3 s -a .y 3 3 6 6 o o 1 43 03 tU •a ca o <u tU 00 fa 03 cu a 3 C tU 00 3 03 J3 CJ X tU =4 O ol DM 3 -o tU CJ 3 3 -6 tU CU • Cu >> 03 .-3 3 M § j § •3 03 B t+H O ° * U o cn tU cn C H c ^ S 2 .a U -B a tu -ft _tu -* 'fc » a o -cs -a a ~ cs 00 ft cs ft a •S & cs s; c a, _ 00 .2 -ft tu -ft CU SP 3 O u o o >. o 03 o co *3! 33 co •O b tD . . S tu a « o 03 o r^, i l l o 3 tu 3 a cj tu a; tiU s s o J3 J O ' 3 > 03 3 _o o a H c 03 13 o 'cO J3 rv 00 CU co to a o _ 3 > OH 3 T3 u jj g <U 3 O " 3 4J O O — C O o3 g> § g h= iu u cfl 0) T3 tU O O T3 03 •a c u •r; tu „ * a I « H 3 U 0 3 a u tU c/5 q ^ =5 3 03 a iu 2 = s ^ 'S « 5^ o <U tu = tt 3 CU U U CS C ~ a tu cu S CU «J cu a 'S •SH CJ -a t - t I § oo - . co , a 3 -a a ^ 3^ a "5 a, S . b 5 « -5 -I a a fe •S a s; <H CU a> a , tU Oo § 1 - 5 s -a • a a S »o « » . s 5 y -w 3 £ § * a,<e, cs 2 u -3 ^ 3 ej .-3 ^ a to cj 3 -ft fc "5 a t; cs 3 : '—v to' cu 5; oo a 3 cu a -a oT-S-- I 1 | a 3 a cu S K <U a CJ . CS •cs ^ a Cj 1 ^ cs a a cu a, 3 a I 3 a a a -a a,.* .1 § •« 'S •S a1 Oo Co Cj §• fc *H »> a c5 -it. i2 a cs > —. -k cj a a. co 1-? . a co j a CJ J- "CS a S .Si a h k a a "o to 00 fc a § -CJ =4 *^ 3 s a a -a -a cs cs o 03 CJ F i CfH tU T3 3 03 IO cu to CU >. O ~Cu s <u T3 cu I IO — o* -3 a 3 o CJ cU OH X CU <U 3 , -co 00 i2 " 8 3 cU O , N 3 tU _ cn cu cn tU cU - cS O H 1 * -CO g 2 J? P5 03 O o3 CU 00 3 o3 00 03 3 U <u •73 a - ^ o « S £ a . ' S § 1 ^ CJ cs -a 8 a. CJ "u 3 ?s -a -a ft S 3 to 73 CJ C j ?S to CU f -O ' -^ S3 O 00 cfl 3 to OS a 3 a J3 a iH 10 T3 • tH o tu cn 3 -3 O cU T 3 '> O 3 3 O OH OH _n CO — -cU 03 OH -3 ,0 • CJ 4V 2 CU >^  'S ° 00 Cu -33 a B U tu o o o o _c 0 3 c 03 co B cr .3 CU 3 , h 00 ^ VS <° ^3 J 3 -a 3 03 "= £ ~o tU O 3 tU T3 3 O _ ^ J3 tU 3! M E3 o cs U o cu u e '8 > u 13 ^ .2 tu p tt -g. co tu cs -a cs -a 3 Ka U a S at ts S; a * 3 « cu Cj * a 3 00 : a o Cj 00 : 3 tu o .^g co cj "u § -a -a S cj -a 1 _co a >^ s^ -cs a cs • s i •SI 8 00 l l *- cs Cj tu -ft fc =4 -a ^ cl « "u tu "Cj i t a a, cs tu a -ft cj — tu a k • fc C J S § a , 3 cu "Cj k "2 5* § 4 cu I ^ ^ -£ a « S cj-3 1 to 8 * •3 Co Ci 3 a tu a *s -a CJ a, 3 5r fc IS to *» •s s to a 881 e a cu CU s S to i-3 © cs JS CD CQ c S CO E 3 O a & a •S a, a SS -a 3 -SS S a s -ss a -a <u -js a 00 cu B cu cu -a e a cu -I g> | B B 13 <u B f cu a -ss *~ - 13 - s a oo cu a, 5 00 ss c-i cu a, cu SS a -SS a co a PS ^ •2.-4* 3 "5: - a ^ -a -s <o IS: a B S oo - CS S ~° Co cu -ss Co u is cu .S a a a a, co a ^ -SS cu £ .5 00 SS 1 § SS -SS a cu SS B . j 8 ^ "5 -a a, ^ cu S "cu to cu co ' -SS i 5 -a a cj cu SS a -SS a, B B1 <o oo TS ss S -SS cu .00 Oo '5 o ^ cu 13 13 C >> cs OH •-> I '§> = i cd C CJ 3 & 00 -JS >r> -5 CJ O o £ I 3 ° a co _ CO fi ° <u «—I £ CO 8 oo-c 3 C O 2 £ % B 3 P i ca cd 00 c o CO OH . '•S o ^ ,tj C 'co O 3 -2 § CD o .2 5? o CJ " 00 S3 S c o CJ o « cj o cu "§ a cu c ss cu a > CU « S •« a 3 ^ a ss S i s a a *. "c a a -ss 3 * a " ^ ** 3 c a -2 a ts i s § I ?! a. *-SS 5? I I § a •5 *• *- CJ 13 <3 SS CS v —N a a ts a ^ cd e o GO UH CJ OH TS CJ CJ - O c o 13 C CS co CJ Cu O JS CJ _3 "cd > cu n u § 3 cu a , . ? • s-a I cu a, a ts cu cu fe "S CJ -> JS o 2 £ cd cs TS c cd c« 6 cd co u li o u tt 00 "S .5 - B CJ TJ S >> o s —H cd ° o >-T3 o o, e 3 cu -SS -SS ts SS B 3 a =s cu B B Oo CS cu 3 a Cj a a oo B > o cj CO g Tu ° B e T3 § £ CO .3 CJ CJ 00 i-c =2 o e o ••e _« .& "3 CS co 13 co cd 13 3 C r r - CJ cj CJ • • C >> S .2 « J3 TS cu -ss cu -a si a B 13 <u cu s ^ E s ^ S ss B B ^ 00 •S s -3 -a s a -ss COl a ss Nj -ss SS a ^ i ss a -00 cu SS cj ' s^-S 3 a, °" B " - -SS cu *•» S .S -1? CO cu a c3 co SS •2 a l i 3 a ss a . 1' a, a • a -ss § B a a. cu -SS ,co -ss B -SS to tS SS B <3 3 a ?s ts cu « cu -H --^ H-» 33 >X .S CJ c-i CJ .a, s ts s a B to a *; cu oo -a § s ^  a ts Si ^ is "S ji B a 89 Ashley's boss, for instance, simultaneously used intimidation and criticism at planning meetings which meant "...he would question everything and turn it around, this way and that way. It was like an interrogation for criminals, trying to get us to confess things." As a result, "we'd get exhausted. We didn't even want to answer anymore — it just wasn't worth it." Similarly, Gwen described her boss as a perfectionist who combined criticism with inconsistency as she arbitrarily targeted staff against whom she 'inflicted her standards.' Thus, these eight dimensions of interactional injustice are both conceptually distinct and empirically correlated. Surprisingly, a paradoxical combination of surveillance and abandonment was evident in several cases. Surveillance is defined as careful monitoring and exceptionally close supervision of employees and was exemplified by bosses whose physical proximity to and frequent contact with employees were perceived as both unnecessary and unusual. Abandonment is defined as the failure of bosses to inquire about or respond to employees' physical, cognitive, social and/or emotional work needs. Thus, in this unexpected combination participants reported feeling at once scrutinized and ignored by their bosses. For example, in one case the president of a large labour organization was reported to routinely engage in surveillance of his employees. On one occasion, he hired a consultant to help head-office employees develop better relationships with bargaining units and members. Employee-only focus groups were held off-site but, unbeknownst to the employees, the president had set up a one-way mirror and microphone to surreptitiously watch and listen. At the end of the meeting, employees were informed of the president's surveillance and not surprisingly, they were upset. Catriona, the office manager, described the episode as "an outrageous, sleazy, unethical violation." She added, "had they known he was behind a mirror 90 listening to them, they most probably wouldn't have participated." On a previous occasion, he had also directed Catriona to install a software program to monitor employees' telephone calls (see Table 4 under the surveillance dimension). At the same time that he carefully watched employees, the president also tended to abandon them by neglecting their needs and failing to provide support. In one graphic example, he yelled at Catriona in front of her subordinates to "get the fucking photocopier fixed or get it removed." One week later, she still felt upset about the incident so she approached the president to discuss the matter. However, rather than show concern for her feelings and acknowledge his behaviour and its impact on her, the president denied yelling at her (although he did agree he had yelled), saying with barely controlled rage "are you calling me a liar - if I say I wasn't talking to you, I wasn't talking to you!" (see Table 4, intimidation dimension). Distributive Injustice Definition Consistent with previous organizational justice theory, distributive injustice is defined here as the perceived misallocation of resources as rewards or punishments (e.g., pay, promotions or dismissals) by organizational members and involved cases where organizational outcomes were seen by participants or complainants as unfair. As noted earlier, issues of pay inequity, non-promotion and unjust dismissals illustrated distributive injustice. 91 Procedural Injustice Definition Consistent with previous research, procedural injustice is defined in this study as perceptions of unfairness which involve the procedures used to determine the allocation of resources (e.g., performance reviews upon which salary or promotions are based). Across cases, procedural and distributive injustice typically occurred concurrently. In a notable exception listed in Table 3, Kathleen had joined the organization in question following a national job search; she then gave up the pursuit of a doctorate and relocated to another province. However, as part of an unannounced organizational downsizing, she was fired on the 364th day of full-time employment, one day short of reaching permanent status. Although she did not disagree with the decision to downsize, she strongly objected to the manner by which it was conducted in which, with no prior warning, she and 20 other employees were given 30 minutes to gather their personal belongings and vacate the building. She said, for an organization within the education service industry, "the half-hour process of getting rid of people was brutal and unprecedented." On this basis, the case was not coded as distributive injustice (no objection to downsizing) but was coded as a case of procedural injustice. Selected Issue: Performance Reviews and Feedback In addition to unfair dismissal procedures, the second of two central issues cited by participants in cases of procedural injustice was that of unfair performance reviews. Both formal and informal appraisal methods and the communication of results were frequently faulted for being neither descriptive nor prescriptive. Patrick, who was a highly committed, career^minded regional field executive said of his formal performance review: 92 It was a general paragraph - there were no specific examples as to where I had fallen short. He denied me any opportunity to discuss the results...just 'take it or leave it, sign it or don't sign it, get back on track within 6 months and we'll see what happens.' Following Patrick's complaints regarding its accuracy, it was later agreed that the negative appraisal was to be retracted. However, following a disappointing loss in a promotion competition, Patrick was surprised to discover that the then 18 month-old appraisal had been released to the selection committee without his knowledge or consent. After several unsuccessful promotion applications, he was told that he would 'continue to be pushed aside' until the negative performance review was resolved, despite prior agreement to retract it. Several month later, he was laid off, with no opportunity to resolve the review. In another case, Eva, a clerk in the health care industry, had been given a new job at a new institution on the basis of her previous positive work history. However, the first suggestion that her performance was problematic occurred during her 3-month performance appraisal, the first sentence of which read "Eva is a liar." In addition to what she perceived as defamatory language, she was affronted that her assessment was completed by her 'absentee boss' whose frequent off-site meetings and conferences meant that she (the boss) had limited knowledge of Eva's performance. Regarding informal performance reviews and feedback, vague complaints about performance which bosses either would not or could not substantiate were common. In one example, Gwen noted: She (the boss) told me my co-workers were complaining about me, so I said okay, what are the complaints exactly? She said, 'well I can't tell you that but there are a few things and I'm concerned about the morale of the office.' I said that I can't fix what I don't know is the problem. I said who is (complaining) and I'll go to them directly. She said, 'I can't tell you that either because it's confidential' which I respect. So I said I was going to (and did) go to each and every one of my co-workers myself and ask them if they have a problem with 93 anything. There was only one who said something about my handwriting on my intake sheet which was fair enough but everybody else said no. In several cases, there was widespread confusion about criteria on which participants were being evaluated, itself an apparent consequence of managerial inconsistency and criticism: "They wanted me in the office but then they kept telling me to be out of the office. And when I was out of the office they were complaining that I wasn't in the office." Rahim noted that, following 3 years of positive performance reviews from his boss, he unexpectedly began to receive unfavorable feedback from the same boss. When he sought clarification, he was told that he was not a good fit with the organization. Several years later (and no longer with the organization), he was still puzzled by this assessment and the time required by the organization to discover his unsuitability. Summary. Most of the criteria for effective performance evaluation and feedback (e.g., M i l l , 1976) appeared to be violated in participants' formal and informal appraisals: vague feedback which was personal, not tied to organizational or professional goals, i l l -timed, and not checked for understanding were frequently pointed to as sources of injustice. As well, positive feedback was rare and participants were widely confused about how to improve their performance. As a result, they often felt less motivated, more fearful and/or confused and undermined by bosses. As Gwen's quotation above illustrated, substantial energy was expended (if not wasted) by participants in their efforts to clarify, confirm or meaningfully integrate performance feedback. Systemic Injustice Definition. Consistent with Sheppard et al. (1992), I define systemic injustice as perceptions of unfairness which concern the larger organizational context within which interactions with authority figures occur (i.e., interactional) and allocation decisions are made 94 (i.e., distributive) and/or implemented (i.e., procedural). I supplement this definition by introducing four illustrative issues hitherto neither identified nor described in previous research: organizational inconsistency, organizational abandonment, organizational overwork and organizational discrimination. In particular, the first two issues are novel and accordingly are explained more fully. Organizational inconsistency refers to inconsistencies in the application of directives or policies or in the treatment of members across the organization. Examples included one case of an organization which failed to honour its policy of providing support for employees' educational upgrading (see Table 3), and another case of an organization whose implementation of discipline, according to Lisa, relied on favoritism. Also, significant discrepancies between what participants were told about their position in the hiring process and the actual daily duties of and limits to their work were considered illustrative of organizational inconsistency. For example, in one of two cases reported by Catriona, she accepted a job largely from its description as one of substantial autonomy and a team-oriented approach. However, once on the job, she was closely supervised in a work unit with little participation, communication or commitment among employees to the project: she angrily reported "it sure didn't feel like a team." Another emergent issue concerned participants' profound feelings of abandonment by their organizations. In several cases, participants were given significant operating This is different from direct discrimination (defined here as a form of interactional injustice) and systemic discrimination (or in American legal terminology, adverse impact). Whereas direct discrimination as used here refers to harassment or objectionable behaviour by one or two authority figures, organizational discrimination refers to discrimination or harassment by many organizational members and/or by organizationally-sanctioned policies or practices (e.g., a policy constituting discrimination against gay employees). Systemic discrimination, on the other hand, refers to workplace discrimination from employment practices which work to the disadvantage of the following four designated group members: women, visible minorities, Native people and people with disabilities (Stone & Meltz, 1993). 95 responsibilities in implicit or sometimes explicit efforts by the organization to empower employees. In the absence of attendant authority,24 however, and in the presence of authoritarian managerial styles, empowerment was experienced as organizational abandonment. Karl explained: My direct superior was the boss whenever it was an issue that normally wasn't sensitive. But when it came to a sensitive issue or a logistic difficulty, all of a sudden, boom, 'it's your baby, deal with it.' And so I felt that the rug was being pulled out from underneath me. If you circumvent him and go to the owner to say here are these problems, the owner would say 'work twice as hard, it'll work out, keep at it.' In the beginning you figure he must know what he's talking about. So you shrug your shoulders and put your nose to the grindstone. But the owner's prompting to work harder did little to directly assist Karl to complete the difficult construction project he was in charge of. Its subsequent failure brought with it a significant financial loss: "for the money that was wasted, probably everyone in the company could have been sent to Harvard for a Ph.D." Similarly, Denise who was mistreated by the broker she reported to, also experienced organizational abandonment: "the prevailing attitude in our firm is that brokers are always right. Even if brokers are wrong, they're still right - always. I don't feel I get any support from them (higher management) with the brokers." Jennifer echoed, "I would write e-mail messages trying to get answers to questions and I would get no response whatsoever. In terms of verbal questions, they just wouldn't really say anything except this is just the way it is." Such efforts to seek information, guidance or support were typically unsuccessful, with participants left feeling unsafe, isolated and resentful. Responsibility is understood here as an explicit obligation to perform with implicit accountability for performance (Weber, 1947). 2 4 Authority as used here refers to rights in a given position (as prescribed by organizational hierarchy and reporting relationships) to give orders and expect them to be obeyed (ibid.) 96 Decision Guidelines for Classification The following section articulates the key guidelines used to classify cases of injustice according to the four-fold typology. It is noted that, while distinct, these categorizations are not completely discrete. Moreover, the process of assigning categories to cases was inherently subjective. Nevertheless, classification of cases according to the typology presented in Table 3 was systematic and intended to be confirmable. To this end, inter-rater reliability of category classification was computed for 30% of the interview cases (10 of 33). Cases were selected purposively for proportional representation of the full sample according to the following criteria: the incidence of categories; gender, ethnicity, organizational size, industry and the incidence of singular and conjoint classifications of injustice. Using my ratings as the standard, Cohen's Kappa was calculated as a measure of agreement, corrected for chance, between each of two raters and myself (Cohen, 1960; Fisher & Van Belle, 1993). Inter-rater reliability was .80 with Rater 1, .79 with Rater 2 and .87 between the two raters; these values are considered to represent excellent agreement beyond chance (Fleiss, 1981). Inter-rater reliability results are reported in more detail in Appendix J. As well, Appendix I provides the Training Manual used to train the two raters, which includes a detailed account of the decision rules and guidelines by which the cases were classified. Central Distinctions Singular vs. conjoint classifications. Each of the 33 interview cases and 30 archival cases was analyzed and classified as one or more categories of perceived injustice. The majority of cases reflected a single type of organizational injustice and were thus coded as singular. For example, on the basis of an unfair performance review and in the absence of 97 other complaints, one case was coded as a singular case of procedural injustice. In the case reported by Robert, a gay social worker, his exclusive focus of discontent was his organization's initially implicit and later explicit 'heterosexual-only' employment policy. Given no other substantive source of discontent, this case was coded as a singular case of systemic injustice. In other cases, however, two or more categories of injustice within each case appeared significant; these cases were coded as conjoint to reflect the collective and compelling involvement of multiple categories of injustice. In Patrick's case described earlier, for example, the unfair performance review (coded as procedural injustice) was largely responsible for several subsequent failures to secure promotions to which he felt entitled (coded as distributive injustice). At the same time, the organization in question was restructuring amidst much controversy and confusion about its mandate and priorities. High turnover at senior levels of management, extensive reshuffling of personnel and poor organizational communication contributed to much confusion, job insecurity and widespread fear of layoffs among employees. Patrick (and other employees) felt profoundly abandoned by the organization. In sum, because the three issues (organizational abandonment, unfair performance review and non-promotion) were central and deeply distressing elements, this was coded as a conjoint case of systemic, procedural and distributive injustices. Unclassified elements of injustice. In some cases, while elements in participants' descriptions were indicative of certain categories of injustice, these elements did not appear sufficiently significant or compelling to warrant classification as a category of injustice (whether singular or conjoint). Although not explicitly classified, these elements were nonetheless retained in subsequent analyses; their retention supported the qualitative research 98 criterion of credibility (i.e., assurance that the phenomenon was accurately described by portraying its complexity), thereby permitting fuller, richer analyses using comprehensive and complex data. C. Properties of Organizational Injustice With this understanding of the specific categories of injustice, I now take a broader view of the phenomenon of workplace injustice by considering three emergent, recurrent properties of the four categories. Two were consistent with two properties suggested by Strauss and Corbin (1990): location and impact. I labeled the third property timing. In addition, in this section I summarize the use and efficacy of two recurring organizational threats associated with workplace injustice: dismissal and the reference letter. Location: Private or Public? In cases of interactional injustice, unfair interactions with bosses were frequently public, often with little or no effort by bosses to disguise or conceal their behaviour. Examples of such public mistreatment included regularly yelling at participants during employee-management meetings, calling out sexist comments down a hallway or making sexually suggestive comments in meetings. In Suzanne's case, inappropriate behaviour which began privately as sexual comments and touching by her boss escalated into repeated slapping of her buttocks in others' presence. Some behaviours which appeared private were, in fact, public. Despite being yelled at behind closed doors in an office, for example, Joe acknowledged that "everybody knew it was happening — the walls aren't that thick." More commonly, however, interactional injustice was entirely overt: as Ashley reported, "he would do it (yelling, sexual and ethnic 99 comments) in front of other people. It wasn't as if he was hiding it for when we were alone in the office. He didn't care who was within hearing distance." As a result, mid- to senior-level management usually knew of the interpersonal maltreatment. Ashley continued: It was no secret in our department...others would come up and talk to me about his approach and say 'what's going on, I don't think he (a contract consultant) should be here working on this project.' Everybody got very concerned about the way he was treating us. Even people who I didn't really work with on a day to day basis would come up and say, 'how can you stand it? He is such a pig.' And I wasn't soliciting sympathy — these were unsolicited comments. Frequently, organizations neglected interactional injustice, despite being directly asked to address it by participants. This organizational inattention may have been taken by bosses as tacit approval to continue the mistreatment of employees. Ashley observed, "He felt that he was free to come up and do this in front of everyone else." Regardless of its origin, an implicit if not explicit sense of entitlement to mistreat others pervaded participants' descriptions of unjust treatment by bosses. Similarly, other types of injustice were often public, such as unpaid overtime outlined in the employment contract or as a daily expectation. Nevertheless, covert elements of unfairness were also apparent in some cases of distributive, procedural and systemic injustice. For instance, in two cases of systemic injustice, one organization was described as 'full of secrets' while another organization's rumoured (but respected) penalty to employees for disclosing their secret departmental performance ranking was instant dismissal. In this latter case, the amount of unpaid overtime worked by employees positively contributed to their ranking: "if you work through your lunch hour, your ranking goes up. And people do work through their lunch hours...most work late and many come in on weekends." While 100 'volunteering time' was ultimately reflected in salary increments, employees had no idea how (or if) their unpaid hours worked were credited to them. In summary, whether perceived injustices were rooted in interactions with bosses or entrenched in organizational protocol, their public and often unrestricted occurrence resulted in hopelessness or cynicism about positive change among participants. In a representative example, one tour coordinator summed up the widespread apathy among staff from the organization's policy of obligatory overwork combined with organizational abandonment: "Everyone's attitude is that it has always been like that, so why try changing it" (see Table 3, quotation'd' under Systemic Injustice). Impact: Trivial or Significant? In describing this second property of organizational injustice, two factors were considered: its scope (i.e., number of employees affected) and participants' judgments of its impact from retrospective sense-making (Weick, 1979), defined as the process by which people interpret their past decisions and behaviour. Scope Typically, participants were one of several employees in their respective organizations to directly experience injustice. For instance, Joe's boss was apparently very proud of the number of 'quits' he effected through intimidation and degradation; "I made the son of a bitch quit" was often happily proclaimed by the boss. In other cases, participants along with many other employees had received unjust performance reviews within their respective organizations. In Kathleen's case, for example, she was 1 of 20 employees affected by the first wave of an organization's downsizing initiative. Within their respective organizations, then, the scope of perceived injustice was broadly felt. Organizational injustice commonly 101 affected many employees within a given organization, including both observers and recipients of unfair treatment. As indicated in several quotations provided earlier, observers, like recipients, were often severely and profoundly affected by organizational injustice. Moreover, as noted in Chapter 2, one participant's inclusion in this study was based on her reported distress from witnessing the mistreatment of her co-worker. Although most observers who witnessed participants' injustice were generally described as supportive (if quietly so), some observers supported the boss and/or organization rather than the participant. For example, when the president yelled at Catriona in front of others "to get the fucking photocopier fixed" (as described earlier), her subordinate publicly aligned herself with the president by supporting his directive: "Yeah, I didn't like this one anyway.'" Among observers, organizational injustice appeared to reduce respect for and undermine the authority of the employees targeted (also see the example in Table 4 under the degradation dimension). Assessing Impact Through Retrospective Sense-Making Whether the past was yesterday or twenty years ago, participants engaged in retrospective sense-making to interpret their injustice experiences and to assess their impact. As Weick (1979) argues, "all understanding originates in reflection and looking backward"(p. 194). In approximately four cases, participants reported that they did not fully appreciate the fundamental injustice of their treatment until they had left the organization. It was not surprising that variations in impact paralleled participants' positions in the process of experiencing and responding to organizational injustice. Two distinctive patterns concerning perceived impact emerged on the basis of when injustices occurred: 1) experiences which 102 occurred within the year preceding the interview; and 2) experiences which occurred more than 1 year prior to the interview. Longer-held experiences of injustice. The majority of cases (22) involved experiences of injustice which occurred more than 1 year prior to their interview. These participants commonly described their experiences as their first exposure to 'abusive, oppressive' behaviour or 'political games' (described as favoritism or 'brown-nosing'). Despite their unfamiliarity with workplace injustice, they called themselves unduly trusting and naive. They reported that, at the time, their experiences of injustice were 'traumatic' and 'horrible,' with many deleterious short- and long-term personal and work-related consequences (these are outlined in Section D). Joe, who suffered a self-described nervous breakdown, reflected "I could have been and nearly was destroyed." Similarly, Margaret described her experience as tragic at the time; she felt lost in the organization and in her career and she felt preoccupied at home and at work by self-doubt, constantly concerned that she was a failure. A common theme expressed was that participants tended to believe that their experiences were life-changing but not ruinous. Although they felt neither emotionally nor intellectually damaged, they did feel different: less naive, more self-protective and cynical. Ironically, these injurious results for the individual and, by extension, the organization were subsequently cited as personally and professionally constructive. For example, following their experiences, participants were often more comfortable with others at work — they were more assertive, they set limits more easily and they were more mindful of possible injustice. Some called their experiences a necessary 'toughening up' which resulted in improved managerial skills, particularly communication. As Andrew said, 103 I understand now the value of open communications and so I have made a big effort to keep really good lines of communications with the people I manage now. I am always asking for feedback — 'am I helping you enough or do you want me to back off?' Likewise, Victor, who had been caught in an organizational web of fraudulent stock promoters reported that his experience gave him a more pragmatic approach to business: I will never (again) believe that just because you have a good idea, it will work. You have to have the right ingredients and the people who you work with are the most important thing. With the wrong people, it doesn't matter what you do. It's an extremely valuable lesson. I don't thank them...(but) in a weird kind of way, I am glad that I went through the experience. From this quotation, a sense of reconciliation to the injustice was inferred. For other participants, a similar but slightly different feeling of resignation was evident. Whether from rapprochement or exhaustion, participants' apparent peace with their experiences seemed influenced by a willingness to take partial responsibility for their mistreatment. Not surprisingly, those who took partial responsibility seemed less likely to feel victimized by their experiences of injustice. Without excusing or condoning the unfairness, many acknowledged that by not clarifying their needs, by not setting limits or by not informing their respective organizations about their discontent, they exacerbated the situation (if inadvertently). Victor, who was not paid for 8 months of full-time work, said "these people (fraudulent stock promoters) are dangerous and should be put away, but if they tricked me then I was able to be tricked." Such self-identified shortcomings, in turn, became the focus for subsequent personal and professional development. However, the irony for participants' organizations was that benefits from participants' improved managerial skills typically accrued to different organizations; by the time that participants had identified, had responded to and had become reconciled with their experiences of injustice, they had left the offending organizations. It is 104 recalled from Chapter 2 that 15 of 32 participants had quit, with 11 citing their injustice experiences as the primary reason for their voluntary exit. Recent experiences of injustice. Of the 11 participants whose experiences of injustice occurred within the 12 months prior to the research interview (including seven participants who were employed at their respective organizations at the time of the interview), the impact of injustice was more equivocal. That is, while some remained confident that they would survive their experiences with no lasting ill-effects, others described deep fears of chronic damage to their self-confidence, self-efficacy and health. As Shezadi said with much emotion, "it's making me nuts." Not surprisingly, beneath the raw emotions frequently expressed were deep-seated doubts and worries about future employment: would they find other work, would new employment relationships be worse than what they had experienced or were currently experiencing? Like participants with longer-held experiences of injustice, participants with more recent experiences conveyed a similar if less developed sense of reconciliation. However, their reconciliation appeared to result from conditions or limits they had imposed on their employment relationships within their respective organizations. For example, some participants continued to experience the injustice only and until they had secured a letter of reference, completed the 6-month probationary period, or found alternate employment. For others, an apparent respite from mistreatment made it easier to stay, 'if only for a while.' Regardless of the specific conditions set by participants, their use appeared to assuage the impact of organizational injustice. 105 Summary Longer-held experiences of injustice (those having occurred more than 1 year prior to their interview) were associated with an overall impact that was at once aversive and transformative: naive, unassertive, idealistic employees became more forthright about work-related needs and more pragmatic. In this transition, their commitment to their organization expanded to include considerations of their own self. While they still believed that their mistreatment was inherently immoral ("it was just plain wrong, it never should have happened"), these participants also acknowledged unexpected benefits (e.g., improved managerial skills). As they reflected, most participants described an innocence lost not to destructive cynicism but to a healthy sophistication rooted in self-protection. Andrew captured the sentiments expressed by several participants as he summed up the impact of his experience of injustice on subsequent employment relationships: I am going to do my job and I will be really loyal to whoever I work for, but if I start feeling like I can't speak my mind or if I start feeling (punished), I am going to be out of there a lot faster. My attitude now is that I'll walk if I have to. I just don't want to go through that again. Consequently, most of these dedicated and wary participants were able to use (or were in the process of learning to use) the experience to their advantage. Exceptions Two exceptions to the pattern of an initially severe but ultimately beneficial impact of organizational injustice were notable. In Chien-chu's case of pay inequity (see Table 3, under Distributive Injustice, Example 'a'), he reported that, compared to the forced relocation of and separation from his parents in his home country of mainland China, pay inequity "was nothing." However, the job with which this experience was associated was his first as a new immigrant to Canada; Chien-chu acknowledged that he had few reference points for 106 workplace (or societal) comparisons of (injustice. As well, his assertion that it was not a particularly bothersome or serious experience was countered by his voluntary participation in a study on organizational injustice. In the second exception, a seemingly minor personality conflict led to an engineer's involuntary reassignment and unexpected dismissal 24 years ago. Despite a graduate degree, over 14 years' professional experience and no prior problems with organizations or bosses, John could not find work. After a year he landed a job in Pittsburgh, but this and many other contract jobs did not last. As he said, his dismissal marked "the start of a very sorry career history." Full-time, steady work was scarce. John could find only casual employment in low-skilled jobs (night security, fruit picking) until i l l health forced him to retire. At the time of the interview he lived alone in a low-income senior citizens' housing complex. Timing: Frequency and Predictability Timing was the third emergent property of organizational injustice. Most participants experienced the timing of injustice as frequent and often unpredictable. The question of timing was most relevant for experiences of interactional and systemic injustice, which have been previously described as iterative (i.e., involving multiple instances). For this reason, cases of interactional and systemic injustice figured more prominently in the results reported here. Joe's description of interactional injustice reflected many participants' experience of the repetitive, relentless nature of their mistreatment by bosses: "By the time I left, we were having yelling matches every day (lasting) probably between 1 to 2 hours." Across cases, daily if not weekly interactions perceived as unjust were customary. Similarly, cases of systemic injustice were often described as repeated incidents of unfairness which seemed 107 relentless and ubiquitous. For example, whether formalized in the employment contract or whether an implicit norm, unpaid overtime was a weekly if not daily expectation by several organizations studied here. Participants typically could not refuse overtime work; in many cases, they believed that their refusals would reduce their performance rankings and chances for promotions. The Paradox of Unpredictability Despite their frequency, instances of injustice were often seen to occur unpredictably. As Ashley noted, "out of the blue, he'd come up with these things (sexual and/or ethnic comments) and I'd think, where is that coming from? It would always catch me off guard." In this unprotected and vulnerable state, she and several others were often unable to respond overtly: "I would freeze...it would shock me so much I was just dumbfounded. I just sat there." Similarly, dismissals were often sudden, leaving participants dazed as they complied with orders to remove their effects immediately. Regardless of the type of organizational injustice, many participants reported a sense of shock, even in cases where one would expect that, by their regularity and predictability, experiences of injustice would be less likely to catch participants unaware. The Use of Threats Dismissal Surprisingly, explicit threats of dismissal were not common, perhaps because their efficacy rested on following through with threatened firings. Joe noted, "He (the boss) would say to me, 'You're going to do that or you're going to be fired.' (But) that became a very hollow threat after you heard it thirty or fifty times, and the entire time I was there, he never once fired anybody." 108 However, the threat of dismissal was a potent weapon when it was periodically acted upon. One case involved an organization with a "myth of an employee who said 'union' and was fired on the spot." Here, one of the top-performing telephone operators was a transgendered woman named Helen, who was awaiting surgery to become biologically female. Helen had used the women's washroom for several months before one employee complained to higher management that she was uncomfortable with Helen's use of the facility. That day, Helen was told that if she used the washroom again she would be fired. She complied, but the overwhelming majority of her co-workers were affronted by what they saw as an assault on Helen's dignity and protested by petition: Nancy, the participant, was 1 of 80 employees who signed the petition. Higher management did not acknowledge or respond to the petition. An employee who then confronted his superiors over Helen's mistreatment (and a newly-instituted dress code) was summarily fired. Subsequently, both public support for Helen and public complaints ceased because of what Nancy described as "employees' dependence on and fear about their jobs." Reference Letter The most common threat was the letter of reference. Specifically, implicit threats to withhold letters of reference were frequently reported. These generated much anxiety among participants and helped, in part, to sustain their perceived dependence on bosses. For example, even when asked in an exit interview why he was leaving, Andrew revealed nothing about perceived systemic injustice. Believing he would be refused a reference, he refused to talk. He also feared that the organization would seek retaliation by surreptitiously damaging his reputation. 109 In Joanne's case, neither distance nor time could reduce her fear of a damaged reputation and of being refused a reference letter. Although she had been on medical leave for lVz years and her rehabilitation was taking place hundreds of miles away from the work setting, she was unable to disclose her injustice-induced stress to her care team for "fear that if I need to find a job she won't give me a reference, although she never gave it to anybody else so why should I get it?" Less commonly, explicit threats surrounding letters of reference were veiled as promises, illustrated by Ashley's boss (a contract consultant) who promised her a strong recommendation and indicated that his organization was happy with her and maybe was interested in her 'coming on board.' D. Antecedents to and Consequences of Organizational Injustice Demographic Differences as an Individual-Level Antecedent Some participants attributed their mistreatment, at least in part, to differences in age, experience, gender, race, sexual orientation, education or knowledge between themselves and an authority figure or among others in their respective organizations. Whether cited as a reason for non-promotion or interpersonal mistreatment, these demographic factors emerged as a distinct source of participants' perceptions of unfair treatment. The degree rather than the direction of the difference appeared salient: thus, bosses much older than participants were described as equally unfair as bosses much younger than participants. As Faith said of her boss who was twenty years younger than she, "I think he is resentful that he has this older Joanne described how she and other employees circumvented the situation: "Because she would never give (ex-employees) a good reference, some have had a heck of a time finding work again. So some have used fellow staff members for a reference. I actually answered a phone call once from (an employer) who wanted a reference (for an ex-employee). So I pretended I was the boss and I felt terrible about that. But she (the ex-employee) had told me the only way she was going to find work was to use her fellow staff members." 110 woman to work with...I am being treated like I am a little old lady and I have never been up against that before." These demographic differences contributed not only to participants' perceptions of workplace injustice but also to feelings of isolation in their respective organizations. The degree of demographic difference (regardless of the specific criterion or variable) represented an imbalance of power to participants, such that they felt less powerful or powerless. For example, as the only employee of Iranian descent among Europeans and North Americans, Shezadi felt completely unable to challenge what she perceived as widespread discrimination from fellow employees on the basis of her nationality. Consequences for Individuals Table 5 synthesizes the wide-ranging consequences of organizational injustice for individuals with resulting conditions (e.g., frustration ), feelings26 (e.g., anger) and behaviours, both at work (e.g., sabotage) and at home (e.g., sleep difficulty), over time (i.e., immediate, intermediate and long-term). In all cases, participants reported that injustice perceptions caused these consequences. As a result, minimal inference was required to determine the consequences of organizational injustice as outlined in Table 5. It is noted that the work behaviours and conditions cited by participants as consequences were both active and passive. Passive consequences included participants' reduced effort, impaired concentration and disengagement, defined as physical, cognitive or emotional withdrawal from work roles (Kahn, 1993). Conversely, active consequences included intentionally making mistakes or engaging in sabotage. Over time, the ongoing experience of organizational injustice grew to dominate participants' lives outside of work as As used here, feelings refer both to participants' emotional and physical sensations. I l l T A B L E 5 C O N S E Q U E N C E S O F O R G A N I Z A T I O N A L INJUSTICE F O R INDIVIDUALS Immediate Intermediate Lone-Term General Positive Negative I s anxiety nervous breakdown less naive high opportunity costs 1s frustration 4> self-esteem relationship strain more self-protective marital separation 4- self-confidence 4s stress burnout more assertive /r> depression failure to achieve personal, 1s anger, resentment professional potential 'f* hopelessness commitment to self I s cynicism "emotionally dead" I s guilt 4s irritability, impatience At Work 4- self-efficacy 4- concentration I s disengagement questioning career I s tardiness choice more responsive to subordinates 4* participation 4* initiative better communication sabotage hating work 1s time off 4- and ^  effort looking for other work I s mistakes (intentional and more mindful of possible injustice unintentional) 4* decision-making ability looking "busy" when not reduced tolerance for injustice missed deadlines intention to quit quitting At Home weight gain/loss stress-related symptoms clearer boundaries between work/home malign organization sleep difficulty less time for or withdrawal from others reduced focus on outside dominant focus on work clearer personal priorities and values activities (i.e., 'what's important') 112 their capacity to focus on non-work activities became more limited and they lost sleep, for example. Table 5 indicates that, over the long term, consequences included positive personal and professional outcomes, such as a more assertive style or clearer personal and professional priorities. However, as previously indicated, because these participants invariably quit their respective organizations, these benefits typically accrued to different organizations. The Impact of Expectations Across cases of interactional and, to a lesser degree, systemic injustice, there was evidence that participants' motivation and performance were affected by their own expectations (self-fulfilling prophecies) and others' expectations (Pygmalion effect). Effects on motivation and performance from organizational injustice were initially affected by others' expectations wherein participants commonly displayed the Pygmalion effect (Rosenthal & Jacobson, 1968) by becoming what bosses had expected them to become. For example, Faith described how, after being treated as if she were 'an idiot' by her boss, she b e c a m e an idiot in interactions with him to meet his low expectation of her. Similarly, Brenda's productivity declined markedly to meet her boss's low performance expectations (see Table 4, the c r i t i c i s m dimension). Notably, participants were distressed by what was, for them, uncharacteristically low motivation and poor performance. In unjust employment relationships, participants' expectations of themselves lowered as they made more mistakes and struggled with muddled thinking, flustered speech, resulting in even poorer performance. The impact of this self-fulfilling prophecy of poor(er) performance worried several participants. As Faith said, "it is eroding my self-confidence to the extent that I'm losing respect for what I do know, for the skills that I do have." 113 In short, whether stemming from self or others' expectations, the feelings of guilt, inadequacy, stupidity and hopelessness ("no matter what I do they're not going to praise me...so what's the point?") linked by participants to low expectations often reduced their sense of self-efficacy (Bandura, 1982). That is, many participants reported that they no longer felt confident that they could perform well enough to achieve their own personal or professional goals. Chapter Summary In this chapter, I presented a four-category typology to classify cases of injustice. This typology departs from existing efforts to categorize types of injustice by delineating interactional injustice as a distinct, substantive component of participants' injustice perceptions. As well, the behavioural dimensions of interactional injustice which resulted from inductive analyses of participants' descriptions were identified and summarized. Emergent properties of organizational injustice were also identified and described: these concerned location, impact and timing, suggesting that workplace injustice was typically public, profound, frequent and yet unpredictable to participants. Lastly, the presence of substantial demographic differences between participants and other organizational members was noted as an individual-level source of injustice perceptions. Wide ranging individual consequences of workplace injustice (general, work- and home-based) were also identified. 114 C H A P T E R I V O R G A N I Z A T I O N A L - L E V E L A S P E C T S O F O R G A N I Z A T I O N A L I N J U S T I C E In this chapter the focus shifts from the individual-level aspects of organizational injustice reported in Chapter 3 to organizational-level considerations of organizational injustice. This chapter contains two main sections. In the first, I examine the organizational antecedents to and consequences of workplace injustice. In the second section, I describe six recurrent images and six themes that broadly characterize injustice at work, resulting from synthesizing individual- and organizational level aspects of the phenomenon. A . Organizational Antecedents to and Consequences of Workplace Injustice This sections reports findings concerning the relative effects of a set of four emergent conditions on participants' perceptions of unfair treatment at work. Derived from inductive analyses of the interview data, these four conditions — structural, procedural, cultural and global — are defined and described in Table 6. As well, each condition is rated with a scale which approximates the frequency of citations of conditions by participants (high, medium, low). The relative frequencies of conditions is a measure, albeit indirect, of their relative importance. In addition, Table 6 outlines relevant properties for each condition with varying intensities or the prevalence (e.g., weak, high) of each property where appropriate. Participants' perceptions of the functional (i.e., job-related) and emotional significance of conditions are also listed. It is understood that while functional and emotional significance are not inseparable, each is a distinct source of meaning for participants. It is also noted that 115 TABLE 6 ORGANIZATIONAL ANTECEDENTS OF ORGANIZATIONAL INJUSTICE Conditions Properties Significance to Participants Functional Emotional STRUCTURAL Hierarchy of authority (H) ambiguous <T> role ambiguity 4> commitment stress, frustration Centralization (H) high 4* autonomy stress, frustration Pay (mixed) a) satisfactory (H) b) unsatisfactory (L) a) ^ commitment b) Is disengagement a) frustration, resentment conflict about exit b) (intention to) quit Formalization (M) low 1s job involvement Is commitment frustration, resentment conflict about exit Physical work setting (L) a) open, communal b) close, individual 4> privacy autonomy under surveillance, invasion of privacy PROCEDURAL Managerial style (H) authoritarian 4 autonomy disrespect, fear Communication (H) absent, one-way, indirect, unclear 4s intra-sender role conflict Is confusion abandonment, alienation, mistrust, stress Performance appraisal (H) general, personal, unsubstantiated confusion about expec-tations, work priorities stress, frustration, fear Individual and organizational decision-making (H) haphazard, irrational, idiosyncratic 1s ambiguity ^ inefficiencies Is disengagement stress, apathy Routinization (mixed) a) high: formal, rigid, structured (H) b) low: informal, chaotic, disordered (L) a) 4 autonomy 4 initiative b) Is confusion about expectations, work priorities a) under surveillance b) stress, job insecurity, self-doubt Socialization (M) weak 4 commitment isolation Hiring and Promotions (M) informal, unclear selection and/or placement criteria unrealistic job previews 1s procedural ambiguity 4 self-efficacy stress, resentment, betrayal, self-doubt Frequency of participants citing condition: (H)igh = 7+, M(edium) = 4-6; L(ow) =1-3. 116 T A B L E 6 O R G A N I Z A T I O N A L A N T E C E D E N T S O F INJUSTICE (cont.) Conditions Properties Significance to Participants Functional Emotional C U L T U R A L Control (H) tight 4* autonomy 4- innovation resentment, resistance Conflict tolerance (H) low 4< problem-solving I s conformity fear, resentment, apathy Supportiveness (H) low /t s disengagement abandonment, alienation Organizational mission, goals (M) unarticulated, ambiguous 1* confusion about work priorities 4/ collective identity disconnection, isolation Team orientation (M) low 4* interdependence 4* collaboration alienation, isolation Outcome orientation (M) high 4* long-range planning, strategic thinking stress, dehumanization G L O B A L Uncertainty (H) high 4* job security i1- competitiveness fear, scarcity mentality Turbulence (L) high 4- job security fear, self-protection Frequency of participants citing condition: (H)igh = 7+, M(edium) = 4-6; L(ow) =1-3. many of the functional meanings of conditions identified by participants are cognitive in nature. As indicated, the four antecedent conditions outlined in Table 6 were adduced from participants' descriptions of workplace injustice. For example, the presence of 'high routinization' (a procedural condition) as indicated by the importance of and time spent on form-filling was a direct means by which Gwen perceived that her boss treated her unfairly: I was abbreviating words on the intake form, so I'd put CL for common law or M for married. But it wasn't good enough. Then I started to compare my work to my co-workers and I noticed they were abbreviating but they weren't being told off for it. So I brought it to her (the boss's) attention, 'if this is supposed to be so important, shouldn't it be consistent with all of them' but I was the one being picked on. The functional significance for Gwen of high routinization was reduced autonomy. Emotionally, she felt scrutinized by her boss and angry at the boss's apparent lack of managerial consistency. Structural Conditions Organizational structure is widely accepted as a set of components by which an organization coordinates its activities and controls its members. The structural conditions described in Table 6 have been widely studied in organizational research: hierarchy of authority (i.e., the allocation of authority and its concomitant reporting structure; Weber, 1947); centralization (i.e., the locus of decision-making authority at the top of the organization; Pugh, Hickson, Hinings & Turner, 1968); and formalization (i.e., the degree to which jobs are standardized throughout the organization and their protocols written; Pugh et al, 1968). In the context of organizational injustice, these structural conditions indicated that organizations appeared rigid yet disordered. Other counter-intuitive findings included the 118 result that low formalization and good salaries did not induce high levels of innovation and initiative. Instead, it appeared that the unlikely combination of both high centralization and ambiguous hierarchy of authority mitigated the potential benefits one might expect from relatively unstandardized jobs and good pay. High centralization meant that decision-making authority primarily rested with top management; thus, whether professionals, mid- to lower-level managers or clerical/line staff, many participants were structurally excluded from participative decision-making. Moreover, ambiguous hierarchy of authority meant that participants were often unsure whom they reported to, which reduced their efficacy and commitment and increased their stress. Together, high centralization and ambiguous hierarchy of authority resulted in confusion about what others expected of participants and about their role, and in frustration toward and stress from their employment relationship. The emergence of this pattern across all three sub-samples made it a compelling finding which is discussed more fully in Chapter 6. Whether organizations were large or small, unclear reporting structures were common. Brenda, a clerk in a company of six employees (including the owner) said, after 4 months on the job, "even now I'm not entirely sure who I report to. I'm often quite confused who to go to, so I'll pick one and they'll say, 'no, go to the other' - it's kind of complicated." Similarly, after 6 months' employment, Faith said of her organization of 100 employees "I have no idea who I report to. Either it's the head writer, but there are two of those, or the head of the department. It's just not clear and it should be." Finally, Jennifer echoed a similar theme in describing her organization of 150 employees: "It was a weird reporting structure because the work leader was actually responsible for the day-to-day work and who 119 did what, when. But we were both responsible to the departmental supervisor who I didn't really have any contact with." In those organizations with clear reporting structures, low formalization and satisfactory pay, participants enjoyed the considerable discretion and autonomy on the job; they called the work itself fun, interesting or challenging. However, these intrinsically-motivating jobs became a significant source of frustration and conflict as participants described their increasing displeasure with their work due to unjust treatment. In Ashley's case, the feelings evoked by experience of interactional injustice after 7 years of fulfilling work were common: It got to the point where I dreaded going to work. Sunday night I would start to get very anxious and that had not been typical of me at all because I had really liked working there...I thought this is a really great project, it was a really great opportunity but it was not worth it. I hated going to work. In short, five structural conditions with varying levels or intensities of properties were identified from interview data as generative influences on participants' experiences of injustice at work. In particular, the combination of ambiguous hierarchy of authority and high centralization played a significant role in the initiation and maintenance of injustice perceptions among participants, for whom both unfairness at work and its effects (e.g., anxiety, reduced commitment) were typically unfamiliar given their previous work history. Procedural Conditions Procedural conditions concern processes associated with organizational functioning (e.g., communication, socialization). In total, seven procedural conditions were identified; of these, authoritarian management styles, poor communication, poorly-conducted performance reviews and haphazard decision-making appeared to have the strongest effect on perceptions of organizational injustice (indicated, in part, by their relative frequencies listed in Table 6). 120 Consistent with descriptions of structural conditions, participants' portrayals of procedural conditions emphasized widespread ambiguity and disorganization. In particular, poor communication appeared to be a significant source of confusion and frustration for many participants which exacerbated their perceptions of injustice. Jennifer's effort to seek clarification about organizational and job priorities was representative of several others' experiences: I would ask them a question, like 'what are your expectations, how would you like me to function?' and there would just be silence - they just wouldn't say anything. It was not a two-way dialogue. I was not getting any feedback on what I asked, if my ideas or attitudes were appropriate. I just felt really uncomfortable. Decision-making at both individual and organizational levels was similarly characterized as disorganized, irrational and ill-informed. Organizational decisions requiring multiple managers appeared consistent with Cohen, March and Olsen's (1972) notion of 'organized anarchy,' which described decision-making marked by ambiguous preferences for goals or solutions, an absence of systematic cause-effect relationships and limited participation by employees. In the present study, disorganization also appeared to pervade the socialization process across several organizations described. For instance, a common complaint related to subsequent injustice perceptions was that organizations exerted little effort to assist participants' adaptation to the workplace. Hiring procedures were also vague and disordered in at least six cases. Specifically, unrealistic job previews (i.e., failure to inform job candidates of the desirable and undesirable aspects of the work; Premack & Wanous, 1985) caused substantial confusion and were a significant source of injustice perceptions. For example, in the second of two cases reported 121 by Catriona, she described being 'misled' in the hiring interview about the nature of the job and the organization's philosophy: In the interview process I was asked how I felt about supervision. So I talked about my philosophy: of equality, so that the Executive Director's work isn't seen as more important than what the file clerk does; that everybody is part of achieving the goals of the organization; that I believe in people working cooperatively and in trusting people to do their jobs; that I believe in personal autonomy so I have a hands-off approach to supervision. They all nodded and wrote this down on their little pads of paper. So when I got hired I thought...that I had a fair degree of autonomy to do my job the way I wanted. But their perception of supervision was the exact opposite of what I had articulated in my interview. Similarly, Karl's hiring interview at an engineering firm was illustrative of several participants' experiences of ineffective hiring interviews: At my second interview I met with my boss. He seemed to be really laid back (like the owner) and there was no real discussion of defined responsibilities....! knew I had a job, I knew it was in a field I was interested in, but I wasn't really sure about (my) responsibilities. I didn't really understand how exactly I would fit in. Even (my) job title wasn't discussed. In the face of such ambiguity, he and several other participants made assumptions which later, to their detriment, proved false. As he said, "I just presumed that they knew what they were talking about, that they probably had their systems in place. You don't expect to be held by the hand, but you do expect that systems are there so that you don't necessarily crash and burn." Unfortunately, with the failure of the project he did 'crash and burn' — psychologically: "without a system there in the beginning and work increasing threefold, it was impossible to control. And so I burned out. Psychologically, I really suffered because your esteem is shot, people are pointing fingers at you." In short, several participants were angered by unrealistic job and/or organizational previews; these emerged as a source of perceptions of workplace injustice. These unrealistic 122 previews created false expectations which, when unmet, were directly linked to subsequent turnover in several cases wherein participants quit or were dismissed. In a graphic case, Joe's efforts to portray the job and organization accurately had serious implications for that organization's selection process. He was in charge of hiring for his department. Within a context he recognized as organizational injustice, he acknowledged that he avoided hiring candidates who were exceptional or experienced for fear "they wouldn't last." Instead, he sought "people who were tough enough to handle the abuse and hang on." In this way, the hiring process was subverted from a search for stellar performers to a search for mediocre, inexperienced ones in order to perpetuate the organization's unjust treatment of employees. In sum, organizational structures and procedures indicated that organizations described in this research were rigid, ambiguous and disorganized. Participants experienced such work settings as high-pressured, stressful environments marked by uncertainty; these conditions appeared antecedent to experiences of injustice. Moreover, many of the organizations described in this study showed little attention to or support for participants' efforts to reduce ambiguity and, by extension, possible opportunities for injustice. Cultural Conditions The conditions included here refer to dimensions which describe organizational culture, widely understood as a system of shared meaning held by members of the organization (Schein, 1990). The conditions listed in Table 6 were first identified as preliminary patterns in the data and later compared against those identified in empirical studies of organizational culture (e.g., Hofstede, Neuijen, Ohayv & Sanders, 1990; O'Reilly, 123 Chatman & Caldwell, 1991). Taken together, these conditions and their specific properties are intended to capture the essence of an organization's culture. Collectively, the configuration indicated by these conditions and their properties in the work settings studied here portrayed a culture of control, suppressed conflict, the valuing of job relations rather than human relations, emphasis on results and production, and a competitive and self-oriented individualism. For example, when asked to describe what is was like to work at his organization, Andrew captured this common culture well: It was very much a 'toe the line' company and there was a lot of butt-kissing, 'yes sir' 'no sir,' because of that. I had seen others who didn't toe the line, who spoke out on issues with opinions different from senior management - they got ostracized, sent to 'corporate Siberia' so to speak. To get ahead you had to play the game or brown nose. It was chaotic, unorganized, short-term in its focus. Problems were never solved by your plan. One-year plans got thrown out the window - it was 'make your numbers for the quarter, month or week.' There was no leadership in terms of any vision. It was not a good atmosphere. In this case and several others, low conflict tolerance emerged as a pivotal norm (Schein, 1980). In other words, participants experienced strong expectations by bosses and/or senior management to suppress conflict or disagreement. To violate this norm by publicly expressing dissent was to threaten one's group and/or organizational membership. As Consuela said: There was the general joke: 'what's the easiest way to get fired? Just say no to the CEO.' So if you wanted to keep your job, you just shut up. The general consensus was that you were stupid if you bought into...the notion that they cared about what you thought and actually asked questions with substance that made them squirm. In many of the organizations described here, there was a strong sense of individualism with little supervisory or peer support. Employee-management relations were frequently depicted as neutral at best and adversarial at worst. Most participants reported that their relations with fellow employees were not supportive, although there were isolated pockets of 124 support among select co-workers. Generally, work relationships were fragmented and characterized by non-recognition such that participants frequently felt ignored. If they did receive attention, it was often in the form of gossip (both harmful and innocuous). Faith said of her organization of 100 employees, It is very cold. I go in and no one says good morning. I say good morning to the two fellows on either side of my cubicle and if I didn't, they wouldn't say good morning to me. If I am there first, they don't say good morning when they come in. I did an experiment one day -1 didn't say good morning to anybody to see if anybody would speak to me. And a whole day went by and no one spoke to me at all. In sum, this example illustrates that conformity, compliance and coldness emerged as features of a common culture, along with control, suppressed conflict and individualism (noted earlier) to form a significant influence on injustice perceptions. Collectively, these conditions formed a composite picture of a 'culture of injustice,' used here as a system of shared meaning held by participants in the context of organizational injustice. Implications for this concept are further considered in Chapter 6. Given limitations in selection and socialization procedures, it is not surprising that the 'person-culture fit' (O'Reilly et al., 1991) was often weak. This lack of fit was a source of considerable distress for many participants, for whom intense alienation, isolation and resentment resulted. Global Conditions As used here, global conditions reflect broad influences on organizations from internal, organizational sources (structural, procedural and cultural) and external, environmental ones. Two global conditions emerged from inductive analyses: uncertainty and turbulence, both typically reported as high. In conditions of high uncertainty, both large-scale change marked by complexity occurs in the absence of full information (Galbraith, 1977) while high turbulence refers to extreme and rapid 125 changes (Emery & Trist, 1965). Conditions of high uncertainty and high turbulence were common in the organizations described. Collectively, the result was a destabilizing influence on employment relationships with direct and indirect impact on perceptions of injustice. Examples included large-scale organizational restructuring which was poorly communicated to participants (if at all), the influence of union certification efforts (or their threat as perceived by organizations) and organizations' cyclic patterns of widespread lay-offs and re-deployment of employees. Nancy related the destabilizing effects of her department's periodic but unremitting retrenchments which employees called "purges:" When you have gone for 6 or 7 months working 40 to 60 hours a week, all of a sudden it is a slow period, nothing is on sale and they start cutting you back. So then you go through 'the purge' where they cut back the people they don't really want, they just get rid of them and cut back hours very drastically. And it went back very quickly to the part-time job it was supposed to be initially. They had only promised us 12 hours, but once you have been working full-time at a job, it becomes your bread and butter, it is your means of survival, and it makes rent hard to pay during those times when hours are cut back. Similarly, Andrew noted the intense fear which typically accompanied high uncertainty: "when I joined the company it was right after a pretty major restructuring and people were walking on egg shells - you didn't want to walk too far out of line." Notably, participants who occupied clerical and line staff positions described their departments and/or respective organizations as unstable, volatile and marked by high turnover. In Nancy's department 97 (described above), the turnover rate was reported to be 40% per year. She felt confident in this estimate by helping to determine it through a staffing analysis conducted during a union drive. 126 Sources of high uncertainty and high turbulence were varied: external factors (e.g., competition, availability of resources, etc.) appeared to drive these conditions in the manufacturing and education/health services industries while other industries and organizations seemed more influenced by internally-driven instability and unrest. Regardless of its source, the high uncertainty and high turbulence which characterized several work settings were antecedent to participants' perceptions of organizational practices and policies as unjust. Organizational Consequences Table 7 lists the immediate, intermediate and long-term consequences for organizations from participants' injustice perceptions. As with Table 6, the data reported here were taken directly from participants' descriptions. However, some inference was required to identify the long-term consequences outlined here. As previously noted, workplace injustice brought few long-term benefits to offending organizations. It is speculated that, over time, chronic effects of individual and/organizational misconduct would include reduced competitiveness and profitability and, in some cases, would threaten corporate survival. B. Characteristic Patterns: Images and Themes In this section, I examine both vivid images and thematic patterns across participants' descriptions of their phenomenological experiences of injustice, of the perceived consequences of injustice and of the broader organizational contexts to synthesize and summarize the individual- and organizational-level considerations involved in workplace injustice. 127 TABLE 7 CONSEQUENCES OF ORGANIZATIONAL INJUSTICE FOR ORGANIZATIONS Immediate Intermediate Long-Term Positive Negative 4s absenteeism conflict resolution improved skills among unjustly ^ cohesiveness treated benefit other organizations ^ problem solving vU collaboration litigation /t> inefficiency coordination reduced responsiveness and 4* morale competitiveness product quality 4> service delivery reduced long-range profitability and compromised survival ^ interdependencies misspent resources Images of Organizational Injustice Six images were notable: five concern perceived acts against participants, both direct (confinement, aggression, dehumanization, surveillance) and indirect (injurious indifference) while the sixth (desperation) illustrates participants' experience of workplace unfairness as one of survival. Confinement The first image portrayed organizational injustice as confinement and was indicated, in part, by language associated with prisons and policing. For instance, when Joe finally quit, his co-workers couldn't believe it because "everybody thought that I was a lifer." Similarly, 128 two participants likened meetings with bosses to criminal interrogations. Images of enslavement dramatized participants' confinement, whether at the hands of individuals ("he was beating us to be these obedient, subservient slaves") or the organization ("we are slaves to the organization"). Andrew's experience of an unjust employment relationship was like being "trapped, even quarantined." Once outside of this context, however, he and others reported a sense of freedom and spontaneity in new employment relationships. Andrew vowed that, in future, 'he'll walk if he has to.' His new capability for self-directed, self-liberating movement was in sharp relief to his (and others') experience of injustice as oppressive confinement. Aggression Across both cases and categories of injustice, images of guns, weaponry and hostile engagement were common. In one case, Jennifer said the that organization "brought in the big guns" for her first performance feedback and that, by its neglect of the consequences of workplace injustice, the organization was "shooting itself in the foot." Nancy said that employees' needs were often "shot down" during management meetings and one manager responsible for firing employees (and many were) was called "the Axeman." Victor described his work life as a "constant battle" to convince the fraudulent stock promoters with whom he worked that the product he was marketing did, in fact, need to be manufactured. Nevertheless, the prospect of "making the big killing" with stock options made leaving the organization difficult for him. In another case, Gwen called her boss a perfectionist who "inflicted her standards on others." As well, Denise's strategy for revenge entailed three stages of sabotage, each increasing in grievousness and culminating in the third stage of "all-out, full-scale war." 129 Several participants characterized organizational injustice as a process of being beaten, whether by individuals or by their respective organizations. For instance, Denise "felt battered" by an employee assistance counsellor from whom she expected help in coping with interactional injustice: It seemed like I was getting beat up at work and then at the session I was getting beaten up too...She (the employee assistance counsellor) kept standing up for him (the boss), taking his side, asking me about all the things that I did that might have ticked him off. She even made fun of me at one point..! almost needed counselling to get over the counsellor. Injurious Indifference Indifference seemed to be an ubiquitous image which pervaded participants' experiences of injustice and their descriptions of their respective organizations. For example, it was at the root of the abandonment dimension of interactional injustice (i.e., bosses' profound neglect of participants' work-related needs) and cases of systemic injustice marked by organizational abandonment. As well, several bosses were called psychopaths and sociopaths (noted for their lack of conscience, among other characteristics) while others were called bullies. Images of damage (both willful or unwitting) to others through indifference were evident in many of the quotations already provided. In one graphic example, Victor described senior executives in the organization who "sue people knowing they will lose but also knowing people can't defend themselves" and whose blatant disregard for responsibility was evidenced by their daily practice of throwing out summons, writs and notices of impending suits. In other cases, sudden and drastic dismissals (whether individual or part of a collective restructuring initiative) were devastating to participants. Involuntary reassignments also had a severe impact on morale, if not income. This was clearly 130 exemplified in Denise's case in her position as a sales assistant to stock brokers. She observed, "it has been said of the investment industry that it is driven by greed and fear — that's what it felt like working there." A major responsibility for sales assistants, whose bonuses (the substantive part of their income) are at the discretion of the broker, was to help build brokers' financial portfolios. After she had helped her broker to rank highest for the office's RRSP contest and to build his "book" from $15 million to $30 million, Denise was summarily reassigned to a new broker whose portfolio was substantially smaller, thereby significantly reducing her income by reducing her bonuses. Her reassignment was consistent with a common organizational practice to reassign assistants to new brokers without consultation or consent. The practice of involuntary reassignment seemed all the more injurious given "there was no cap on our overtime or on how much time we spend working late - if work has to be done, we have to do it." By the same token, in the previously-described case of the transgendered employee named Helen, her denial of women's washroom privileges was widely seen as injurious by the 80 employees who signed the petition to protest both the decision and the way in which it was communicated. Higher management demonstrated their indifference by ignoring the petition. Dehumanization A related but distinct series of images suggested that participants also experienced organizational injustice as objectification or dehumanization. Denise said, "I was treated like a piece on a Monopoly board, or a chess piece." Nancy described the dehumanizing atmosphere in her organization: "In the phone room we were an expendable commodity - you 131 were fired on a moment's notice. There were always resumes, you were told, 'if you have a problem there is a whole file cabinet full of resumes' and that was held over our heads." The physical setting of the workplace sometimes contributed to participants' sense of objectification. For example, Nancy (and many others at the organization) experienced both indifference and dehumanization from the work environment: telephone operators worked in tight, crowded stalls where "you are elbow-to-elbow with the person beside you with a bit of a wall, a narrow space that moved so that if you moved your chair back a bit you would hit the person behind you." Moreover, because higher management "didn't have anything to do with the phone room, they couldn't care less about it. You could see that from the carpets, chairs, old terminals - the oldness and negligence (sic) of the place itself." In contrast, the floor above which housed owners, presidents and marketing and where clients were brought "was a different world: it was fancy - a high-classed world of BMW's, lots of money in a professional atmosphere." In this case, higher management was literal. Perceived differences between the two floors conveyed privilege and personal attention to those of 'the managerial elite' but dehumanizing, personal neglect to those below. Lastly, this same organization introduced a dress code, despite the fact that clients never saw the telephone operators. Many employees enjoyed what Nancy called "alternative lifestyles," but the dress code, which took effect immediately, prohibited unusual hair colours, shaved heads and body piercing. Nancy deplored that her co-workers so quickly changed their hair, removed all jewelry and covered their tattoos; "they...became robots right before my eyes." 132 Surveillance The image of surveillance, represented conceptually in the interactional injustice dimension of surveillance (see Table 4 ) , was identified on the basis of images of close scrutiny frequently reported by participants. For instance, in one organization where a union drive was underway, managers "hovered like hawks" and "patrolled with attitude rather than guns" as they tried to prevent pro-union lobbying. Nancy reported of this experience that "you couldn't even whisper because the walls could talk." In another case, Karl complained that "there was always an unspoken pressure of not spending money because everything was a penny watch from the owner of the company." The physical settings of workplaces were conducive to a sense of surveillance as well as dehumanization and confinement. For instance, telephone operators required to work in tightly-packed stalls (described earlier) had little personal, private space, which facilitated what employees experienced as 'the managerial gaze.' Some participants also described scrutiny by their non-managerial co-workers in more open and communal work settings. For Jennifer, being watched or listened to so intently felt invasive: I would be on the phone talking to a client about a request for something and all of a sudden the person next to me would start shouting out responses that I was supposed to give to the client, like where to find something. So suddenly I would know that everybody is listening to my every word on the phone. In sum, participants' perceptions of unrelenting scrutiny were pervasive and caused deeply-felt discomfort. Work as Survival Many participants reported that working in unjust employment relationships became a daily endurance test, a battle for emotional, psychic or spiritual survival. As Karl described, "I was not living life anymore. I was just seeing each day through to get through the chaos." 133 Joe echoed this sentiment by describing himself as in "self-preservation mode," unable to think of starting a family and unable to fully engage in pursuits or interests outside of work. He was also a part-time graduate business student whose focus increasingly excluded all non-work activities and interests, which were seen as 'luxuries' or distractions to be dispensed with in his struggle to survive at work. In one graphic example, an organization was said to regularly maintain a portion of their workforce on "lifeboat," described by Consuela as "a very sarcastic term for people that are on overhead, meaning you were being paid but you didn't have work. It was such a volatile environment to be in that they called it the 'lifeboat', they were 'saving' you." However, instead of feeling relief at being rescued by the organization, employees' safety in lifeboats felt precarious; they were often highly stressed, unsure when and if there would work and uncertain whether being 'put on lifeboat' was itself an act of preservation or punishment by the organization. In short, these images conveyed a sense of desperation among participants in the face of daily organizational adversity. The experience of work as chronic crisis or disaster management meant that opportunities for participants to develop or use their conceptual or intellectual skills were severely restricted as they fought for their emotional and spiritual survival in the workplace. Themes Across participants' perceptions of organizational injustice, six key themes emerged. Four concerned types of behaviour by organizations — domination, blame, ambiguity, profligacy — while the remaining two — isolation and fear — referred to participants' feelings within employment relationships they perceived as unjust. 134 Toward Domination A central theme throughout participants' injustice experiences was that the use of control and power, whether expressed in interactions with bosses or in organizational policies, represented relentless efforts to dominate employees. Even when seen as unintentional, injustice often resulted in perceptions of domination. Faith summarized, "the aura of that whole organization is that you get talked down to by the top and you don't stand up. We are peons, slaves to the organization." Many described injustice in their respective employment relationships as 'power games' or 'abuses of power.' Joanne observed, "I began to realize that... if it wasn't her idea or her thought, you couldn't use it, you couldn't do it." While recuperating at home from a serious workplace injury, she reported, "after I lost part of my hand and I had quite a serious head injury, she (her boss) would call me up to make sure I wasn't talking (to her rehabilitation team) about anything that happens there." In another exemplary case, Ashley imputed control as her boss's immediate motive in his long range goal of domination: Whether it was because I was a threat and stood up to him...for him it was about establishing control and power in that project. I think I represented someone who knew more than him, and he needed a way to reign me in so he could predict what I was going to do. Concomitant with the desire to dominate was a profound lack of trust, evidenced in part by 'micro-management practices' (Wright & Smye, 1996) such as an inability to delegate or an over-focus on controlling expenditures. Karl described in his organization a widespread management style "which manages prices as opposed to managing the ball game." In the view of several participants, domination or efforts toward it promoted corporate stagnation rather than growth. 135 Blame Many participants reported that, along with failing to take responsibility for mistakes, authority figures repeatedly blamed a variety of others (e.g., subordinates, clients, nature of the work, political situations, the economy) for a plethora of problems (e.g., high turnover, low profits, poor communication). In one case, so-called 'accountability sessions' which were described by higher management as constructive, regular supervisory meetings turned out to be structured, systematic personal and professional shaming sessions, according to Catriona. Moreover, pervasive blame served to perpetuate injustice, as this quotation by Denise illustrated: If management just took the time to say something to the brokers in cases where they mistreat the assistant, it would mean something. But they try to sweep it under the rug or blame the assistant. I certainly felt blamed, they were always trying to make it look like it was my fault. But maybe it's the broker's fault and the organization is just too afraid to acknowledge that so they need to find a scapegoat. Ambiguity Previous sections have acknowledged the major role of widespread structural and procedural ambiguity on perceptions of injustice. Ambiguity in the form of unclear job responsibilities, vague reporting structures and equivocal expectations, for example, meant that organizational responsibility for mistreatment could be evaded more easily. Profligacy Whether involving money, human capital, or time, a theme of wanton waste of resources emerged across cases. Organizations' squandering of real or potential energy, expertise or assets during recession-driven downsizing and cutbacks was deeply resented by several participants who saw both services to clients suffer and perquisites disappear. In many cases, resources were reluctantly misspent by participants in their self-protective efforts 136 to resist perceived injustice. For example, several participants were frustrated by the time needed to 'process emotions,' or worry about or deal with injustice-related 'office politics,' all of which diverted their attention from their work. Inadequate training also wasted several participants' contribution by reducing their efficacy and job performance. As Brenda reflected: "I'm certainly not getting what I need to do a good job...if I were taught it right I could be a pro by now." In one exemplary case, profligate spending was seen to initiate or necessitate organizational injustice. One region of a federal, non-profit organization mandated to serve children received a bequest in excess of $300,000 to assist youth. The region put $50,000 into a youth fund and used the remaining $270,000 to renovate the regional office. Patrick revealed that "their legal interpretation was that the office is a service centre which services the adults and ultimately serves the kids." However, this decision was highly controversial and ultimately divisive. Patrick, who disagreed with office renovation as a priority over youth membership, was told he 'was not a team player.' Meanwhile, "they ran over budget so they didn't have money to refurnish the new office. They couldn't use the old furniture once the new office was built, they had to buy all brand new furniture. So they actually borrowed against the $50,000 fund for youth services to finance the new office furnishing." A radical restructuring (including the participant's ultimate lay-off) was then implemented to reduce organizational operating costs in light of growing debt and declining youth membership. More philosophically, Eva lamented "the waste of natural resources" as her supervisor rejected, through her mistreatment of staff, the goodwill being offered her "without even having to ask for it. She has complete strangers (employees) who are willing to go the distance for her and this is what she does to them. It doesn't make sense." 137 Fear That participants felt unsafe within unjust employment relationships was not surprising. However, the pervasiveness of the fear was remarkable; it commonly defied boundaries of time or space (i.e., geography). For instance, aversive associations remained so strong that some participants could not bear to see the corporate logo or walk past the company building, even several years after their experiences of injustice. As previously noted, Joanne said that she remained afraid to inform her rehabilitation team (located in a city hundreds of miles away) of the chronic stress from interactional injustice which she continued to experience, even though she had been off the job for 1 Vz years and had vowed not to return. Despite her understated, thoughtful manner, her overriding fear was that she would not be believed, that others would not understand "how bad it was, they'll say 'oh, you're just being a baby because you didn't get along with your supervisor.'" In debriefing, she acknowledged that her voice dropped to a near-whisper as she became afraid during the interview from having challenged her boss's unrelenting domination by expressing what she had been repeatedly been ordered to keep secret. Isolation In some cases, participants' sense of isolation was profound. Eva reflected this deeply-felt sentiment in the following quotation describing injustice from a tyrannical supervisor: "the management is no help and the union is no help. You're on your own. There is no court of law that will touch it. She (the supervisor) can do whatever she wants with you, practically." Manifestations of isolation were sometimes varied; some were exemplified geographically, others professionally and still others were emotionally-based. In an 138 illustrative case, contrary to her previous experiences in the engineering industry, Suzanne worked alone in a trailer in a small town with periodic visits from her sexually-harassing boss: David said he didn't want me to get caught up in the politics and all the things that were going on. He wanted me to be able to focus on the project and that made sense, but I always felt that I was kind of isolated from everybody else, including his boss and Human Resources. So, I basically worked by myself with David. Initially it never bothered me but after awhile it started to bother me - he dealt with everything external, and so it was only us at lunch. Thus, she was isolated professionally, both as an engineer and as a woman in a male-dominated industry and job site. Closely connected to this was her emotional isolation; she was without co-workers with whom to talk, problem-solve or confide. Her isolation appeared to be part of a deliberate and strategic effort to facilitate the boss's harassment of the participant. Two properties of isolation were noteworthy. First, across cases a temporal aspect to participants' sense of isolation emerged. For Suzanne, isolation both preceded and accompanied sexual harassment. In contrast, Andrew's feelings of alienation and isolation were tell-tale consequences (for him) of perceived injustice. Second, paradoxically, the alienating effects of injustice were countered by its potential to bring people together. Although some described it as a lonely experience, Irene said, "the only time we got along was (in agreeing) this is a hellish place to work at so we will kind of stick it out for now." Chapter Summary Organizational antecedents of workplace injustice identified from participants' descriptions of their respective organizations were outlined in this chapter. These were grouped into four conditions: structural, procedural, cultural and global. Also, the 139 consequences of injustice for organizations were reported. A set of six images and six themes which characterized experiences of injustice emerged from inductive analyses across multiple levels (individual and organizational). This and the preceding chapter have described and analyzed the phenomenon of organizational injustice. In the next chapter, I focus on how participants responded to injustice at work, namely by its resistance with voice and silence. 140 CHAPTER V RESISTING ORGANIZATIONAL INJUSTICE WITH VOICE AND SILENCE This chapter reports, across four sections, analyses concerning the ways that participants responded to organizational injustice. Findings suggested that responses appeared to be strategic efforts by which participants resisted unfairness at work. Before resisting injustice, participants first had to identify unfairness in their respective employment relationships. Accordingly, in the first section of this chapter I describe the key processes by which participants became aware of individual and/or organizational conduct as unjust. In the next section, I outline participants' varied responses to perceived injustice and categorize them in a typology of voice-based or silence-based strategies to resist injustice. Inductively-derived conceptual meanings of voice and silence are also considered, along with participants' explanations for staying in employment relationships which they recognized as unjust. In the third section, I report findings from the projective exercise portion of the interview in which participants were asked to imagine that they could speak with impunity to the person(s) involved or responsible for the perceived injustice. The fourth and final section reports findings concerning the availability of voice systems across organizations and participants' assessments of their efficacy. As in Chapters 3 and 4, both the archival and interview data were combined to yield the results reported here. However, because the interview data again provided richer information than the archival data about participants' emotional and psychological states, this source figured more prominently in analyses and was used to illustrate concepts and describe processes introduced in this chapter. Nonetheless, the archival data was an important source 141 of information as it provided a strong, triangulated measure on which to ground and provide support for theory emerging from the interview data. Tables and figures with supporting data are provided throughout this chapter. As in Chapters 3 and 4, the written text is not intended to restate or review the displayed data but to present the meaning made from tables and figures by noting patterns and themes, making contrasts and comparisons and drawing and verifying conclusions. A . Identifying Organizational Injustice In Chapter 3 it was reported that interactional and systemic injustices tended to involve repetitive yet ambiguous instances of unfairness. For example, Brenda, who was employed with the organization in question at the time of the interview and was experiencing ongoing interactional injustice said: There are good days, mostly when he (the boss) is not in...I come home from work and I think things are picking up, maybe I'm starting to fit in, maybe I was imagining all the bad stuff and I don't have to look for a new job quite so quick (sic). But then because it was good, the next day is terribly bad and I just scream in my car on the way home. In contrast, distributive and procedural injustices were typically traced to specific incidents involving violations of justice which were readily and unequivocally identified as such by participants (e.g., an unjust dismissal; a performance review which relied on unfair procedures). Accordingly, consideration of the process of identifying injustice applied to cases of interactional and systemic injustice which, as a result, figured more prominently in the results reported in this section. 142 Toward A Trigger Point in Identifying Organizational Injustice As noted previously, some participants reported that their experiences of workplace injustice sometimes comprised incidents which were perceived as progressively severe, resulting in feelings of chronic discontent. These participants typically described a sense of 'mounting problems' which they first ignored, believing them to be isolated events which would likely not recur. Similarly, others minimized the impact of deteriorating relations with hierarchical superiors, believing that they were immune to the possibility of harmful effects. However, as the frequency and severity of these incidents increased, several participants reported reaching a specific trigger point or critical threshold, after which they clearly recognized individual behaviours and/or organizational conduct as unjust. Participants typically reached this trigger point by relying on either an internal source or a combined internal and external source to identify organizational injustice. Internal Source to Identify Injustice When injustice was identified using participants' own judgments and standards, an internal source was said to be relied on. In most cases, the direct experience of injustice was a necessary and sufficient condition to convince participants of its unfairness. In particular, one case illustrated the critical role of participants' internal process to identify injustice; a similar reliance on internal sources was described in over half the interview cases classified as interactional and/or systemic injustice. Initially, Joe witnessed the mistreatment of co-workers by his boss. His use of social comparisons minimized the impact of his boss's abusive behaviour: "I knew that other people in the organization got screwed. I did have a very good relationship with (the boss) compared to everybody else, but it wasn't a good relationship." He was able to strategically protect 143 himself: "I was always able to make sure my interests were looked after." Social comparison and self-protection may have helped him to accept his boss's mistreatment as a necessary motivational device for a sluggish staff: "I always knew that the abuse was wrong, but I always believed he had goals. For a long time I really believed what he was saying — that people weren't performing and it was their fault." Thus, his support for his boss's highly outcome-oriented managerial style mitigated his identification of it as unjust. As well, Joe's relief at escaping mistreatment may have precluded deeply-felt identification of others' mistreatment by his boss as unjust. However, Joe's identification of injustice began when he began to experience maltreatment: "it was only when I started getting treated the same way he treated a lot of people that I began to realize how unfair it was....The fact that people wouldn't get things done wasn't their fault — it was more of an organizational problem." Full identification of injustice took longer still. Even after a nervous breakdown from injustice-induced chronic stress, he engaged in self-blame rather than boss-blame: "to that point, I didn't realize it was wrong. I thought / was doing something wrong, I thought I just couldn't achieve what I should be." It took quitting and being safely settled in another job that he saw his interactions with his ex-boss as profoundly unjust, if inevitable: "nobody could have achieved what he was asking in the ways he was asking us to do it." Despite their reliance on internal sources to identify injustice, some participants responded to injustice in ways that were incompatible with their self-identity as hard-working, responsible employees. As Andrew said of his reduced work effort in response to organizational abandonment, "it just wasn't me." Participants' distress with inconsistencies in 144 work identity typically occurred before their trigger point was reached: their own behaviour did not make sense to them as they had not yet fully identified injustice. In sum, these findings supported the view that identifying organizational injustice was a process of retrospective sensemaking (Weick, 1979) by which participants came to understand their actions within a given context, in this case their actions within the context of organizational injustice. Both Internal and External Sources to Identify Injustice For some participants, external corroboration along with internal judgments were required to identify injustice, despite its direct experience. In these cases, participants reported a growing but vague discomfort that 'something was not right' at work. Typically, they undertook explicit comparisons with others to confirm that they were, in fact, routinely targeted for mistreatment. Others contacted ex-employees for corroboration, advice and support. Regardless of the procedure, these efforts represented ways that participants combined internal and external assessments of the fairness of the situation as part of their process of identifying workplace injustice. For Suzanne, her boss's sexual comments and slaps on her buttocks were troubling but unnamed as unjust until her co-workers, who witnessed the boss's behaviour, identified it as such: A couple of the guys went into his (the boss's) office and said, 'What the hell are you doing. You're just going to get yourself into so much trouble.' He was like, 'What are you talking about? It's no big deal - it doesn't mean anything.' So he had been warned by them, and this I didn't know at that stage. Later, I took the whole team out to the bar. As they started drinking, they started talking about all these things (the boss's sexually-harassing behaviour). And I was like, 'Oh my God, other people are realizing this.' And that is when I started thinking, 'Gees, maybe this is really wrong,' instead of just, whatever. 145 In sum, these participants' progress toward their trigger point of awareness was facilitated by others' identification of their mistreatment. The recurring emphasis across cases on participants' identification and sensemaking of their injustice experiences is consistent with Weick's (1979) view that the problem of meaning is a central issue in experiences, namely experiences of organizational injustice. B. Voice and Silence as Strategies to Resist Organizational Injustice The terms voice and silence, applied in this research to describe participants' responses to workplace injustice, were adopted from earlier organizational studies on dissatisfaction responses and justice. However, as indicated in Chapter 1, these concepts remain poorly understood from theoretical and empirical perspectives. Few studies have been undertaken to inductively and systematically develop the concepts of voice and silence beyond largely colloquial meanings as speaking up and staying silent. Thus, the present research builds on and refines the extant but elementary meanings of these terms within the specific context of organizational injustice. As previously indicated, thematic analyses of workplace injustice portrayed it as including experiences of domination, blame, aggression and ambiguity. Findings reported in this section suggested that voice and silence represented forms of resistance chosen by participants in response to their injustice perceptions. Specifically, voice and silence represented categories of strategies by which participants attempted to resist acts of or the effects of organizational injustice. Participants' motives indicated that, as cognitions or behaviours comprising their responses were largely strategic efforts toward clear goals of stopping injustice, reducing its frequency or mitigating its effects. 146 The concept of strategy as used here is consistent with Strauss and Corbin's (1990) action strategy, by which participants manage, direct, or handle a phenomenon in its natural context. In the present research, however, strategies were adopted by participants to resist the phenomenon of organizational injustice. Like action strategies, strategies to resist organizational injustice appeared to be both processual and purposeful. That is, they changed over time and were goal-oriented. For example, sabotage was one strategy adopted by Denise who developed a three-stage 'action plan' to avenge her boss's mistreatment of her. She carefully constructed each stage of increasingly-egregious acts of property destruction over several weeks and implemented selected parts of the plan over several months. This strategy was thus both a process and purposeful. The notion of resisting captures the scope of participants' varied efforts to oppose individual and/or organizational conduct perceived as unjust. Of its multiple connotations in general terms, I use those meanings of resisting which range from less active responses (i.e., "to withstand the action or effect of") which were associated with silence strategies, to more active responses (i.e., "to strive against; to stop the course of"; Oxford Dictionary of Current English, 1993: 773) which were associated with voice strategies. No implicit evaluation of success or effectiveness in resisting workplace injustice is intended or assumed. Together, participants' strategies to resist injustice and their associated characteristics are outlined in a typology that depicts the three major forms of resistance to organizational injustice: formal voice, informal voice and silence (see Table 8). Conceptually, these forms of resistance represented the substantive responses undertaken by participants to oppose the dominating, blaming, aggressive and/or ambiguous nature of unfairness at work. In Table 8 I label strategies using verbs rather than nouns to emphasize their processual nature (Weick, 147 cu u c —< CO cd — > CL, =2 C3 S « ~ .S3 « <D c (U .S3 S3 a C3 C/3 O 3 O tU a o ca cj © © ft* © © •a co CU CJ '© !> It 5 u .o T 3 CD X T3 tU X T3 T3 X U l OJ T3 O cn 3 O a -a o a ca ca t © fee J? a5 S c*> to cu IS j© a © •§ as be .a -3 ca ca CO f © •ft « CO •o CJ •© > "is £ rt o - 1 Cl, X CO -a S3 x w a S <u X si C3 tU 3 D C/3 J3 C<5 cu Cj S a •a a -s a> -a a O .a ca ca CO CO ^CJ *J cn •rt u. cu CJ as u « O W) s co 3 co Q ca ca •e C<5 a a> ca a 4? •3 © a .is! cj I a © fc. ca I? « •ts ca ca ca -2. 0 5 a fc? a © •e © tse a CO rt 'Si CU «rt 58 rt •+•* CO 1 4 8 1979). Before clarifying the meanings of these concepts, it is instructive to outline the process by which they were identified as resistance strategies. Methodology To generate Table 8,1 began by compiling a list for each case of the numerous actions by which participants handled workplace injustice. I then used several qualitative analytic approaches described by Miles and Huberman (1994) to generate meaning and develop theory from these lists of actions. I first used 'clustering' to inductively sort them into groupings on the basis of identified commonalities; here, emergent commonalities included participants' motives of resistance through revenge and self-protection that influenced the selection of various actions and the meanings that participants ascribed to each action. In most cases, motives were clearly identified by participants. In the few cases where inferences were required, I tested the validity of these inferences by comparing them against data from other portions of the interview; only those inferences which were supported by additional data were included here. As I sorted the actions, I noted the emergent properties of broader groupings or clusters which, when aggregated, acted as rules for inclusion within a given cluster; in this way, each cluster was rendered as internally consistent as possible. I continually re-examined the set of clusters as well as the individual strategies within each cluster to reaffirm their relevance. Where necessary, some actions and clusters were discarded as irrelevant. The remaining clusters of actions were given the collective label of strategies which were then assigned titles which captured their essence (e.g., using formal voice systems, working less, avoiding conflict). 149 From these clusters, I then 'subsumed particulars into the general' (Miles & Huberman, 1994) to identify more basic processes or more general classes underlying them: at this stage, the concepts of voice (both formal and informal) and silence emerged. I then developed these concepts using the process of 'factoring,' (Miles & Huberman) by which recurrent themes or similarities were identified. In this way, common characteristics which served to differentiate among the concepts emerged (e.g., acknowledgment of injustice, desired remedial change). More broadly, in generating Table 8 I used a prototype approach rather than an essentialist approach to categorization (Rosch, 1978). Thus, voice and silence are viewed as distinct categories, albeit with somewhat indistinct boundaries. In terms of philosophical approaches to classification, nominalism (Rich, 1992) underlies the conceptualization of voice and silence as forms of resistance; accordingly, I accept that voice and silence, as intro-duced here, are artificial constructs or consciously-chosen devices to generate culturally-specific and temporally-bounded knowledge. The Concepts of Voice and Silence Findings from this study suggested that, in some instances, responses to organizational injustice were overt and deliberate efforts by which participants actively resisted acts of injustice or their effects through using formal voice systems or directly confronting one's boss about perceived mistreatment, for example. Other means by which participants resisted injustice were more subtle and covert in nature, such as engaging in sabotage, working less and, in some cases, working more. Still other responses were yet more hidden and disguised: avoiding conflict or forgoing formal voice by refusing to acknowledge injustice at exit interviews are some examples. From inductive analyses of the interview and, to a lesser degree, the archival data, three distinct but related concepts which 150 describe how participants responded to injustice emerged: formal voice, informal voice and silence. Voice: Formal and Informal As indicated in Table 8, voice encompassed two distinct but related constructs: formal and informal voice. Formal voice is conceptualized here as overt, deliberate action that is intended to resist acts of or the effects of organizational injustice which involves and/or results in the public acknowledgment of injustice to persons within the organization who can effect remedial change. Three strategies were identified as examples of formal voice: a) 28 using formal voice systems, b) confronting, and c) seeking outside help. Informing the organization of injustice through established organizational voice systems was manifested by activities such as laying a formal harassment complaint, disclosing injustice during an exit interview and informing a new boss of past injustice and its effects to salvage a project. Confronting as a resistance strategy involved informing the perpetrator of injustice that his or her behaviour was unacceptable by refuting unjust accusations or yelling back at one's boss. In seeking outside help, participants publicly acknowledged workplace injustice to those who could effect remedial change, such as union officials or professional associations. Closely related but nonetheless distinct, informal voice is conceptualized as partially disguised but deliberate action that is intended to resist acts of or the effects of organizational injustice which may involve the public acknowledgment of injustice to persons within or outside the organization who can provide support but who cannot effect remedial change. Five strategies were identified as representative of informal voice: a) using informal Recall that these include suggestion boxes, harassment officers or employee-management meetings (see Appendix D). 151 voice systems, b) working less, c) working more, d) seeking validation, and e) sabotaging. Activities which exemplified using informal voice systems included developing and/or invoking organizational support networks (e.g., lobbying co-workers for collective job action), exchanging information covertly through the construction of an 'underground' information channel and disclosing injustice to clients. Seeking validation included documenting perceived injustice through diaries and logs, frequent monitoring of equity by comparing treatment with co-workers and bringing a witness to employee-management meetings. Together, these activities were commonly undertaken in preparation for the subsequent expression of formal voice. Whereas formal voice was largely overt and typically unambiguously expressed to its targets, Table 8 indicates that informal voice was often partially disguised, masked or otherwise indirectly expressed to organizational authority figures. Although more subtle, informal voice nonetheless represented intentional resistance: some participants, for example, adopted the strategy of working more: "The best way that I could fight back was to prove him wrong through my success." Others chose working less by reducing their daily effort, calling in sick or taking long-term stress leave to resist injustice by managing its effects through stress management or by restoring perceptions of equity. As well, acts of sabotage such as surreptitiously shredding the boss's mail, erasing client datafiles from computer storage and speaking badly of the organization "to whomever will listen" were deliberate actions taken by participants using informal or unofficial channels to resist injustice through actively revengeful strategies. In sum, formal and, to a lesser degree, informal voice involve the public acknowledgment of the perceived injustice. Similarly, the degree of change sought by 152 participants was maximal with formal voice (e.g., stopping the injustice through direct intervention) and less so with informal voice (e.g., alleviating aspects of the experience of injustice). Both formal and informal voice were associated with an external focus (again, informal voice less so than formal voice); that is, they reflected behaviours directed away from or outside of the individual experiencing the injustice. Silence In contrast, I conceptualize silence as covert but deliberate action that is intended to resist acts of or the effects of organizational injustice by maintaining the 'status quo' and which does not publicly acknowledge injustice to persons within the organization but which may involve the public acknowledgment of injustice to persons outside the organization who can provide support. As a form of resistance, silence embodied efforts to withstand injustice, rather than to stop or to actively oppose it. Two major strategies — avoiding conflict and forgoing formal voice — comprised the ways that participants resisted injustice using silence. Avoiding conflict reflected a constellation of activities such as consciously withholding signs of injustice-induced emotional distress, acquiescing to or pleasing of hierarchical superiors to evade injustice or its effects, joking, isolating oneself and ignoring injustice. Forgoing formal voice was exemplified by participants who withdrew formal complaints and refused to acknowledge injustice at exit interviews, even when directly asked to do so. In this context, silence described those actions taken by participants which remained fully disguised or hidden from the organization. For instance; while some participants sought personal support by acknowledging their experiences of injustice to friends and family, the organization was not involved in or directly affected by silence-based strategies of resistance. Thus, unlike voice, silence was indicated by an internal focus on the individual experiencing 153 the injustice, rather than on the organization. In short, silence as conceptualized here is not intended to suggest the absence or omission of actions in response to workplace injustice; instead, silence as expressed by participants conveyed a constellation of cognitions and behaviours which represented fully-disguised resistance. Characteristic Patterns Several recurrent patterns or themes concerning the strategies adopted by participants across cases were noteworthy. In many instances, these emergent themes highlighted distinctions among silence, formal and informal voice. These thematic distinctions are included in Table 8. Timing. Whether participants' strategies were diverse or similar, they were sometimes employed simultaneously and sometimes in sequence. For instance, several participants informally voiced their discontent by working less while they also adopted silence, at the same time, by avoiding conflict to deflect ongoing mistreatment. On the other hand, some participants like Suzanne used resistance strategies in sequence: her prolonged and exclusive use of silence was followed by formal voice which brought a rapid and decisive end to her boss's unfair treatment of her. These examples illustrate a key pattern concerning the simultaneous or sequential use of strategies: participants tended to use silence and informal voice strategies at the same time but they tended to use silence and formal voice in sequence. When silence and formal voice strategies were used in sequence, there was no pattern to their order of appearance: silence could be preceded by or followed by formal voice. In Ashley's case, formal voice was followed by silence: she formally voiced to her superiors her discontent due to interactional injustice. Following their inaction, she then used a simultaneous series of informal voice 154 (e.g., seeking validation) and silence (e.g., avoiding conflict) strategies to resist the ongoing injustice. Blame. A recurring theme of internal or self-blame was reflected in self-deprecatory language by some participants who frequently called themselves 'stupid,' 'naive' or 'dumb' in describing their failure to recognize injustice sooner than they did or in describing their overall response to the situation. For example, Joe summed up, "he (the boss) sure got a number one sucker in me." These participants believed that they should have seen and/or appreciated the significance of 'early warning signs' of mistreatment, despite having had no prior experience of injustice at work. The implication in this and similar comments was that participants were at least partially responsible for workplace injustice. In contrast, other participants placed blame firmly on certain individuals and/or the organization for conduct they viewed as unjust. A key pattern here was that participants who tended to remain silent were more likely to blame or criticize themselves whereas those tended to who voice their discontent blamed others (i.e., individual and/or organization). This finding supports the conceptualization of silence as internally-focused and voice as externally-focused strategies to resist organizational injustice. Optimism. Generally, the use of voice was associated with a sense of optimism about the possibility of positive change whereas silence was associated with deeply-felt pessimism, hopelessness and cynicism about change. Analyses of participants' motives support this finding. Specifically, from participants' perspectives, formal and informal voice represented personally-constructive change efforts directed toward stopping the injustice or countering its 155 effects. Thus, seeking revenge through sabotage was seen as an effective way to resist injustice. For instance, in Shaun's case formal voice through legal redress was very satisfying, if arduous: "They fired me for their perception that they would lose money by having a union in there. So my feeling was 'I will cost you every dollar I damn well can, and it is going to cost you big.'" Similarly, Denise's use of informal voice through seeking validation by documentation in the form of a diary of both her perceived mistreatment and her sabotage to avenge her mistreatment was a substantial source of comfort; her diary, which she read whenever she needed consolation, was a satisfying reminder to her of the actions she had taken to get back at her boss. However, silence through avoiding conflict was typically associated with an unwillingness to engage in change efforts because of their perceived futility: Jennifer, who intentionally ignored repeated acts of injustice, said "Nothing will change - what's the point in doing anything?" Collaboration. Consistent with its external focus, voice appeared to be a collaborative experience. In particular, formal voice involved direct contact with organizational representatives (e.g., bosses, ex-bosses) or external agents (e.g., unions, lawyers) for remediation while informal voice involved unofficial contact with organizational affiliates 1 (e.g., co-workers) for support. In contrast, silence involved less contact with others, and those contacted were not affiliated with the organization (e.g., family and friends). Efficacy. Efficacy was defined here as the degree to which strategies selected by participants fulfilled long-term goals. Many strategies to resist organizational injustice appeared effective initially by providing immediate self-protection or security, for example. 156 However, in at least nine cases, such apparent success was mitigated by deleterious consequences experienced by participants from the chronic stress of workplace injustice. Over time, silence appeared less effective than voice in resisting organizational injustice. For example, participants who ignored injustice usually reported that their mistreatment (whether by individuals or organizational protocols) increased rather than decreased in the long run. Similarly, many participants realized that avoiding conflict by pleasing was ineffectual: they never seemed able to do enough or find the right words to placate the boss, for example. Although generally more effective than silence, informal voice yielded mixed results concerning long-run efficacy. For example, the informal voice strategy of working more was not effective over time; several participants who thought they could effect an end to the injustice by working harder reported subsequent self-described 'nervous breakdowns.' However, the strategy of using informal voice systems appeared effective in providing participants with the emotional endurance needed to manage injustice over the long term. Similarly, formal voice sometimes resulted in enduring remedial changes sought by participants. In other cases, however, the results of formal voice were a "complete disaster" with violations of promised confidentiality, for example. As well, the strategy seeking outside help brought mixed success: several participants reported their frustration with union representatives who repeatedly failed to return telephone calls or who failed to provide support. For example, Jennifer said to her union representative, 'what am I supposed to do?' She said 'smile, relax' but she didn't really have any suggestions as to what I was supposed to do, if anything." Likewise, John perceived that the professional association from which he 1 5 7 sought counsel regarding his unfair dismissal lacked the commitment to effect remedial change: "it turned out to be a big zero - there was nothing there, no muscle." Additional Patterns Joking was used both in formal voice and in silence. An example of its use in formal voice occurred with Suzanne, who strategically used humour to disclose sexual harassment to an ex-boss who remained with the organization and who could (and did) effect remedial change: I called up my past boss, we had no reporting relationship at all, just 'a friend keeping in touch' thing. We were talking about Frank (current boss) and I purposefully jokingly said "Well, I think somebody should put him through sexual harassment training again." He said "what are you talking about?" And I said I wasn't sure what I wanted to happen with that, I just wanted to let someone else know. And so I was kind of trying to make it jokingly, but I said there are just some things that Frank does that aren't very appropriate....He said "I think this needs to be reported" but I said, "no, I don't want to cause any problems -1 just want people to know so that if things happen you know, so people are aware of it after I leave." Strategic joking was used both by bosses seen to perpetrate injustice and by participants against whom injustice was perpetrated. For both bosses and participants, the primary function of joking was to deflect, disguise or otherwise minimize responsibility for their actions, whether those actions involved perpetrating or responding to injustice. For example, in several archival cases of sexual harassment, bosses defended their behaviours as 'jokes' which complainants misunderstood and/or misrepresented. Similarly, at times complainants initially used joking as a silence-based strategy to reframe their boss's injustice toward them as jokes, thereby deflecting it. However, joking did not appear, over time, to reduce the frequency or severity of episodes of injustice. One unexpected finding was that only four participants consistently used voice-based or silence-based strategies. Given wide-ranging differences across both participants and 158 organizations studied, some variation of strategies adopted to resist organizational injustice may be expected. What was surprising, however, was the diversity of strategies adopted within participants: most participants used strategies from both voice and silence approaches. I had also expected that 'prototypical profiles' of voice and silence responses to organizational injustice would emerge. However, one consequence of participants' use of diverse strategies was that such profiles did not emerge. No patterns emerged between the types of organizational injustice and resistance strategies. Other Sources of Resistance: Self-Attributions for Staying Four main reasons were cited by participants to explain why they maintained their employment relationships despite ongoing organizational injustice. They are as follows: 1) social conditioning; 2) loyalty or obligation; 3) opportunism; and 4) perceived immunity. These explanations were not mutually exclusive and did not preclude the use of voice or silence; in some cases, these factors appeared to influence the selection of voice or silence. Social Conditioning Several participants attributed maintaining the employment relationship to a predisposition to or to a familiarity with injustice due to maltreatment in childhood. For example, Eva said that her supervisor-to-be was an ungracious woman who looked at her coldly and said during the hiring interview, "I guess the job is yours, nobody else has applied." The supervisor also had a widespread reputation for "savaging" staff of which the clerk knew. However, she was unable to act upon these portents of impending injustice because "I was so used to taking shit from people as part of my abusive background. I could get along with any one...but there were all the little things you say to delude yourself. I should have just run, just run." 159 A comparable experience was reported by Karl who linked the paternalism in his Germanic family of origin to his willingness to accept orders from authority figures in a family business: "I had the assumption that the guy above you and the owner knew what they were doing, so you didn't dispute them...you just carried through with it (the work)." This misplaced trust led, in part, to a serious project failure for which he was held responsible. However, he felt the organization bore 70% of the responsibility for the failure due to its unclear procedures and unstated expectations and because it did not provide him with the authority commensurate with his organizational responsibility. As part of his 30% responsibility, he accepted that he failed to see it as "a dysfunctional organization," suggesting that his upbringing predisposed him to accept behaviour which he ultimately came to see as unjust and unacceptable. Influence on formal voice. Social conditioning also influenced some participants to formally voice or not to formally voice. For example, Eva explained her reluctance to formally voice her discontent: "I am not effective in standing up for myself due to childhood abuse." Conversely, other participants attributed their willingness to formally voice to an upbringing which fostered self-esteem and personal pride. Loyalty and Obligation For Karl, both his sense of partial responsibility for the project failure and his self-described masochistic tendencies influenced him to continue to stay with the organization (where he still worked at the time of the interview). Together, these influences formed a perceived obligation to 'make good' or redeem himself (and the organization) on a second project, despite 'hating every minute of it.' In contrast, others' senses of loyalty seemed to stem from a general dispositional tendency (whether genetic or learned) to be highly 160 committed with a strong work ethic. In one case, Margaret lost a promotion as project leader to an external candidate whose own work was considered by her and others to be sub-standard. However, for the good of the project, the team and the organization, she supported his promotion by correcting his mistakes and informally assisting the team: "There were many nights when people would call me at home at four in the morning because his work had blown up, and I had to help them through." In another case, Victor's self-described 'fanatical enthusiasm' for a revolutionary environmental product blinded him to the dangers of working with 'sociopathic stock promoters.' As he said, "I was so absolutely dedicated to it. I really believed in it -1 just didn't know it was impossible." His extreme commitment to the product, along with a perceived duty to honour unfulfilled promises he had made to clients, 'made it hard to walk away.' For other participants, staying was less due to altruistic loyalty than the need to meet financial obligations (e.g., mortgages). Regardless of its source, loyalty was commonly associated with conflicting feelings about the decision to exit. Opportunism: Money, Ambition, Risk-Taking Whether to secure a letter of reference, gain experience or make a lot of money, many participants cited opportunistic reasons for staying in an employment situation which they recognized as unjust. Some were highly motivated to stay by extrinsic factors such as pay and other benefits; Victor clarified that "the salary wasn't too exciting for me but I wanted the (stock) options because I knew this was going to be enormous if it worked." Intrinsic factors such as power and responsibility which the employment situation offered, despite perceived injustice, were also motivating influences: Victor reported, "I ran everything which was absolutely exciting and I didn't get paid anything at all. I have never had that much fun 161 working or been as enthused in something before, which is why I allowed much more (injustice) than anyone else would." Similarly, the excitement and stimulation which accompanied the chaos and uncertainty in many of the work settings described in this study were attractive features to some participants. Victor said, "when it worked it was an absolute high. It was like 'look what I did. I got people to do this.'" In short, for some the risk-taking integral to the swings of emotional and financial highs and lows was a compelling reason to stay in volatile and unfair workplaces. Combinations of extrinsic and intrinsic factors also influenced decisions to stay. Joe summed up his conflicting feelings: "He (the boss) was a son of a bitch, he treated me like shit but he also taught me a lot." However, in considering whether the benefits outweighed the costs, he responded in a poignant moment of public admission with tears in his eyes, "No, on balance it was not worth it. Because I could have been somewhere else doing something else. I tell people that it was worth it, but it wasn't." Perceived Immunity Many participants stayed, at least in part, because they minimized or underestimated the impact (emotional, physical and cognitive) of chronic, daily injustices. As Joe said, "I thought I could take it, I really thought I could take his shit. I thought that for years. But it's not until you start being ground down, until I had the breakdown that I realized, 'no, you're wrong, you can't take this' - it gets to you after awhile." Whether it was pride, stubbornness or a belief in the protective power of their resiliency and determination, several participants Poignant, to me, because of his notable composure and restrained, unemotional demeanor during the interview. 162 reported that these 'defenses' were slowly but steadily eroded, at which point quitting became a last-ditch strategy to survive. C. Projective Exercise Results In this section, I report findings from the projective exercise portion of the interview in which participants were asked to imagine that they could now speak with impunity to the person(s) involved or responsible for the perceived injustice and say whatever they wished to say. The projective exercise was adapted from Scott's (1990) concept of the private transcript (i.e., verbal and non-verbal behaviours hidden from dominant others; see Chapter 1). I focused here on the discrepancy or gap between participants' private and public transcripts (i.e., verbal and non-verbal behaviours expressed in the presence of dominant others). For the purposes of this research, participants' public transcripts were seen as their interview data which described public responses to injustice. Thus, to assess the gap I analyzed differences in content between interview portions which related what they reportedly expressed to dominant others (e.g., hierarchical superiors) and what they expressed during the projective exercise. Contrary to expectations, assessing the gap between participants' public and private transcripts proved problematic. The primary reason was the difficulty in associating a given participant's transcripts (public and private) with voice or silence, given participants' lack of consistency in responses and the absence of emergent profiles of voice and silence (as noted earlier). Accordingly, accurate inferences about the gap could not be made from these data. Despite the difficulty assessing the private versus public gap, there were several notable findings concerning private transcripts themselves. 163 Patterns Across Private Transcripts The Effects of Voice and Silence on Anger, Frustration and Revenge In the few cases where participants consistently used silence, their private transcripts typically reflected an angrier tone (indicated by swearing) and expressed a greater desire for revenge than did the private transcripts of those who consistently responded to perceived injustice with voice. For example, Andrew, who primarily used silence to manage his experience of injustice, said in the projective exercise: I think what you (the boss) do is complete bullshit. I hope I get the chance someday to come back to this company and be your manager because I am going to fry your ass. If I ever get this chance, you will be gone in a moment -you and the rest of the managers in the company. This example indicates that resisting injustice with voice mitigates participants' injustice-induced anger, frustration and/or desire for revenge. However, because it is based on approximately three cases, this finding is necessarily tentative. Additional support (if indirect) was found in the way that Joe used mixed strategies (i.e., voice and silence) to resist interactional injustice. Because he repeatedly confronted30 his boss (i.e., formal voice), he had little to say during the projective exercise. He explained, "I'm not angry because every time I was mad at him I yelled at him. I didn't have a problem expressing my feelings to him at the time that I was unhappy. For example, I would say, 'you're wrong. That's unfair.'" Surprisingly, in cases of interactional injustice wherein participants used both voice and silence, many declined the opportunity to 'wax vindictively' against their bosses, even when prompted to do so in the interview, because "it wouldn't be worth it." Instead, several Joe's use of confrontation did not include expressing his deep-seated discontent at the boss's abusive treatment of him and others: "I never let him know my emotional feelings about things....I don't know if I ever had the balls to say 'I can't do what you want me to do' because that would have meant I was admitting failure and he would have said "Well, if you can't do it, I'm going to have to get somebody else who can." 164 expressed pity for bosses and speculated about unhappiness in their bosses' childhoods or current situations which might have made them act unjustly. Given the reported impact of injustice and the intensity of emotions often expressed during the interview, I did not expect this degree of participants' compassion in the projective exercise. Acknowledgments, Accounts and Apologies Regardless of participants' use of voice or silence, in the projective exercise over half said that they wanted the person or persons responsible for their perceived injustice to acknowledge, explain and/or apologize for their conduct. Some participants also sought to inform the individuals and organizations responsible of how the injustice affected them. By far, the most frequently asked question during this portion of the interview was why: why were participants targeted to receive injustice, why did the boss act unjustly, why did the organization appear to condone workplace injustice and why was the injustice not stopped (or not stopped sooner). These questions were typically asked with a tone which combined genuine curiosity with firm insistence for an answer or explanation. Volume: The Relationship between Silence and Fear Another unexpected finding concerned the connection between silence and fear as indicated by the volume of several participants' conversations during interviews. It was noted in Chapter 3 that, at several points during one interview, Joanne's voice dropped to a near-whisper from fear of retribution from expressing what she had been repeatedly been ordered to keep secret and fear that her experience and its devastating personal (i.e., marital separation) and professional effects would not be believed. In four other interviews, participants (all female) spoke so quietly that transcribing the interviews was problematic. In 165 Consuela's case, the entire interview required re-recording with special audio equipment to artificially amplify her voice. Together, it appeared that describing their experiences in their entirety rekindled extant fear and/or generated new fear. As a consequence, lowering their volume emerged as a spontaneous strategy to manage describing the effects of injustice. Although there were no apparent gender differences across participants' use of voice or silence, the fact that problems with volume occurred only among female participants provided indirect, preliminary evidence of gender-based differences in voice and silence. D. Voice Systems In/Action In this section I report results concerning the availability of voice systems for participants to acknowledge injustice and express their discontent, including participants' assessments of their utility. Availability of Voice Systems Voice systems refer to organizationally-sanctioned procedures and policies by which employees can participate in decision-making (Robinson, 1994; Sheppard et al. 1992), or express their content or discontent; the present research focuses on the expression of discontent. Voice systems range from formal (e.g., formal grievance procedures in both unionized and non-unionized environments) to informal (e.g., open-door policies) avenues. As indicated in Chapter 2, a criterion for participation in this study was the availability of at least one voice system in participants' respective organizations. Robinson's (1994) measure of the availability of voice systems (in turn adapted from Spencer, 1986) was determined by asking participants to respond with 'yes,' 'no' or 'I don't know' to whether each of nine voice 166 systems was available in their organization (see Appendix D). In addition, participants were asked to identify the one (or two) voice systems which they believed were most effective in encouraging employees to express discontent. Some participants also volunteered their views on why certain voice systems discouraged use. Index of voice systems. Consistent with Spencer (1986) and Robinson (1994), I generated a composite index of voice systems by summing the number available to participants in their respective organizations. The number of available voice systems ranged from one to seven with an average of three per organization (s.d. = 2). 3 1 The modal response was one; of those organizations with only one voice system, the open-door policy was cited most frequently as the one available to participants. Assessments of Efficacy Contrary to Spencer (1986) but consistent with Robinson (1994), high opportunities for voice (indicated by higher numbers of available voice systems) were not always associated with participants' high expectancies for problem resolution or high levels of effectiveness (indicated by participants' assessments of their organizations' response to employee voice). For example, in one case involving an organization with seven voice systems, Consuela described what happened when she complained: I wrote an e-mail to the man who was responsible for the woman who screwed up in the travel office. Although he was a VP, he was the one telling us there is an open-door policy and anybody can go in and talk to him. I used e-mail instead. But he fired my e-mail to the woman in the travel office — I don't even know if he took the damn time to read it — so then she came storming into my office and chewed into me. It blew up in my face, her coming in and barging in my office. What kind of an open-door policy is that? If I had wanted to talk to her, I would have gone to her. 3 1 These descriptive statistics are based on an N of 30. 167 Moreover, she informed the V P of her dissatisfaction with the voice system implementation, but received no response. She described several other examples of so-called 'complete disasters' following her own and others' efforts to voice their complaints, which led to widespread mistrust and cynicism among personnel of the organization's genuine commitment to hearing employees' concerns. In the second of two cases where seven voice systems were available to organizations, Ashley, who was happy with the organization and her job, reported that the organization's response to her voicing effort exacerbated her perceptions of injustice: I told my bosses about it (mistreatment from a contract consultant) - they could tell that I was really upset about it, but they felt really tied to do anything because they didn't want to jeopardize the financing. They thought if they made waves, even though he was an outsider, that it would jeopardize the project and they weren't willing to do that. They knew it (the consultant's behaviour) was inappropriate, but they weren't willing to do anything about it....They made it very obvious that they weren't going to pursue anything. A n d their preference was to remove me rather than to deal with the situation. She described her expectations in using the organization's open-door policy: What I was looking for was some direction from them about how to deal with it. Not necessarily that they were going to go blow the whistle, but help me deal with or get through the situation — what I should say, how I should act, what kind of behavior I could change to lessen the situation. But they didn't give me any of that. The organization's inaction had the following impact: I felt I had no support -1 felt really let down. I could understand that they didn't want to jeopardize the financing, because it was a major deal for the hospital. But at the same time I couldn't believe that they would sit back and watch two of their senior staff go through this. M y esteem for them lessened a lot after that. She quit shortly thereafter, citing this experience as a contributing factor in her decision to leave the organization. 168 These two cases were illustrative of the finding that open-door policies, in particular, were widely seen as ineffective despite their prevalence (they were reported as available in 30 of 33 organizations). Approximately one-third observed that their respective organizations had open-door policies 'in theory but not in practice,' suggesting significant disillusion with or cynicism about their efficacy. As Faith observed, "as much as there is an open door, you have to be careful of what you say when you walk through it." Joanne echoed this sentiment in describing the organization's open-door policy, "the word that went around there all the time was that even though you were told that the door is open, God help you if you walk through it." At the same time, ironically, slightly less than one-third of participants identified the open-door policy as the voice system with the greatest potential to encourage employees to express discontent or acknowledge injustice. In considering the conditions required for its effective implementation, participants cited the importance of interpersonal skills of the authority figure, particularly communication skills. In accessing an open-door policy, participants pointed to their need to feel not simply listened to but 'heard and understood' and their need to feel valued by their respective organizations. Many acknowledged that they expected 'action' after using the open-door voice system. However, in their views action could take many forms, from ongoing monitoring of the situation, to 'coaching' with a problematic situation or to providing a coherent, plausible account to justify the lack of direct remedial action. Regular employee-management meetings, as a voice system were also favored by participants, although to a lesser degree, than open-door policies. Reasons cited included the need for face-to-face communication for personal recognition of employees and for team-169 building, the opportunity to demonstrate managerial accountability and to provide a forum for interactions and exchange. However, participants clarified that for such meetings to be effective, employees must feel safe, uninhibited and able to be honest. The meetings themselves sometimes became an avenue to perpetuate injustice: several participants described instances of public degradation or intimidation by bosses (e.g., yelling at employees, calling them stupid). Of the remaining voice systems, each was cited by one or two participants as most effective in encouraging discontent. In sum, the open-door policy and employee-management meetings clearly emerged as the two voice systems believed capable of effective management of employees' concerns and of building the employment relationship. Several participants reported dissatisfaction with the use of counsellors connected to organizational employee assistance plans as voice systems for work-related discontent. Of the three participants who reported using counsellors, one described her experience as 'neutral' in terms of problem resolution while the remaining two reported that their perceptions of injustice were increased following their sessions. This quotation described the neutral counselling intervention reported by Jennifer: A l l the time I wondered 'what am I doing wrong now?' but I didn't know what. So I started to feel really, really scared - very, very scared. The company had an employee assistant program so I went and started talking to the counsellor and her advice was 'do your job, work hard and produce good quality work -what can they say about your performance if your work is really good and you are doing your job?' So that's what I focused on. However, this approach ultimately was not effective. Despite her attempt to avoid censure by redoubling her work efforts, mistreatment from her boss and the larger organizational system continued until her dismissal several months later. In the remaining two cases, counsellors' impartiality seemed questionable if not absent. As previously described, Denise's counsellor 170 immediately took the side of management and ridiculed her. In Vera's case, she felt racially discriminated against by the counsellor, whom she believed had an inherent conflict of interest which precluded the voice system from responding to her own and others' needs. Suzanne's case was one notable exception to the many negative experiences with voice systems. As previously noted, she worked in a small town in B.C. in a trailer with her boss who began sexually-harassing soon after her arrival. The organization had offices throughout the province and provided a telephone counselling service for its more isolated employees. Although this gave employees from remote areas anonymous access to a qualified counsellor, Suzanne was precluded from calling because its regular day-shift hours of operation meant that her opportunity to call was severely restricted: with increasing frequency and severity of the harassment, she felt safe to call only when her boss left the trailer, yet his absences and their duration were unpredictable. However, when Suzanne did disclose by 'joking' to her previous boss about her present boss's behaviour, the response was both immediate and effective. Despite being three hours away by car, the previous boss arranged lunch with her that day and called ahead to the human resources department to set up a preliminary meeting. Following lunch, he accompanied her to the meeting and the organization's harassment investigation process began. Suzanne described the conduct of personnel as respectful, efficient and sensitive; she was deeply relieved and grateful that they never once asked why she had not previously acknowledged or directly dealt with the situation. In short, most participants described their own or others' experiences with their organizations' voice systems as ineffective at best and injurious at worst. Whether because of 171 poor interpersonal skills, general indifference or an insincere commitment to employees' welfare, the voice systems reported here tended to further participants' frustration, cynicism and injustice perceptions, thus providing substantial evidence for the frustration effect (i.e., increased frustration or dissatisfaction from the exercise of voice). As Robert summed up, "for voice to be used, there needs to be a message that we (the organization) want to hear a voice." Across most of the organizations described, that message was missing, lost or otherwise unaccounted for. Chapter Summary Resisting injustice at the workplace was a complex and varied process for participants which first required reaching a trigger point or critical threshold to identify the experience as unjust. The concepts of voice and silence were defined and described as strategies by which participants resisted perceived unfair treatment. I also considered reasons cited by participants for not terminating their employment relationships (or doing so sooner). In addition, findings from the projective exercise suggested that voice mitigated participants' anger from injustice. Contrary to expectations, the gap between public and private transcripts could not be assessed. During the projective exercise, participants' most consistent (and insistent) demands were for an acknowledgment, an accounting for and an apology for the injustice and/or its effects. Finally, participants' assessments of their organizations' voice systems were reviewed, with evidence of the frustration effect and, to a lesser degree, the fair process effect from their use of organizational channels for the acknowledgment of injustice and the expression of discontent. 172 C H A P T E R V I D I S C U S S I O N A N D C O N C L U S I O N S In this final chapter, I integrate the results of this study with existing theory and research and consider their implications. To this end, I first present an overview of the significant findings of the study. In the next two sections, I examine findings concerning the phenomenon of organizational injustice and the concepts of voice and silence, respectively, in relation to existing studies and consider implications from this research for managerial and organizational practice. In the fourth section I introduce a process model of voice, silence and organizational injustice which synthesizes the main findings of the study. In the final section, I consider limitations of the study and several avenues for future research. A . Overview of Significant Findings This study is consistent with a long-standing tradition in organizational research which links consideration of human needs, motives and attitudes to organizational functioning. Known as the human relations approach (Argyris, 1957; McGregor, 1960), this tradition is based on the premise that organizational effectiveness depends on individuals' satisfaction of personal and professional desires. Present findings suggest that participants' work needs were consistent with those identified by Argyris: self-actualization, competence, responsibility and commitment. Also, when the satisfaction of these needs was blocked, patterns were observed which were consistent with Argyris' findings of individuals' decreased psychological energy, as indicated by consequent apathy, withdrawal and feelings of alienation among participants. Furthermore, results were also consistent with Argyris' 173 notion of organizational atrophy, described as a state wherein increasing energy is needed by the organization to complete simple tasks. In the present research, this impaired organizational functioning is the direct result of organizational misuse of human energy which was required by participants to resist organizational injustice. Thus, the present results suggest that organizational injustice may be a source of atrophy. More broadly, this research has examined issues consistent with those studied within the parameters of the human relations approach (i.e., job dissatisfaction, perceived injustices, concerns of individual dignity and self-worth at work). As well, it takes its place in a similar but more recent qualitative tradition of 'atrocity stories' in organizational settings, which include accounts of professional incompetence and its consequences, for example (see Coffey & Atkinson, 1996). However, the present study departs from previous approaches in several ways. First, its focus is the phenomenon of organizational injustice and processes integral to it; job dissatisfaction, motivation and the study of work-related needs are studied to the degree they relate to or shed light on workplace injustice. Second, with its inclusion of contextual and situational variables, this study represents a contingency approach to organizational research: it recognizes that work behaviour is a complex result of many interacting forces, including individual, organizational and environmental influences (Lawrence & Lorsch, 1967). Also, unlike previous studies, this research relies on multiple case studies (i.e., multiple participants across multiple settings) to generate knowledge, rather than on single case studies or surveys. The central findings of this study are as follows: 174 1. a four-category typology of organizational injustice which includes interactional injustice (interpersonal mistreatment by a boss) as a distinct category 2. delineation and description of the behavioural dimensions of interactional injustice 3. identification of antecedents and consequences of workplace injustice for both individuals and organizations 4. emergent images (6) and themes (6) to characterize the phenomenon of organizational injustice from a synthesis of its individual- and organizational-level aspects 5. the concepts of voice (formal and informal) and silence as strategies to resist organizational injustice 6. a process model of voice and silence in organizational injustice B. The Phenomenon of Organizational Injustice A Typology of Organizational Injustice The emergence of conceptually distinct yet empirically correlated experiences of injustice reported by participants led to the development of a typology of organizational injustice. Cases were classified as reflecting perceived unfairness on the basis of individual interpersonal treatment (interactional injustice), outcomes (distributive injustice), procedures (procedural injustice) and/or the organization as a whole (systemic injustice). These categories were intended to be exhaustive of injustice experiences across both the interview and archival cases and were derived from theoretical interplay between the data, academic research (e.g., Bies & Moag, 1986; Sheppard et al., 1992) and popular accounts (Dumaine, 1993; Wright & Smye, 1996) of organizational injustice. 175 In addition, the finding in this study that distributive and procedural injustice tended to occur conjointly is consistent with findings from organizational justice research which support the integral connection between outcomes and procedures (e.g., Greenberg, 1986; Sheppard et a l , 1992). Together, these results and the incidence of other conjoint combinations (e.g., some cases reflected all four categories of injustice) indicate that the historic focus on procedural aspects of (injustice has neglected the role of additional influences on organizational injustice. Interactional injustice. Previous research (e.g., Bies & Moag, 1986) has focused on interactional justice — the perceived fairness of interpersonal treatment during the implementation of organizational procedures. However, my findings support recent but growing evidence which suggests that interactional mjustice is a distinct, substantive category of injustice perceptions, independent of its role in the implementation of procedures. I introduce a set of eight behavioural dimensions which emerged from inductive analyses of data which characterize interactional injustice: intimidation, abandonment, inconsistency, degradation, criticism, inaccessibility, surveillance and manipulation. The concept of interactional injustice as a distinct form of workplace injustice is consistent with popular accounts of bullying bosses (Dumaine, 1993), also described as 'sorcerers and tyrants' (Wright & Smye, 1996). As well, it is supported by Ashforth's (1994, 1997) studies of the 'petty tyrant,' defined as 'one who lords his or her power over others' in the workplace. Specifically, Ashforth's portrayal of tyrannical bosses reflect six behavioural dimensions: arbitrariness and self-aggrandizement, belittling others, lack of consideration, forcing conflict resolution, discouraging initiative and noncontingent punishment. Collectively, these dimensions appear conceptually consistent with those I introduce. As 176 well, the bosses described in my study as perpetrators of interactional injustice are consistent with what Ashforth (1997) calls infrequent petty tyrants who display 'full-blown pathology,' unlike the more common phenomenon of managers with tyrannical tendencies. Moreover, both Ashforth's petty tyrants and bosses described in my research as perpetrating interactional injustice suggest that their deep-seated assumptions and beliefs reflect McGregor's (1960) Theory X : that employees are lazy, selfish and stupid and prefer close supervision. In other research, Baron (1988) describes the concept of destructive criticism which is similar to the criticism dimension of interactional injustice in its effects (i.e., decreased self-efficacy, increased anger and stress) but different in its fundamental meaning. Whereas destructive criticism is defined as feedback that is general, inconsiderate and attributes employees' poor performance to internal causes, criticism in the context of organizational injustice emerged as specific and frequent; as well, the giver of feedback typically blames others (e.g., employees) while shirking personal responsibility for mistakes or problems. Consistent with Wright and Smye's (1996) accounts of the effects of bullying bosses on employees, my study found that when observers witness injustice toward an aggrieved party, they lose respect for that party. In addition, injustice perceptions are exacerbated when the organization appears to condone unfair treatment through inattention or inaction. Bosses' sense of entitlement to mistreat others (described by Wright and Smye and confirmed in this study) may derive, in part, from their interpretation of organizational inaction as support for his or her managerial approach. Essence of Injustice The results of this study are consistent with Folger's (1993) finding that the two primary reasons for employees' perceptions of mistreatment at work are: 1) not receiving 177 things (i.e., resources) such that employees perceive inequity and; 2) being treated as things such that employees do not feel respected as humans. Injustice perceptions are clearly associated with intentions to quit and subsequent turnover, suggesting that employees' expectations of fair treatment may act as a requirement for employment. Moreover, consistent with legal dictates that employers are responsible for the behavior of their employees (i.e., "vicarious liability") and that organizations are responsible for providing a workplace free from harassment, several participants in this study believed that their organizations had an obligation both to identify injustice and to redress it when informed of its occurrence. Their subsequent turnover (called 'principled turnover' or 'silent dissent' by Parker, 1997) indicates that employees' unmet expectations of respectful treatment reduces organizational reserves of human or intellectual capital and thus reduces, over the long-term, competitive edge and profitability. These results are also consistent with the importance of expectations in perceptions of injustice as reported by Rutte and Messick (1995). However, in this study some participants did not appear to have strong expectations of just treatment and, as a result, they did not quit. Some stayed because of social conditioning (i.e., abusive childhoods, cultural imperatives) against making complaints or causing conflicts. No evidence was found to support the equity sensitivity construct (Huseman et al., 1987) which suggests that those who perceive themselves as under-rewarded are 'benevolents' who prefer a lower equity ratio because they are givers rather than takers. On the contrary, the intense distress reported by many participants at their perceived inequity precluded their consideration as benevolents. Findings from other research provide additional insights into why some participants stayed despite workplace injustice. Research on female employees, for example, suggests 178 that perceptions of injustice are hard to maintain because cognitive bias encourages women to attribute discrimination to personal rather than systemic factors32 (Crosby, 1984). Other studies suggest that, to reduce chronic feelings of anger, some disadvantaged employees rationalize their discontent by adjusting their expectations downward to allow them to accept rather than change perceived inequities (Martin, 1986, 1993). In the present research, findings were consistent with both explanations. More broadly, some participants' attributions for staying indicated evidence of the fundamental attribution error (see Miller & Lawson, 1989) whereby they underestimated external causes (e.g., others' misconduct) and overestimated internal causes (e.g., engaged in self-blame). However, more research is needed to explain the puzzling phenomenon seen in this and other studies (e.g., Crosby, 1984) that those who have the most to gain are often least likely to speak up. Additional Patterns The finding reported in this study that the withholding of a reference letter was widely used as a threat in employment relationships perceived as unjust is consistent with a recent estimate that two-thirds of Canadian organizations provide little or no reference information about former employees ("Reference checks," 1997). Clearly, not providing reference information in itself does not an unjust organization make. However, this threat may be more prevalent in organizations which, knowingly or not, manifest additional types of injustice (e.g., interactional injustice, etc.). An unsolicited example provided by Irene supports, albeit indirectly, this possibility. She contrasted her experience of injustice at one organization with another in the same industry which she identified as an exemplar of justice toward and Because of difficulty inferring discrimination from individual cases and the limited availability of aggregate data, women often cited internal or personal faults (e.g., interpersonal style, work record) to reduce or negate their dissatisfaction with external outcomes (Crosby, 1984). 179 consideration for employees. This organization not only willingly provided reference information but also assisted in her job search by engaging in simulated hiring interviews to practice her job interview skills. My findings also suggest that seemingly small events, in fact, play a significant role in injustice perceptions. This emergent paradox is consistent with what Frost, Mitchell and Nord call 'the importance of the trivial,' described as "the contemporary state of any human system [such as organizations] that is the product of many small events" (1997: 519). Jennifer referred to it in her longing for "the little things that make you feel like a human being" in her emotionally barren workplace where managers valued productivity much more than people. Joanne echoed, "it was the subtle little things that hurt." In short, the incessant, relentless neglect of the importance of the trivial appears to be a characteristic feature of unjust employment relationships and appeared to heighten injustice perceptions. Antecedents to Organizational Injustice Ambiguous Structures And Procedures Findings concerning performance reviews are consistent with nearly forty years of reported deficiencies with performance appraisal procedures (e.g., Greenberg, 1986b; McGregor, 1957). In the present research, many of the procedural factors that are determinants of fair appraisals were absent. For example, opportunities for clarification or redress were typically unavailable, despite the presence of organizational voice systems designed to permit such exchange. Like Greenberg (1986b), this study found that employees were usually unaware of what criteria determined their appraisals and who was responsible for their completion. Contrary to McGregor's (1957) prescriptions for sound performance appraisal, findings from this study suggested that participants were not given the opportunity 180 to establish personal performance goals, there was little emphasis on analysis and much attention to questions of personality rather than performance. Similarly, few of Sashkin's (1981) guidelines for effective performance appraisals were in evidence in the present research. Collectively, these findings are consistent with extant research which suggests that, contrary to popular belief, accuracy is not the primary concern in appraisals (Longenecker, Sims & Gioia, 1987). Instead, as Longenecker et al. argue, political considerations appear to play a key role in the performance review process, indicated here by bosses' attempts to use the appraisal to their own advantage or for their own purposes (e.g., termination of employees on the basis of personality rather than performance). Findings from this study also indicate that the ambiguity of employee selection interviews was a source in itself of injustice perceptions, with deleterious effects on individuals and, by extension, organizational performance from unrealistic job previews and unclear job priorities. Thus, findings from this study support earlier research linking unrealistic job previews to turnover by precluding accurate assessment of the job demands -person style fit (Premack & Wanous, 1985). The present results are also consistent with those of Graves and Karren (1996) who report that unstructured, idiosyncratic employment interviews are linked to turnover, subsequent litigation and organizational ineffectiveness. Many of the practical recommendations which they cite (e.g., developing consensual selection criteria, providing training for interviewers) would likely reduce this source of injustice perceptions. In addition, results from the present study which indicate that role ambiguity is a structural antecedent to injustice are consistent with substantial evidence from role theory research which identifies role ambiguity as a significant source of job dissatisfaction and stress (e.g., Kahn, Wolfe, Quinn & Snoek, 1964; McGrath, 1978). 181 Together, procedural and structural antecedents to organizational injustice depict work settings which are rigid yet disordered with widespread ambiguity (e.g., unclear reporting structures, idiosyncratic decision-making, weak socialization processes). In particular, a compelling finding was the combination of ambiguous hierarchy of authority with high centralization; employees in such conditions cannot participate in decision-making and they experience confusion about their roles and to whom they report. This combination appeared to be a substantial source of stress and frustration. At this time, little research has been conducted against which to compare these findings. On this basis, then, these findings are accepted as preliminary. Given this ambiguity, it is perhaps not surprising that the role of politics was described in largely pejorative terms. Within such a 'lawless' environment, the role of politics was consistent with Pfeffer's (1981) conceptualization as activities to acquire, develop and use power to obtain one's preferred outcome when there is uncertainty or disagreement about choices. As evident here, however, politics was viewed in cynical terms (i.e., 'brownnosing,' 'taking the right people out to lunch') to describe efforts by employees for personal gain (i.e., getting a promotion) in such ambiguous environments. With such a pejorative connotation, politics appeared to contribute to injustice perceptions by reinforcing employees' beliefs that their use violated due process or resulted in the misallocation of resources. The pivotal role of structural and procedural ambiguity as a source for perceptions of injustice may be explained, in part, in terms of Weber's (1947) three bases of authority by which managers are provided control to make organizations function effectively. Of the three bases, rational-legal authority (i.e., employees' belief in the legality of rules and the inherent 182 right of those in authority to issue commands) is most relevant for this research.33 Findings from this study support Weber's argument that employees prefer rational-legal authority over other types of control, such as favoritism or politics (particularly as described above). However, across many of the organizations described here, widespread structural and procedural ambiguity precluded the development and subsequent use of rational-legal authority. That is, the rules by which those in command could exercise control were neither clear nor consensual, and more fundamentally, there was significant confusion about the identities of those in authority who were to issue commands. In the absence of clear rules and a hierarchy of authority, the opportunity for rational-legal authority was significantly limited. Indeed, in the organizations described there was strong evidence that the base of authority was not legitimate (i.e., not widely-held across organizational members as right and just). Moreover, injustice perceptions seemed exacerbated by organizations which claimed to rely on rational-legal authority, but in practice relied on other means. As a result, employees' expectations of the benefits of the rational-legal approach were not met. In sum, given the widely-accepted premise that organizational effectiveness rests on a coordinated and efficient authority structure (Pugh et al., 1968; Weber, 1947), we may expect that organizations whose injustice perceptions are linked to these procedural and structural antecedents will experience reduced effectiveness from reduced managerial control. Although Weber's (1947) rational-legal authority was identified for large organizations (i.e., bureaucracies), evidence from this research suggests, regardless of the size of their respective organizations, participants' beliefs in the legality of rules was widespread, thus providing support for the relevance of rational-legal authority for the smaller organizations described in this study. 183 Cultural Conditions Six cultural conditions, which resulted from comparisons of emergent patterns in the data against existing research on organizational culture (e.g., Hofstede et al., 1990; O'Reilly et al., 1991),were identified as sources of injustice perceptions. Together, they suggested the following configuration: tight control, suppressed conflict, the valuing of job relations rather than human relations, emphasis on results and production, and a competitive and self-oriented individualism appeared to be sources of injustice perceptions. In addition, low conflict tolerance emerged as a pivotal norm (Schein, 1980) whereby conformity was seemingly required by bosses and/or senior management; to violate this norm by publicly expressing dissent risked varying degrees of ostracism (e.g., 'being sent to corporate Siberia') at best or dismissal at worst. As well, recurrent themes of conformity, compliance and coldness were evident as sources of injustice perceptions. Taken together, these conditions and themes formed a composite picture of a culture of injustice, described here as a system of shared meaning held by participants in the context of organizational injustice. Little research exists against which to compare this portrayal of a culture of injustice. However, support for the concept is found in the three types of abusive organizational cultures described by Wright and Smye (1996). These are the culture of sacrifice, the win/lose culture and the culture of blame which, in varying ways parallel elements of the culture of injustice described here. For example, the culture of sacrifice demands conformity and compliance while the highly competitive win/lose culture rests on individualism and is described as cold and unfriendly. Further study is needed to examine the relative impact of the individual cultural conditions described here on injustice perceptions and their collective influence on the notion of a culture of injustice. 184 The means by which this cultural configuration emerged is, at present, unclear. However, existing research attests to norm formation around critical incidents and identification with leaders as two avenues for culture creation (Schein, 1990) which may explain, at least partly, how the cultural conditions reported here came to precede workplace injustice. Schein describes how the norm that 'authority is sacred' might develop which echoed several scenarios described in this study. For instance, one participant described her public disagreement with a new boss who then harshly criticized her in front of others. Schein suggests that it is in these pivotal moments that the norm against challenging the boss would begin to be created; thus, if co-workers concur with silence to the boss's counter-challenge and the employee apologizes, the foundation for the norm is set. The boss's counter-challenge need not be perceived as unjust, but if the pattern recurs employees learn that challenging authority is negative. In this way, the norm may first become a belief, then an assumption which, in turn, becomes a condition which fosters injustice: bosses who believe that their authority will not be challenged may then feel unrestrained, if not entitled, to mistreat employees. Schein (1990) also suggests that employees identify and internalize the values and assumptions of organizational leaders, especially founders or dominant figures who model to employees organizational beliefs and values. Thus, leaders' values and beliefs play a significant role in shaping emerging culture. In the present study, this mechanism may explain the impact of three owner-operators whose authoritarian, critical management style appeared consistent with Theory X (McGregor, 1960): these founders demonstrated low tolerance for conflict, a strong outcome orientation and low supportiveness for others, and were dominant forces in many areas of organizational functioning (i.e., setting and 185 monitoring policies and practices). By identifying with such leaders, participants and other employees in their organizations internalized Theory X beliefs that they (and others) were, in fact, stupid and lazy. Within such a belief system, a culture may develop in which injustice is believed necessary to keep such unproductive, untrustworthy staff in line. Given the poor selection and socialization procedures in many of the organizations described, evidence of weak 'person-culture fits' (O'Reilly et al., 1991), indicated by profound feelings of isolation and alienation among participants, is not surprising. My findings are consistent with O'Reilly et al.'s result that a strong person-culture fit is positively related to satisfaction and turnover and provide support for the notion that organizational cultures (in particular, those perceived as unjust) can have a strong impact on individuals. Consequences of Organizational Injustice Decreased self-esteem, performance, self-efficacy along with increased frustration, anxiety, isolation and helplessness among participants are some of the central consequences reported in this study. These findings, for the most part, are consistent with the few direct studies of unfair treatment at work. For example, Ashforth (1997) reports decreased work unit cohesiveness and increased stress and frustration as key effects of petty tyranny. Also, previous research (e.g., Jackson & Schuler, 1983) cites burnout from the accumulated stresses of overwork. The three stages of burnout — emotional exhaustion, cynicism and futility (Maslach, 1982) — are consistent with descriptions of burnout reported in the present research which resulted, in part, from organizational overwork. Several participants reported an early sign of burnout whereby employees work even harder but the overall quality of their performance deteriorates because of exhaustion (Maslach). 186 Furthermore, findings from this study strongly suggest that organizational injustice is a significant source of stress which merits much greater attention in continuing research on worksite stress management intervention (see Ivanevich, Matteson, Freedman & Phillips, 1990). For example, care must be taken to ensure that, ironically, voice systems as stress management interventions (i.e., employee assistance programs) do not become, in themselves, a situational stressor by perpetuating injustice. The present study departs from what Ivanevich et al. recognize as the traditional focus on individual interventions (e.g., teaching individual responsibility and coping strategies) and outcomes (e.g., physical and psychosomatic complaints) by including an organizational focus with attention to organizational outcomes (e.g., productivity, turnover) toward eliminating sources of workplace injustice. In sum, much remains to be done toward understanding both the antecedents to and effects of organizational injustice. For example, the interrelationships among and the relative impact of consequences remain unexplored. Given previous findings which suggest that self-efficacy acts as a moderating variable to determine whether job control positively or negatively influences work stressors (Schaubroeck & Merritt, 1997), one of many possible questions for further study is whether self-efficacy also acts a moderator for injustice-induced stress or other consequences. Similarly, within the context of workplace injustice the roles of cynicism and withdrawal in relation to frustration (i.e., whether they precede or result from this condition) require clarification in further research. Costs Individual. Long-term effects of workplace injustice on individuals included both detrimental and beneficial consequences (e.g., improved professional skills). However, the 187 turnover associated with injustice meant that benefits from employees' personal and professional development accrued to different organizations. For some, the experience of injustice was transformative; employees evolved from a state of self-described naivete to sophisticated wariness and conditional commitment to the organization (contingent on justice). However, that some participants were able to use their experience to their best advantage points to their resiliency in the face of adversity and is in no way intended to reduce organizational responsibility to abstain from misconduct. More broadly, my findings were consistent with what Argyris (1957) called the organizational dilemma, referring to the inherent conflict between the desire of individuals to maximize their psychological energy with the tendency of organizations to engender psychological failure through strategic and structural configurations which require dependence, submissiveness and which provide minimal challenge. In the present research, some organizations appeared to have incorporated structures, such as flatter hierarchies with team-based departments, to build in flexibility and responsiveness. Because these configurations are also intended to increase innovation and efficiency among employees, we would expect concomitant increases in employees' psychological energy. However, findings indicated that reduced psychological energy and, in some cases, psychological failure among individuals occurred despite such structural adaptations. Thus, the fundamental dilemma may not lie with conflicting organizational and individual needs but with conflicting organizational and individual perceptions of justice. In this way, potential benefits from organizational structures may be offset by authority figures who perpetrate interactional injustice, or by the retention of procedures perceived as unjust, for example. 188 Organizational. There is strong evidence from this research to suggest that organizational costs due to injustice are significant. As Jennifer said: In the light of reason, it (the injustice) makes no sense whatsoever. It's a waste of money. If an organization is trying to accomplish its goals and it's people who accomplish those goals, then any behaviour which does not foster, encourage and support its employees to accomplish its goals may as well pack up and go home. What advantage is it to keep training people, to not make them comfortable, to not sort things out and not make it a productive relationship? Empirical evidence suggests that injustice is associated with a multitude of negative effects, including productivity (in terms of work quality and quantity), client relationships and overall functioning (i.e., increased turnover, absenteeism). These findings are consistent with the theorized effects of stress and dissatisfaction on organizational functioning (e.g., Ivanevich et al., 1990). However, a systematic assessment of financial costs of workplace injustice across individual- and organizational-levels remains to be determined. In roughly 25% of cases reported here, additional substantive costs arose from litigation and other forms of third-party intervention to directly challenge workplace injustice or its effects. As Shaun said, "they fired me for their perception that they would lose money by having a union. So my feeling (in suing them) was, 'I will cost you every dollar I damn well can and it is going to cost you big.'" Societal. Jennifer, who felt emotionally unfit for work following the injustice, concisely captured the larger implications of organizational injustice as she contemplated whether social assistance (i.e., unemployment insurance, welfare) could provide adequate financial support until she felt fit enough to return to the workforce: The impact of this (the injustice) on me ripples out into society and through the social network, my personal life, throughout the rest of my future, and the health of our society. It is not just a narrow, isolated event. These have very profound ramifications (about) how our society functions and how we treat each other as human beings. 189 To varying degrees, organizational injustice results in the emotional and physical alienation of unfairly-treated employees. It also violates the fundamental sanctity of safety in the workplace and with it, the human spirit. Clearly, questions concerning societal effects of workplace injustice are of interest; for example, what are the implications of injustice experiences for support for or compliance with other bases of authority (e.g., governmental)? At present, detailed consideration of these broader questions is outside the parameters of the current study. Nonetheless, the issue of the societal implications of injustice at work represents an important area for further theoretical and empirical review. Incidence of Organizational Injustice: Present Estimates and Future Trends From the assumption commonly-made in the E V L N literature of the ubiquity of dissatisfaction and from my own and others' experiences, I expected a higher response rate (at least three-fold) than the 15% of U B C part-time business students who ultimately participated. The unexpectedly high percentage who reported 'no injustice' (49% of the U B C part-time graduate business student population) suggests that, as a more extreme form of dissatisfaction, the incidence of workplace injustice is significantly lower than that for general dissatisfaction. However, estimates of the incidence of workplace injustice may under-represent actual levels given the tendency described by Crosby (1984) and confirmed in this study that those who experience mistreatment tend to deny or minimize its occurrence and/or effects. It may also be that, paralleling Ashforth's (1997) findings on the incidence of petty tyrants, work environments with 'full-blown' injustice are much rarer than those exhibiting tendencies toward unfair treatment. In any case, future research which explicitly restricts participation to those who perceive injustice in particular is needed. 190 More broadly, the lack of comparable studies against which to compare this response rate and the absence of societal estimates of the incidence of organizational injustice makes it difficult to track current trends or to forecast future trends. One might suspect an increased incidence of workplace injustice in the future, given the catalytic role of the changing work environment. That is, current economic pressures from globalization and shrinking resources are among two influences affecting organizational structures and functioning, such that current organizational changes are unprecedented in their scope and the speed with which they are implemented. The 1990's economy demands that many organizations respond quickly and efficiently (see research on organizations in 'high velocity' or high turbulence environments; e.g., Brown & Eisenhardt, 1997; Eisenhardt & Bourgeois, 1988). In an era where restructuring is synonymous with downsizing and job loss, what Kathleen called the 'half-hour oil-patch process' of firing employees is notable because of its acceptance as common organizational practice. However, flatter organizational hierarchies and the a reliance on teamwork to coordinate labour do not preclude authoritarian managerial styles in project or team leaders, as this study confirms. As a result, Ashforth's (1997) prediction that the incidence of petty tyranny and, by extension other types of injustice, will decrease in future is questionable. On the contrary, such structural changes may continue to create conditions which are antecedent to organizational injustice. Together, the results of this study and an additional trend cited in the popular press support speculation that organizational injustice is on the rise. Specifically, across Canadian workplaces, organizationally-sanctioned overwork in the form of unpaid overtime is increasing (Beauchesne, 1997). Two reasons are cited for the increase: 1) downsizing and layoffs have reduced the workforce but not the amount of work; among remaining 191 employees, many believe they must stay until the job is done; 2) fear of being laid off. In turn, these reasons are supported by findings in my study, which adds a third: promises (implicit or explicit) of rewards (e.g., promotions). Thus, although expectancy theory (Vroom, 1964) might predict that employees' motivation to engage in unpaid overtime would decrease given the typically unclear performance - reward linkage (e.g., departmental ranking - salary increment) and effort - performance linkage (overtime - ranking), chronic and intense fear of being laid off may constitute a boundary condition. While organizational injustice may be increasing, tolerance for it may also be decreasing. This latter interpretation is consistent with Ashforth's (1997) speculation of the reduced tolerance for petty tyranny. However, even if the incidence of organizational injustice is steady, we may expect continued interest in and concern about organizational injustice, in part due to a natural extension of the increased societal awareness of childhood and domestic abuse to the workplace. Recall that several participants made direct links to their childhood experiences and/or social conditioning which predisposed them to accept injustice. Just as societal tolerance for childhood and domestic abuse is decreasing, so it may for injustice or abuse in the workplace. The quiet but firm insistence of many participants that their mistreatment was fundamentally unacceptable supports such speculation. C. Conceptualizations of Voice and Silence Voice This study suggests that voice encompasses two related but distinct constructs: formal voice and informal voice. Existing theory emphasizes the function of and positive consequences from voice within employment relationships perceived as just. This study 192 suggests a different (although arguably complementary) view of voice within the context of organizational injustice. For example, it suggests that formal voice may be defined as a collection of strategies to manage organizational injustice which involve the public acknowledgment of injustice to persons within or outside the organization who can effect remedial change. This study also suggests that informal voice strategies publicly acknowledge injustice but to persons who can provide support rather than effect remedial change. Thus, in this research voice is examined within the context of ongoing employment relationships perceived as unfair, rather than confined to the realm of managerial or organizational fairness. This study found evidence to support the dualistic use of formal voice by employees for instrumental and for value-expressive purposes. That is, consistent with the value-expressive view, the absence of opportunities for participants to be heard was a significant source of injustice perceptions (e.g., being prevented from questioning or commenting on performance appraisals). However, contrary to Sheppard et al., (1992) and Tyler (1987), but consistent with Cohen (1989) and Shapiro (1993), simply voicing discontent did not lead to perceptions of justice: participants' perceptions that their viewpoints were not seriously considered by or understood by the authority figure invariably led to injustice perceptions. In the present study, participants also used formal voice instrumentally to attempt to influence authority figures to secure more just outcomes or to create more just procedures (Leventhal, 1980; Thibault & Walker, 1975). Together, these findings support Shapiro's (1993) use of both value-expressive and instrumental roles of voice in models of procedural justice. 193 Silence As previously noted, the concept of silence has received little attention in organizational research. However, it has been studied in broad anthropological, sociological and linguistic terms (Bruneau, 1973; Jaworski, 1993; Tannen, 1985), from a phenomenological perspective (Ehrenhaus, 1988) and in sociocultural power relations (Scott, 1990). Across these disciplines, silence is widely accepted as a means of communication in itself, rather than the mere absence of speech or voice. Consistent with this position, my findings suggest that silence reflects a set of strategies to resist injustice which privately acknowledge injustice but nevertheless are intentional efforts to avoid conflict and to forgo formal voice. In fact, evidence suggests that silence belies a range of motives which, to varying degrees, reflect covert self-protection and resistance while communicating overt compliance. Thus, consistent with Molseed's (1989) research on gender, silence and group structure in relation to power, the present results suggest that silence is not simply the absence of voice but is a form of communication in itself. However, unlike Molseed (1989), no evidence was found to support gender differences in the use of silence. Similarly, contrary to evidence suggesting gender-based differences in communication styles (e.g., Tannen, 1990) and in the social conditioning of silence and powerlessness (Lips, 1994), the present study found no evidence of gender differences in voice and silence. However, preliminary evidence suggests that women may experience more guilt, shame and fear than men with the public acknowledgment of their injustice experiences, as indicated by the four interviews with female participants whose low volume made transcription difficult. 194 Contrary to previous research (e.g., Hirschman, 1970), my findings indicate that silence is not synonymous with loyalty such that loyal employees are not necessarily silent. Instead, this study suggests that loyalty and/or organizational commitment appear to have little direct impact on either voice or silence; they neither activate nor suppress these responses to organizational injustice. Rather, loyalty and commitment appear to influence the intensity of injustice perceptions: participants who appeared highly committed were distressed both by the negative consequences of injustice for the organization (e.g., reduced productivity, profligacy) and by those personal responses (e.g., acts of sabotage, reduced effort) which contradicted their own self-image as hard-working, responsible employees. An assumption tested in this research was that silence and powerlessness are indicated by the gap between participants' private (verbal and non-verbal behaviours kept from dominant others) and public transcripts (those behaviours expressed in the presence of dominant others). However, unexpected difficulties in assessing the gap using grounded theory precluded further consideration of this assumption. It may be that other qualitative methods of analysis (e.g., narrative and conversation analyses) which attend more closely to textual content may be more effective in assessing the gap. These approaches are increasingly used in organizational research (see Hatch, 1996 on narrative analyses and Tulin, 1997 on conversation analysis). More generally, support was found for Scott's (1990) general notion of private transcripts in some participants' creation and use of 'underground information networks;' these informal networks developed in response to workplace injustice support Scott's premise that when public disagreement is unsafe, objections or complaints are channeled into hidden, private and thus more safe domains. 195 Conceptual Connections: Relationships among Formal, Informal Voice and Silence One unexpected finding in this study was the difficulty in classifying participants' responses to organizational injustice as discrete categories of voice or silence. As previously noted, I had expected that prototypical profiles of participants' responses would emerge. The surprising result that participants chose a variety of voice-based and silence-based strategies to resist injustice speaks to the complex meaning of injustice experiences for participants. Although Table 8 distinguishes among the concepts of voice (formal and informal) and silence, the question of how these three major forms of resistance are related has not yet been addressed. To fully determine the patterns of conceptual relationships requires more elaborate theory than is currently possible given the available data. Nevertheless, it is possible at this point to speculate about the ways that these resistance strategies might be seen to interrelate as a preliminary step toward theory development. Several conceptualizations of the relationships among formal voice, informal voice and silence are possible. One approach is to view them as essentially independent or orthogonal. This approach is challenged, however, by the emergence of a common set of characteristics (e.g., acknowledgment of injustice) and commonly-held motives (e.g., resistance through self-protection and revenge) which linked the concepts. If related, questions concerning the nature and degree of their conceptual connectedness then arise. To what degree does conceptual overlap occur and what is the significance of points of shared meaning? One approach to conceptualization proposes substantive conceptual overlap, such as a Venn diagram of three overlapping circles or a continuum, for example. Although stronger evidence is needed to empirically support the notion of a voice-silence continuum, some findings from this study provide a preliminary 196 indication that voice and silence might be conceptualized as endpoints on a bipolar continuum of forms of resistance to perceived injustice. For example, the apparent exclusivity of formal voice and silence (i.e., that participants tended to adopt one or the other) is consistent with their conceptualization as endpoints on such a continuum. The use of a common continuum suggests that their relationship is one of conceptual complements rather than opposites: thus, Jaworski (1993) observes that silence does not preclude voice but rather keeps communication channels open to facilitate subsequent voice, if desired. The voice-silence continuum is also consistent with previous research in linguistics which views speech and silence in relativistic rather than absolute terms (see Jaworksi, 1993) and which uses a prototype approach (Rosch, 1978) rather than an essentialist approach to develop conceptual meanings. Emergent conceptualizations are well-suited to placement on a continuum where the ending of one concept and beginning of another cannot be precisely measured. In this application, informal voice would occupy a middle position on the continuum. Figure 1 illustrates how such a voice-silence continuum might be graphically represented. It is noted that assumptions of bipolarity and related explanations of organizational phenomena have been criticized for their inability to adequately reflect theory which captures the inconsistencies, paradoxes and contradictions which are so characteristic of human behaviour (Bobko, 1985). Such limitations can be addressed, in part, by rejecting a priori assumptions of bipolarity and by permitting evidence of bipolarity to emerge from the data. When warranted and valid, bipolar assumptions can enhance constructions of organizational phenomena (Bobko, 1985). As applied to the present data, a continuum such as that proposed here could capture the underlying variations by aggregating them to portray the set 197 FORMAL VOTCH INFORMAL VOTCF. SILENCE Acknowledgement of injustice and discontent Focus of response Impact of response Degree of desired change public external organizational maximal individual minimal private internal Figure 1. The Voice-Silence Continuum of Strategies to Resist Organizational Injustice. of strategies adopted by individuals, thereby addressing, at least in part, concerns that bipolar conceptualizations artificially and inaccurately restrict the complexity of data. In sum, there is insufficient evidence to support a voice-silence continuum. However, of the possible conceptualizations of relationships among voice (formal and informal) and silence, that which recognizes a two-way relationship between informal voice to each of formal voice and silence but which does not indicate conceptual connections between formal voice and silence appears to best fit the data. Because the data represent inherently partial versions of multiple realities of participants' experiences and because alternate modes of inquiry, theoretical foundations and analytic procedures using this data are expected to reveal equally valid (if divergent) interpretations, several conceptualizations of the interrelationships among formal voice, informal voice and silence are possible. Given the paucity of prior 198 conceptualizations of voice and silence in the context of organizational injustice, those briefly acknowledged here are recognized as rudimentary, requiring further detailed study. Fair Process and Frustration Effects As indicated in Chapter 1, previous research emphasizes organizational justice and the positive consequences of voice, namely the fair process effect by which employees feel greater satisfaction from voicing suggestions or complaints. However, within the context of injustice, strong evidence supports the potential for voice to engender both positive and negative reactions, namely the frustration effect by which employees feel greater frustration and dissatisfaction from voicing. Like those of Westerman and Rosse (1996), findings in this study suggest that the frustration effect is related to employees' perceptions that their respective organizations' commitment to voice is not genuine, illustrated by the sentiment expressed by many participants in this study that "there were voice systems in theory but not in practice." Moreover, my findings suggest that silence may also result in either fair process or frustration effects. However, of the responses described here, silence strategies appeared more likely to result in frustration effects, perhaps as an extension of the futility and hopelessness which prompted their use. Findings from this research support previous research which suggests that voice (Hirschman, 1970; Rothschild & Miethe, 1994), silence (Jaworski, 1993) and both voice and silence (Scott, 1990) are political responses. Participants' motives indicated that both responses were used as strategies to effect varying degrees of change, whether by directly or indirectly influencing superiors, and as strategies to resist perceived misconduct and unjust reprisals and thereby assert their dignity (see Rothschild & Miethe, 1994). Ironically, in some cases, participants' resistance of others' misconduct to assert their dignity sometimes 199 meant engaging in their own misconduct (e.g., sabotage). This result is also consistent with recent research which conceptualizes theft as an attempt by employees to redress distributive injustice through restitution and retaliation, which in turn is influenced by perceptions of interpersonal mistreatment (Greenberg & Scott, 1996). In the present study, the conceptualization of sabotage as an informal voice strategy to express discontent through resistance is consistent with research which suggests that sabotage is usually a rational act by which employees express a myriad of deep emotions (LaNuez & Jermier, 1994; Jermier, 1988). Voice System Efficacy This study revealed strong evidence that assumptions of voice system efficacy within contexts of organizational injustice are not warranted. Instead, findings suggest that such voice systems not only may be ineffective in resolving perceptions of injustice but, in fact, may contribute to or exacerbate injustice perceptions. Thus, findings from this study are consistent with Robinson's (1994) finding that the number of voice systems was p o s i t i v e l y related to employees' destructive behaviour. Further research is needed to examine why and how voice systems are perceived as ineffective. Like Robinson (1994) and Allen and Lucerno (1996), findings from this research indicate that the very presence of voice systems increases employees' expectations that some remedial action will by taken by the organization to resolve or redress injustice. Together, these results suggest that when such expectations are not met, employees become frustrated and possibly aggressive. Evidence also indicates that participants recognized and resented what they perceived as a political bias in many voice systems, seen as "designed fundamentally to preserve and protect the power of those who currently govern the 200 organization" (Sheppard et al., 1992: 154). This perception may partly explain participants' widespread cynicism and mistrust of organizational voice systems as reflecting the organization's sincere commitment to their welfare. In sum, voice system theory has largely ignored inductive analyses of users' experiences to develop knowledge or to inform practice. Despite their thoroughness, analyses of existing voice systems tend to focus on assessments of their efficacy by high-level management or by designers rather than users of voice systems (e.g., Ewing, 1989). Inclusion of both users' and designers' perspectives in assessing efficacy will no doubt increase accuracy and explore the role of voice systems in perpetuating workplace injustice. Implications for Managerial Practice and Organizational Functioning Despite their preliminary status, findings concerning voice system effectiveness point to several recommendations to reduce barriers and improve efficacy. First, it must be recognized that high opportunities for voice (indicated by larger numbers of voice systems) do not, themselves, confer perceptions of efficacy. In fact, if ineffective, higher opportunities for voice will likely increase employees' frustration due to raised but unmet expectations. On the basis of participants' assessments, it appears that organizations may benefit from fewer but more responsive voice systems. In particular, two systems appear potentially effective: an open door policy and regular employee-management meetings. However, to prevent frustration effects described in the study, these systems must be carefully and properly implemented. Effective implementation requires interpersonal skills training of managers or executives who are responsible for handling open-door complaints or for conducting meetings. Specifically, the development of active listening skills and interviewing or 201 counselling techniques which convey to others that they are being heard and understood are critical. As well, managerial training in dealing with others' emotional behaviour (Ostell, 1996) — particularly anger, anxiety and depression — would likely reduce the opportunity for authority figures to exacerbate existing perceptions of injustice or create new ones. Ongoing assessments by both users and non-users of the organization's implementation of these systems would identify barriers and provide directions for subsequent improvement. D. A Process Model of Voice, Silence and Organizational Injustice The theoretical framework presented in Figure 2 indicates a perceived violation of justice is typically precipitated by triggers, whether a series of adverse interactions, procedures or outcomes or by one triggering event (i.e., unjust dismissal). Both individual and organizational characteristics play causal roles in producing the precipitating event(s). For example, the degree to which the employee is demographically different from either his or her boss or from the workforce is important. Also, organizational attributes such as a disordered but autocratic milieu resulting from unclear hierarchy of authority and authoritarian management styles are antecedent to perceptions of injustice. Once employees have identified that an injustice has occurred, they respond with voice- and silence-based strategies to resist it. Individual factors likely play a role in employees' choices of resistance strategies. For instance, individuals from different cultural backgrounds and with different social and familial conditioning will respond in different ways. As well, employees' perceptions of the efficacy of voice systems determine, in part, their response; those who perceive them as ineffective will likely forgo formal voice and respond with informal voice and/or silence. 202 8 u s <U 3 o 1 CU cn C O U !2 .a « bX) c 1 S a o CL, c/T 3 o o cs a _o « N 'S u 'S > M S O g 3 n H < O. w • • • Voice ce Voice Voi C3 Forrrii Inforrr e CU -a u c s JO %-i "3 c o C a o 'J3 *° — a, 25 c 2 r- 3 a a jo _ w S b — § 1 8 B  5 M 2 P g ^ t i O u C ,3 — • w • W • • • • <u o 3 03 c ' - 4 — » aj N *3 aj T3 C CC3 tU O a ju cu" o o > cU -a o c<0 <D o o CU CN l-l 3 Pu 203 Both voice and silence have the potential to result in both frustration effects, leading to perceptions of injustice, and in fair process effects, leading to perceived resolution to the situation. Furthermore, the organization's response to formal voice itself may be a source of injustice perceptions by means of the frustration effect; alternately, the organization's response may lead to fair process effects. In this way, the model then repeats in a recursive loop such that different choices are available during subsequent iterations. Thus, an employee whose use of silence results in the frustration effect and the exacerbation of injustice perceptions may subsequently choose to use formal voice. More specifically, when the use of voice or silence does not resolve the situation, a number of outcomes are possible. At an individual-level, various feelings and conditions may occur: anxiety, frustration, stress, for example. In addition, work-related functional consequences may result. Organizational consequences include (but are not limited to) turnover, absenteeism and productivity. These consequences may result from the use of any one strategy of formal voice, informal voice and silence. E. Conclusions Contributions The central thesis of this dissertation is that organizational injustice is a complex, dynamic phenomenon which is resisted by employees with various voice-based and/or silence-based strategies. Through a process model developed from inductive analysis, the roles of voice (both formal and informal) and silence and their relation to the fair process and frustration effects are placed within a context of organizational injustice. Theoretically and methodologically, this study represents an alternative approach to investigating 204 organizational injustice and the roles of voice and silence with its qualitative inquiry in a field setting and with its in-depth, individual and processual focus on reactions to organizational injustice. It is hoped that the resulting cross-level model of voice, silence and organizational injustice will provide new directions for future qualitative or quantitative inquiries into how, when and why we choose among the varieties of voice- or silence-based strategies to resist workplace injustice and with what results. Limitations and Directions for Future Research This study represents the first in an emergent program of research on organizational injustice. Methodologically, it shares many of the limitations commonly observed in qualitative research: the reliance on self-report data, self-selection in recruitment and the relatively small number of participants which limit the application of emergent knowledge to all employees' experiences of organizational injustice. However, findings from this study are not intended as universal statements toward a grand theory of workplace injustice. Rather, they are viewed as theoretical formulations from in-depth empirical research which offer conceptual insights into the phenomenon of organizational injustice and related processes. Although the theoretical implications of this study may not apply to all employment relationships perceived as unjust, they nonetheless provide empirically grounded perspectives of employees' experiences of and responses to perceived injustice at work. In terms of organizational research, this study has not directly addressed the perspective of the organization or employer in the development and maintenance of injustice perceptions. Just as employers can act unjustly, so too can employees. In the absence of organizational misconduct, what are the manifestations and implications of employee-perpetrated injustice? How do organizations respond and with what results? Although it is 205 beyond the scope of the present research, how organizations and employers define and experience injustice is an issue of interest, nonetheless. One immediate direction for future research is to examine the role of voice and silence within two organizations in the same industry, one identified a priori as an exemplar of injustice and the other as an exemplar of justice, using an ethnographic case study approach involving both participant-observation and interviews. This would allow further clarification of the distinctive features of each phenomenon (organizational justice and organizational injustice) as well as conceptual refinements to voice and silence as strategies to resist injustice perceptions. More generally, additional research is needed to assess the utility of the process model proposed here and to assess the efficacy of voice systems by identifying both barriers to effective functioning and necessary improvements. Detailed survey research of both users and non-users of voice systems would help reveal functional limitations and explore more systematically the role of employees' expectations on their beliefs about and satisfaction with voice systems. In summary, this study introduces a grounded theory of the phenomenon of organizational injustice and responses to it. Emergent theory is connected to extant research; for example, Greenberg's (1990b) call for conceptual integration among the distributive, procedural and organizational justice literatures is extended by connecting this body of research with these historically disparate literatures. As well, this research informs practice by identifying barriers to effective organizational voice systems from assessments of their availability and efficacy. Future research and applications are required to assess the overall contribution of this work, judging its value, perhaps, against Herbert Simon's 206 criterion that "good theory is theory that works when it is applied to practical probl (Dessler, 1979). REFERENCES Adams, J. 1963. Toward an understanding of inequity. Journal of Abnormal and Social Psychology, 67: 422-436. Adams, J. 1965. Inequity in social exchange. In L . Berkowitz (Ed.), Advances in experimental psychology. N Y : Academic Press. A l l e n , R. E . & Lucerno, M . A . 1996. Beyond resentment: Exploring organizationally targeted insider murder. Journal of Management Inquiry, 5(2): 86-103. A l l e n , N . & Meyer, J. 1996. Affective, continuance and normative commitment to the organization : A n examination of construct validity. Journal of Vocational Behavior. 49(3): 252-276. Argyris , C . 1957. Personality and organization. N Y : Harper. Ashforth, B . 1994. Petty tyranny in organizations. Human Relations. 47: 755-778. Ashforth, B . 1997. Petty tyranny in organizations: A preliminary examination of antecedents and consequences. Canadian Journal of Administrative Sciences, 14: 126-140. Atkinson, P., Delamont, S. & Hammersley, M . 1988. Qualitative research traditions: A Brit ish response to Jacob. Review of Educational Research. 58:231-250. A v e r i l l , J. 1985. The social construction of emotion: With special reference to love. In K . Gergen (Ed.), The social construction of the person: 89-109. N Y : Springer-Verlag. Bandura, A . 1982. Self efficacy mechanisms in human agency. American Psychologist, 37: 122-147. Baron, R. 1988. Negative effects of destructive criticism: Impact on conflict, self-efficacy, and task performance. Journal of Appl ied Psychology. 73: 199-207. Barry, A . H . 1974. Economic models in political science: The case of "Exit , voice, and loyalty." British Journal of Political Science, 5, 69-82. Beauchesne, E . July 15, 1997. Unpaid overtime rampant, report says. The Vancouver Sun. A 3 . Bemmels, B . 1997. Exi t , voice and loyalty in employment relationships. In D . Lewin , J. Mi tche l l & M . Zaidi (Eds.), The human resource management handbook. Part II, 245-259. C T : J A I Press. Berkowitz, L . & Walster, E . (Eds.) 1976. Advances in experimental social psychology, 9. N Y : Academic Press. Bies, R. 1987. The predicament of injustice: The management of moral outrage. In L . L . Cummings & B . M . Staw (Eds.), Research in organizational behavior, 9: 289-319. C T : J A I Press. Bies, R. & Moag , J. 1986. Interactional justice: Communication criteria of fairness. In R. J. L e w i c k i , B . H . Sheppard & B . H . Bazerman (Eds.), Research on negotiation in organizations, 1:43-55. C T : J A I Press. 208 Bies, R. & Shapiro, D . 1988. Voice and justification: Their influence on procedural fairness judgments. Academy of Management Journal, 31: 676-685. Blau , P. 1964. Exchange and power in social life. N Y : Wi ley . Bobko, P. 1985. Removing assumptions of bipolarity: Towards variation and circularity. Academy of Management Review, 10: 99-108. Boroff, K . 1989. Loyalty - a correlate of exit, voice, or silence? In J. F . Burton (ed.), Industrial Relations Research Association Series: Proceedings of the Forty-Second Annual Meeting, 307-314. Madison, W I . Brockner, J. & Greenberg, J. 1989. The impact of layoffs on survivors: A n organizational justice perspective. In J. S. Carroll (Ed.), Appl ied psychology and organizational settings: 45-75. N J : Erlbaum. Brown, S. & Eisenhardt, K . 1997. The art of continuing change: L ink ing complexity theory and time-paced evolution in relentlessly shifting organizations. Administrative Science Quarterly, 42:1-34. Bruneau, T. 1973. Communicative silences: Forms and functions. The Journal of Communication, 23: 17-46. Buchanan, A . & Mathieu, D . 1986. Philosophy and justice. In R. L . CohentEd.), Justice: V iews from the social sciences: 11-45. N Y : Plenum Press. Butterfield, K . , Trevino, L . , & Ba l l , G . 1996. Punishment from the manager's perspective: A grounded investigation and inductive model. Academy of Management Journal, 39(6): 1479-1512. Cannon, M . 1997, May . The Howe Street mysteries. Report on Business: 38-49. Caproni, P. 1991. Gendered and hierarchical views of workplace aggression. (Doctoral dissertation, Yale University, 1991). Dissertation Abstracts International, 52(3-A): 994. Coffey, A . & Atkinson, P. 1996. Mak ing sense of qualitative data. C A : Sage. Cohen, J. 1960. Coefficient of agreement for nominal scales. Educational and Psychological Measurement, 20: 37-46. Cohen, R. L . 1985. Procedural justice and participation. Human Relations. 38: 643-663. Cohen, R . L . 1987. Distributive justice: Theory and research. Social Justice Research, 1: 19-40. Cohen, R. L . 1989. Fabrications of justice. Social Justice Research, 3: 31-46. Cohen, R. L . 1990. Justice, voice and silence. Paper presented at the International Conference on Social Science and Societal Dilemmas, Utrecht, The Netherlands. Cohen, R. L . 1991. Justice and negotiation. In R. J. Lewick i , B . H . Sheppard & B . H . Bazerman (Eds.), Research on negotiation in organizations, 3: 259-282. C T : J A I Press. 209 Cohen, M . , March , J. & Olsen, J. 1972. A garbage can model of organizational choice. Administrative Science Quarterly. 17: 1-25. Cropanzano, R. & Randall, M . 1993. Injustice and work behavior: A historical review. In R. Cropanzano (Ed.), Justice in the workplace: 3-20. N J : Erlbaum. Crosby, F. 1984. The denial of personal discrimination. American Behavioral Scientist. 2: 371-386. Davidson, M . & Earnshaw, J. (Eds.) 1991. Vulnerable workers: Psychosocial and legal issues. England: John Wi ley & Sons. Denzin, N . 1996. Interpretive ethnography. London: Sage. Denzin, N . & Lincoln , Y . (Eds.) 1994. The handbook of qualitative research: C A : Sage Dexter, L . 1970. Elite and specialized interviewing. I l l : Northwestern University Press. Dessler, G . 1979. Organization and management. N J : Prentice-Hall. Dumaine, B . 1993, October 18. America's toughest bosses. Fortune. 39-49. Ehrenhaus, P. 1988. Silence and symbolic expression. Communication Monographs, 55: 41-57 Eisenhardt, K . 1989. Bui lding theories from case study research. Academy of Management Review. 14: 532-550. Eisenhardt, K . & Bourgeois, L . 1988. Politics of strategic decision making in high velocity environments: Toward a mid-range theory. Academy of Management Journal. 31: 737-770. Elsbach, K . 1994. Managing organizational legitimacy in the California cattle industry: The construction and effectiveness of verbal accounts. Administrative Science Quarterly. 39: 57-88. Emerson, R. 1962. Power-dependence relations. American Sociology Review, 27: 31-34. Emery, F . & Trist, E . The causal texture of organizational environments, Human Relations, 18: 21-32. Ewing , D . 1989. Justice on the job. M A : Harvard Business School Press. Farrell, D . 1983. Exi t , voice, loyalty and neglect as responses to job dissatisfaction: A multidimensional scaling study. Academy of Management Journal. 26: 596-607. Farrell, D . & Rusbult, C . 1992. Exploring the exit, voice loyalty and neglect typology: The influence of job satisfaction, quality of alternatives and investment size. Employee Rights and Responsibilities Journal. 5: 201-218. Feuille, P. & Delaney, J. 1992. The individual pursuit of organizational justice: Grievance procedures in nonunion workplaces. In G . R. Ferris & K . Rowland (Eds.), Research in personnel and human resource management, 10: 187-232. C T : J A I Press. Fine, M . 1983. The social context and a sense of injustice: The option to challenge. Representative Research in Social Psychology. 13: 15-33. 210 Fisher, L . & V a n Belle, G . 1993. Biostatistics: A methodology for the health sciences. N Y : John Wi ley . Fleiss, J. 1981. Statistical methods for rates and proportions (2nd ed.). N Y : John Wi ley . Folger, R. 1977. Distributive justice and procedural justice: Combined impact of 'voice ' and improvement on experienced inequity. Journal of Personality and Social Psychology, 35: 108-119. Folger, R. 1986. Rethinking equity theory: A referent cognitions model. In H . W . Bierhoff, R. L . Cohen & J . Greenberg (Eds.), Justice in social relations, 145-162. N Y : Plenum Press. Folger, R. 1993. Reactions to mistreatment at work. In J. Murnighan (Ed.), Social Psychology in Organizations (pp. 161-183). N J : Prentice-Hall. Folger, R. 1995. A popcorn model of workplace violence. Paper presented at the 55th Annual Meeting of the Academy of Management, Vancouver, B C . Folger, R. & Bies, R. 1989. Management responsibilities and procedural justice. Employee Rights and Responsibilities Journal, 2: 79-90. Folger, R. & Konovsky, M . A . 1989. Effects of procedural and distributive justice on reactions to pay raise decisions. Academy of Management Journal, 32: 115-130. Folger, R., Rosenfield, D . , Grove, J. & Corkran, L . 1979. Effects of 'voice' and peer opinions on responses to inequity. Journal of Personality and Social Psychology, 37: 2253-2261. Folger, R., Rosenfield, D . , & Robinson, T. 1983. Relative deprivation and procedural justification. Journal of Personality and Social Psychology, 45: 268-273. Frost, P., Mi tche l l , V . & Nord. W . 1997. Organizational reality: Reports from the firing line. (4th ed.). M A : Addison-Wesley. Freeman, R. & Medoff, J. 1984. What do unions do? N Y : Basic Books. Galbraith, J . 1977. Organization design. M A : Addison-Wesley. Gersick, C . 1988. Time and transition in work teams: Toward a new model of group development. Academy of Management Journal, 31: 9 - 41. Glaser, W . & Strauss, A . 1967. The discovery of grounded theory: Strategies of qualitative research. London: Wiedenfeld and Nicholson. Gorden, W . 1988. Range of employee voice. Employee Responsibilities and Rights Journal, 1: 283-299. Gorden, W . , Infante, D . A . , & Graham, E . E . 1988. Corporate conditions conducive to employee voice. Employee Responsibilities and Rights Journal, 1: 101-111. Graham, J. 1986. Principled organizational dissent: A theoretical essay. In Staw, B . M . & Cummings, L . L . (Eds.), Research in Organizational Behavior. 8: 1-52. C T : J A I Press. 211 Graham J. & Keeley, M . 1992. Hirschman's loyalty construct. Employee Responsibilities and Rights Journal. 5: 191-200. Graves, L . & Karren, R. 1996. The employee selection interview: A fresh look at an old problem. Human Resource Management, 35: 163-180. Greenberg, J. 1986a. Determinants of perceived fairness of performance evaluations. Journal of Appl ied Psychology. 71: 340-342. Greenberg, J. 1986b. Organizational performance appraisal procedures: What makes them fair? In R. J. L e w i c k i , B . H . Sheppard & B . H . Bazerman (Eds.), Research on negotiation in organizations. 1:25-41. C T : J A I Press. Greenberg, J . 1987. A taxonomy of organizational justice theories. Academy of Management Review. 12: 9-22. Greenberg, J. 1988. Equity and workplace status: A field experiment. Journal of Appl ied Psychology. 73: 606-613. Greenberg, J. 1989. Cognitive re-evaluation of outcomes in response to underpayment inequity. Academy of Management Journal. 32: 174-184. Greenberg, J. 1990a. Looking fair vs. being fair: Managing impressions of organizational justice. In B . M . Staw & L . L . Cummings (Eds.), Research in organizational behaviour, 12: 111-157. C T : J A I Press. Greenberg, J. 1990b. Organizational justice: Yesterday, today and tomorrow. Journal of Management, 16: 399-432. Greenberg, J. 1992, August. Concept development in organizational behavior: The case of procedural justice. Paper presented at the Meeting of the Academy of Management, Las Vegas, Nevada. Greenberg, J. 1994. Using socially fair treatment to promote acceptance of a work site smoking ban. Journal of Appl ied Psychology, 79: 288-297. Greenberg, J. & Folger, R. 1983. Procedural justice, participation and the fair process effect in groups and organizations. In P. Paulus (Ed.), Basic group processes: 235-256. N Y : Springer-Verlag. Greenberg, J. & Ornstein, S. 1983. High status job title as compensation for underpayment: A test of equity theory. Journal of Applied Psychology. 68: 285-297. Greenberg, J. & Scott, K . Why do workers bite the hand that feeds them? Employee theft as a social exchange process. In B . Staw and L . Cummings (Eds.), Research in Organizational Behavior. 18: 111-156. C T : J A I Press. Guba, E . 1990. The alternative paradigm dialog. In E . Guba (Ed.), The paradigm dialog: 17-27. C A : Sage. Guba, E . & Lincoln , Y . 1981. Effective evaluation. C A : Jossey-Bass. Hammersley, M . 1992. What's wrong with ethnography? London: Routledge. 212 Hammersley, M . & Atkinson, P. 1995. Ethnography: Principles in practice (2nd ed.). London: Routledge. Harlos, K . P. 1995, August. Beauty and the beast: Radical feminist power in organizations. Paper presented at the 55th Annual Meeting of the Academy of Management, Vancouver, B C . Hatch, M . 1996. The role of the researcher: A n analysis of narrative position in organization theory. Journal of Management Inquiry, 5: 359-374. Hirschman, A O. 1970. Exi t , Voice and Loyalty. Cambridge, Massachusetts: Harvard University Press. Homans, G . 1961. Social behaviour: Its elementary forms. N Y : Harcourt, Brace & Wor ld . House, R., Rousseau, D . & Thomas-Hunt, M . 1995. The meso paradigm: A framework for the integration of micro and macro organizational behaviour. In B . Staw and L . Cummings (Eds.), Research in Organizational Behavior. 17: 71-114. C T : J A I Press. Hunt, J. & Ropo. A . 1995. Mult i - level leadership: Grounded theory and mainstream theory applied to the case of General Motors. Leadership Quarterly, 6(3): 379-412. Huseman, R., Hatfield, J. & Mi les , E . 1987. A new perspective on equity theory: The equity sensitivity construct. Academy of Management Review, 12: 222-234. lies, P. & Aul ick , R. 1991. The experience of black workers. In M . Davidson & J. Earnshaw (Eds.), Vulnerable workers: Psychosocial and legal issues: 297-322. London: John Wi ley & Sons. Ivanevich, J. , Matteson, M . , Freedman, S. & Phillips, J. 1990. Worksite stress management interventions. American Psychologist, 45: 252-261. Jackson, S. & Schuler, R. 1983. Preventing employee burnout. Personnel, 60: 58-68. Jacob, E . 1987. Qualitative research traditions: A review. Review of Educational Research, 5 1 : 1 -50. Jaworski, A . 1993. The power of silence: Social and pragmatic perspectives. C A : Sage. Jermier, J. 1988. Sabotage at work: The rational view. Research in the Sociology of Organizations, 6: 101-134. Jick, T. M i x i n g qualitative and quantitative methods: Triangulation in action. Administrative Science Quarterly, 24: 602-611. Kahn, W . A . 1993. Caring for the caregivers: Patterns of organizational caregiving. Administrative Science Quarterly. 38: 539-563. K a h n , R . , W o l f e , D . , Q u i n n , R . & Snoek, D . 1964. Organizat ional stress: Studies i n role conf l ic t and ambiguity. N Y : John W i l e y . Kanter, R. 1977. M e n and women of the corporation. N Y : Basic Books. Keeley, M . & Graham, J. 1991. Exit , voice and ethics. Journal of Business Ethics. 10: 349-355 213 K i r k , J. & Mi l l e r , M . 1986. Reliability and validity in qualitative research. C A : Sage. Kitzinger, C . 1991. Lesbians and gay men in the workplace. In M . Davidson & J. Eamshaw (Eds.), Vulnerable workers: Psychosocial and legal issues: 223-240. London: John Wi ley & Sons. Kolarska, L . & Aldr ich , H . 1980. Exit , voice and silence: Consumers' and managers' responses to organizational decline. Organization Studies, 1:41-58. Konovsky, M . A . & Cropanzano, R. 1991. Perceived fairness of employee drug testing as a predictor of employee attitudes and job performance. Journal of Appl ied Psychology, 76: 698-707. Kossek, E . & Nicho l , V . 1992. The effects of on-site child care on employee attitudes and performance. Personnel Psychology. 45: 435-509. LaNuez, D . & Jemier, J. 1994. Sabotage by managers and technocrats. In J. Jermier, D . Knights and W . Nord (Eds.), Resistance and power in organizations, 219-251. London: Routledge. LaTour, S. 1978. Determinants of participant and observer satisfaction with adversary and inquisitorial modes and adjudication. Journal of Personality and Social Psychology, 36: 1531-1545. Lawler, E . 1968. Equity theory as a predictor of productivity and work quality. Psychological Bullet in. 70: 596-610. Lawrence, P . & Lorsch, J. 1967. Organization and environment. IL : Irwin. Leek, J. & Saunders, D . 1992. Hirschman's loyalty: Attitude or behavior. Employee Rights and Responsibilities Journal, 5: 219-229. Leventhal, G . S. 1980. What should be done with equity theory? New approaches to the study of fairness in social relationships. In K . Gergen, M . Greenberg, & R. W i l l i s Eds., Social exchange: Advances in theory and research: 27-55. N Y : Plenum Press. Lewin , D . & Boroff, K . 1994. The role of loyalty in exit and voice: A conceptual and empirical analysis. Paper presented at the 4th Bargaining Group Conference, University of Toronto. Lewis , D . 1990. Self-appraisal versus traditional performance appraisal: Perceptions of procedural justice. Unpublished doctoral dissertation, Tulane University. L inco ln , Y . & Guba, E . 1985. Naturalistic inquiry. C A : Sage. L i n d , E . A . & Tyler, T. R. 1988. The social psychology of procedural justice. N Y : Plenum. Lips , H . 1994. Female powerlessness: A case of 'cultural preparedness'? In H . L . Radtke & H . J. Stam (Eds.). Power/gender: Social relations in theory and practice: 89-107. London: Sage. Longenecker, C , Sims, H . , & Gioia , D . 1987. Behind the mask: The politics of executive appraisal. Academy of Management Executive, 1: 183-193. Macaulay, J. 1985. Adding gender to aggression research: Incremental or revolutionary change? In V . O'Leary, R. K . Unger & B . S. Wallston (Eds.), Women, gender and social psychology: 191-224. N J : Lawrence Erlbaum. 214 Major, B . 1987. Gender, justice and the psychology of entitlement. In P. Shaver & C . Hendrick (Eds.), Review of personality and social psychology: Sex and gender. 7: 124-148. Maneiro, L . 1986. Coping with powerlessness: The relationship of gender and job dependency to empowerment-strategy usage. Administrative Science Quarterly. 31: 633-653. Marshall , C & Rossman, G . 1995. Designing qualitative research. 2nd ed. C A : Sage. Mart in , J. 1986. When expectations and justice do not coincide: Blue-collar visions of a just world. In H . W . Bierhoff & R. L Cohen (Eds.), Justice in Social Relations. 317-335. N Y : Plenum. Mart in, J. 1993. Inequality, distributive justice, and organizational illegitimacy. In J. K . Murnighan (Ed.), Social Psychology in Organizations: Advances in Theory and Research: 296-321. N J : Prentice Ha l l . Mart in , P. Y . & Turner, B . A . 1986. Grounded theory and organizational research. Journal of Appl ied Behavioural Science, 22 (2): 141-157. Maslach, C . 1982. Burnout: The cost of caring. N J : Prentice-Hall. Mayes, B . & Ganster, D . 1988. Exit and voice: A test of hypotheses based on fight/flight responses to job stress. Journal of Organizational Behavior, 9, 199-216. McCracken, G . 1988. The long interview. C A : Sage. M c G r a t h , J . 1976. Stress and behaviour i n organizations. In M . Dunnette (Ed.) , H a n d b o o k o f industr ial and organizational psychology, 1351-1398. I l l : R a n d M c N a l l y . McGregor , D . 1957. A n uneasy look at performance appraisal. Harvard Business Review, 35: 89-94. McGregor , D . 1960. The human side of enterprise. New York: M c G r a w - H i l l . Meyer, J . P., A l l e n , N . J. & Smith, C . A . 1993. Commitment to organizations: Extension and test of a three-component conceptualization. Journal of Appl ied Psychology, 78: 538-551. M i c e l i , M . P. & Near, J. P. 1984. The relationships among beliefs, organizational position and whistle-blowing status: A discriminant analysis. Academy of Management Journal, 27: 687-705. M i c e l i , M . P. & Near, J. P. 1985. Characteristics of organizational climate and perceived wrongdoing associated with whistle-blowing decisions. Personnel Psychology, 38: 525-544. M i c e l i , M . P. & Near, J. P. 1989. The incidence of wrongdoing, whistle-blowing and retaliation: Results of a naturally-occurring field experiment. Employee Responsibilities and Rights Journal. 2:91-108. Mi le s , E . , Hatfield, J. & Huseman, R. 1989. The equity sensitivity construct: Potential implications for worker performance. Journal of Management. 15, 581-588. Mi le s , E . , Hatfield, J. & Huseman, R. 1994. Equity sensitivity and outcome importance. Journal of Organizational Behaviour, 15, 585-596. 215 Miles , M . & Huberman, A . 1994. Qualitative data analysis. 2nd ed. C A : Sage. M i l l , C . R. 1976. Feedback : The art of giving and receiving help. In L . Porter & C . M i l l (Eds.), The reading book for human relations training: 18-19. M E : National Institute for Appl ied Behavioural Science. M i l l e r , A . & Lawson, T. 1989. The effects of an informational option on the fundamental attribution error. Personality and Social Psychology Bulletin. 194-204. Mishler , E . G . 1986. Research interviewing: Context and narrative. M A : Cambridge University Press. Molseed, M . 1989. Gender, silence and group structure. Studies in Symbolic Interaction. 10: 145-165. Morse, J. 1994. Designing funded qualitative research. In N . Denzin & Y . L inco ln (Eds.), The handbook of qualitative research: 220-235. C A : Sage. Morton-Wil l iams, J. 1993. Interviewer approaches. Cambridge: University Press. Mowday, R. 1987. Equity theory predictions of behaviors in organizations. In R. Steers and L . Porter (Eds.), Motivation and work behavior. (4th ed.): 89-110. N Y : M c G r a w - H i l l . Mowday , R., Steers, R. & Porter, L . 1979. The measurement of organizational commitment. Journal of Vocational Behavior. 14: 224-247. Near, J. P. , Dworkin , T. M . & M i c e l i , M . P. 1993. Explaining the whistle-blowing process: Suggestions from power theory and justice theory. Organization Science. 4: 393-411. Near, J . & M i c e l i , M . 1987. Whistle-blowers in organizations: Dissidents or reformers? In B . Staw and L . Cummings, Research in Organizational Behavior. 9: 321-368. C T : J A I Press. Nisbett, R. & Wilson , T. 1977. Tell ing more than we can know: Verbal reports on mental processes. Psychological Review. 84: 231-259. Olson-Buchanan, J. 1996. Voic ing discontent: What happens to the grievance filer after the grievance? Journal of Appl ied Psychology. 81: 52-63. O'Rei l ly , C , Chatman, J. & Caldwell , D . 1991. People and organizational culture: A profile comparison approach to assessing person-organization fit. Academy of Management Journal.. 34: 487-516. Ostell, A . 1996. Managing dysfunctional emotions in organizations. Journal of Management Studies. 33: 525-557. Parker, L . 1993. Control, self-efficacy, dissent and exit. Journal of Appl ied Psychology. 78: 949-959. Parker, L . 1997. Silent dissenters: A model for exploring the sources and consequences of principled turnover. Paper presented at the 57th Annual Meeting of the Academy of Management, Boston, M A . Patton, M . 1990. Qualitative research and evaluation methods (2nd ed.) C A : Sage. 216 Pfeffer, J. 1981. Power in organizations. Cambridge: Ballinger. Phil l ips, D . 1990. Postpositivistic science: Myths and realities. In E . Guba (Ed.), The paradigm dialog: 31-45. C A : Sage. Pinder, C . 1998. Work motivation in organizational behaviour. N Y : Prentice-Hall. Podsakoff, P. & Organ, D . 1986. Self-reports in organizational research: Problems and prospects. Journal of Management. 12: 531-544. Prasad, P. 1993. Symbolic processes in the implementation of technological change: A symbolic interactionalist study of work computerization. Academy of Management Journal. 36: 1400-1429. Premack, S. & Wanous, J. 1985. A meta-analysis of realistic job preview experiments. Journal of Appl ied Psychology. 70: 706-710. Pugh, D . , Hickson, D . , Hinings, C . & Turner, C . 1968. Dimensions of organizational structure. Administrative Science Quarterly. 13: 65-105. Rawls, J . 1971. A theory of justice. M A : Harvard University Press. Reference checks. July 1, 1997. The Globe and M a i l . B 8 . Reichers, A . Wanous, J. & Austin, J. 1997. Understanding and managing cynicism about organizational change. Academy of Management Executive, 11: 48-59. Rich , P. 1992. The organizational taxonomy: Definition and design. Academy of Management Review. 17: 758-781. Robinson, S. 1993. Retreat, voice, silence and destruction: A typology of employees' behavioural responses to dissatisfaction. Working paper, N Y University. Robinson, S. 1994. Explaining retreat, voice, silence and destruction: The impact of work context on employees' responses to dissatisfaction. Working paper, N Y University. Roehl , J. A . 1988. Measuring perceptions of procedural justice. Unpublished doctoral dissertation, George Washington University. Rosch, E . 1978. Principles of categorization. In E . Rosch & B . L l o y d (Eds.), Cognition and categorization, 27-48. N J : Erlbaum. Rosenthal, R. & Jacobson, L . 1968. Pygmalion in the classroom. N Y : Rinehart & Winston. Rosse, J . & Hul in , C . 1985. Adaptation to work: A n analysis of employee health, withdrawal and change. Organizational Behavior and Human Decision Processes, 36: 324 - 347. Rothschild, J. & Miethe, T. 1994. Whistle-blowing as resistance in modern work organizations: The politics of revealing organizational deception and abuse. In J. Jermier, D . Knights and W . Nord (Eds.), Resistance and power in organizations, 252-273. London: Routledge. Rousseau, D . & House, R. 1994. Meso organizational behavior. Journal of Organizational Behavior. 15 (Suppl. 1): 13-30. 217 Runciman, W . 1966. Relative deprivation and social justice: A study of attitudes to social inequality in twentieth century England. Berkeley, C A : University of California Press. Rusbult, C , Farrell, D . , Rogers, G . , & Mainous, A . 1988. Impact of exchange variables on exit, voice, loyalty and neglect: A n integrative model of responses to declining job satisfaction. Academy of Management Journal. 31: 599-627. Rusbult, C . & Lowery, D . 1985. When bureaucrats get the blues: Responses to dissatisfaction among federal employees. Journal of Appl ied Social Psychology, 68: 429-438. Rusbult, C , Zembrodt, I. & Gunn, L . 1982. Exit , voice, loyalty and neglect: Responses to dissatisfaction in romantic involvements. Journal of Personality and Social Psychology. 43: 1230-1242. Rutte, C . & Messick, D . 1995. A n integrated model of perceived unfairness in organizations. Social Justice Research, 8: 239-261. Sashkin, M . 1981. Appraising appraisal: Ten lessons from research to practice. Organizational Dynamics. 37-50. Saunders, D . , Sheppard, B . , Knight, V . & Roth, J. 1992. Employee voice to supervisors. Employee Rights and Responsibilities Journal. 5, 241-259. Schaubroeck, J. , & M a y , D . R., & Brown, F . W . 1994. Procedural justice explanations and employee reactions to economic hardship: A field experiment. Journal of Appl ied Psychology. 79: 455-460. Schein, E . 1980. Organizational psychology (2nd ed.). N J : Prentice-Hall. Schein, E . 1990. Organizational culture. American Psychologist, 45: 109-119. Schaubroeck, J. & Merritt, D . 1997. Divergent effects of job control on coping with work stressors: The key role of self-efficacy. Academy of Management Journal, 40: 738-754. Schwab, D . P . 1980. Construct validity in organizational behaviour. In B . Staw & L . Cummings (Eds.), Research in organizational behaviour, 2: 3-43. C T : J A I Press. Scott, W . 1975. A n overview of the field. In H . Tosi (Ed.), Theories of Organizations: 10-28. N Y : John Wi ley & Sons. Scott, J. 1990. Domination and the arts of resistance: Hidden transcripts. C T : Yale University Press. Shapiro, D . 1993. Reconcil ing theoretical differences among procedural justice researchers by re-evaluating what it means to have one's views "considered": Implications for third-party managers. In R. Cropanzano (Ed.), Justice in the Workplace: 51-78. N J : Lawrence Erlbaum Associates, Inc. Sheppard, B . H . & L e w i c k i , R. J. 1987. Toward general principles of managerial fairness. Social Justice Research. 1: 161-176. Sheppard, B . H . , L e w i c k i , R. J., & Minton, J. W . 1992. Organizational justice: The search for fairness in the workplace. Toronto: Lexington Books. 218 Singer, M . 1990. Determinants of perceived fairness in selection practices: A n organizational justice perspective. Genetic, Social, and General Psychology Monographs. 11.6: 477-494 Spencer, D . 1986. Employee voice and employee retention. Academy of Management Journal. 29: 488-502. Staw, B . 1984. Organizational behavior: A review and reformulation of the field's outcome variables. In M . Rosenzweig and L . Porter (Eds.), The Annual Review of Psychology, 35: 627-666. Stone, T. & Mel tz , N . 1993. Human resource management in Canada (3rd ed.). Holt , Rinehart & Winston: O N . Stouffer, S., Suchman, E . , DeVinney, L . , Star, S., & Wil l iams, R. 1949. The American soldier: Adjustment during A r m y life (vol. 1). N J : Princeton University Press. Strauss, A . 1987. Qualitative analysis for social scientists. N Y : Cambridge University Press. Strauss, A . & Corbin, J. 1994. Grounded theory methodology. In N . Denzin & Y . L inco ln (Eds.), The handbook of qualitative research: 273—285. C A : Sage. Strauss, A . & Corbin, J. 1990. Basics of qualitative research: Grounded theory procedures and techniques. C A : Sage. Swann, W . 1983. Self-verification: Bringing social reality into harmony with the self. In J. Suls and A . Greenwald (Eds.), Psychological perspectives on the self, 2: 33-66. N J : Erlbaum. Tannen, D . 1990. Y o u just don't understand: Women and men in conversation. N Y : Ballantine Books. Tannen, D . & Saville-Troike, M . (Eds.) 1985. Perspectives on silence. N J : Ablex Publishing Corp. Thibault, J. & Walker, L . 1975. Procedural justice: A psychological analysis. N J : Erlbaum. Tul in , M 1997. Talking organization: Possibilities for conversation analysis in organization behaviour research. Journal of Management Inquiry, 6: 101-119. Turner, B . A . 1983. The use of grounded theory for the qualitative analysis of organizational behaviour. Journal of Management Studies, 20 (3): 333-348. Tyler, T. R. 1989. The psychology of procedural justice: A test of the group-value model. Journal of Personality and Social Psychology. 57: 830-838. Tyler, T. R. 1990. W h y people obey the law. C T : Yale University Press. Tyler, T. R. 1991. Using procedures to justify outcomes: testing the viability of a procedural justice strategy for managing conflict and allocating resources in work organizations. Basic and Appl ied Social Psychology. 12: 259-279. Tyler, T. R. & Bies, R. 1990. Beyond formal procedures: The interpersonal context of procedural justice. In J. S. Carroll (Ed.), Applied psychology and organizational settings: 77-98. N J : Erlbaum. 219 Vroom, V . 1964. Work and motivation. N Y : John Wiley & Sons. Walster, E . , Walster, G . & Berscheid, E . 1978. Equity: Theory and research. M A : A l l y n & Bacon. Wanous, J. , Reichers, A . & Austin, J. 1994. Organizational cynicism: A n initial study. In D . Moore (Ed.), Best Paper Proceedings of the 54th Annual Meeting of the Academy of Management, 269-273. Weber, M . 1947. The Theory of Social and Economic Organizations (A . Henderson & T. Parsons, Trans.). N Y : Free Press. Weick , K . 1979. The social psychology of organizations (2nd ed.). N Y : M c G r a w - H i l l . Westerman, J. & Rosse, J. 1996, March. Subordinate appraisals: Fertile ground for the frustration effect. Paper presented at the Western Academy of Management Conference, Banff, Alberta. Withey, M . & Cooper, W . 1989. Predicting exit, voice, loyalty and neglect. Administrative Science Quarterly. 34: 521-539. Wolcott, H . 1990. Wri t ing up qualitative research. C A : Sage. Wolcott, H . 1992. Posturing in qualitative inquiry. In M . LeCompte, W . M i l l r o y & J. Preissle (Eds.), The handbook of qualitative research in education: 3-52. N Y : Academic Press. Wright, L . & Smye, M . 1996. Corporate abuse. O N : Key Porter Books. Y i n , R . 1984. Case study research. C A : Sage. 220 APPENDIX A INTERVIEW PROTOCOL Background I'd like to know a bit about your job and workplace where you experienced the injustice: Typical job duties and responsibilities Job title Involvement with others - independent vs. team Supervision of others How long have (had) you worked for the organization in which the unfair treatment took place? Are you still employed there? Critical Incident Can you describe the unfair treatment you experienced: what happened? who was involved - gender, position (hierarchy)? where and when it happened? were there any witnesses? how did you feel about it? What did you do? CHOOSE EITHER 1(a) or 1(b) I. (a) VOICE i. Individual Focus 1. How long before you told someone? 2. Who did you speak to first (and in what capacity): inside the organization - a boss, a co-worker, shop steward? outside the organization - friend or family member? 3. When you first spoke up, what alternative ways to let the organization know how you felt did you know about and did you consider these? 4. What made you decide to speak up? What were the two or three strongest factors? 5. (If not mentioned) - Did someone encourage you to speak up? 6. Are there any family members, friends or acquaintances who experienced a similar thing who also spoke up? Or did they remain quiet about it? 7. Did you have any reservations about speaking up and if so what were they? 8. At the crucial point in deciding whether to speak up, what could someone have said or done which would have changed your mind? 9. How did you feel after you spoke up? 10. Did you notice any changes in attitudes or behavior: in yourself any one else who was aware or made aware of your experience and response weren't aware but you believe they were reacting to you? 11. Have you had any similar experiences in speaking up in the past? (If yes) Were your feelings or behaviors the same or different this time? 12. What were your expectations in speaking up? 13. Were you surprised by anything about the process of speaking up or by your own responses? 14. What was the outcome? 15. What did you learn? Would you do it again? 16. What mistakes, if any, did you make? 17. How did speaking up affect you? 18. What would you say to someone you care about who finds themselves in a similar situation (i.e., similar injustice) and has not yet decided to speak up? 221 19. Do you think your age had anything to do with the unfair treatment you experienced? Gender? Anything other aspects about yourself that you believe was involved, whether in the issue itself, how you responded or the result of your response? 20. How do you feel about the experience now? ii. Organizational Focus 1. How responsive is the management of your organization to its voice systems? 2. What makes a voice system effective? 3. How would you change the voice system? 4. What's the worst story or incident of injustice that you've heard about it in your organization? 5. Has anything you would call unjust happened to your friends or colleagues at work? 6. Is there anything else you would like to add before we finish? II. PROJECTIVE EXERCISE Now I'm going to ask you to imagine that you can say anything you want to the person or persons involved or responsible for the incident or situation you've described to me and that they cannot do anything about what you say. Tell me now what you'd like to say, what you've always wanted to say to him or her. I. (b) SILENCE 1. When you experienced the unfair treatment were you aware of alternative ways to handle the situation and did you consider these? 2. What made you decide to remain quiet about your experience? What were the two or three strongest factors? 3. (If not mentioned) Did someone encourage you to do so? 4. Are their any family members, friends or acquaintances who experienced a similar thing who also choose to remain silent? How did they handle the situation? 5. Did you have any reservations about not speaking up and if so what were they? 6. Was there a crucial point at which you made your decision to stay silent - what could someone have said or done which would have changed your mind? 7. How did you feel (and do you feel now) after your decision to remain quiet? 8. Did you notice any changes in attitudes or behavior: in yourself any one else who was aware or made aware of your experience and response weren't aware but you believe they were reacting to you? 9. Have you had any experiences staying quiet in the past and if so were your feelings and behaviours the same or different this time? 10. What were your expectations in staying silent? 11. Were you surprised by anything about your process of staying silent or by your own reactions? 12. What was the outcome? 13. What did you learn? Would you stay silent if something similar happens again? 14. Did the experience of not complaining affect your health in any way - good or bad? 15. What would you say to someone you care about who finds themselves in a similar situation (i.e., similar injustice) and who is unsure what to do? 16. Do you think your age had anything to do with the unfair treatment you experienced? Gender? Anything other aspects about yourself that you believe was involved, whether in the issue itself, how you responded or the result of your response? 17. How do you feel about the experience now? ii. Organizational Focus 1. How responsive is the management of your organization to its voice systems? 2. What makes a voice system effective? 3. How would you change the voice system? 4. What's the worst story or incident of injustice that you've heard about it in your organization? 5. Has anything you would call unjust happened to your friends or colleagues at work? 6. Is there anything else you would like to add before we finish? 222 7. What would you have needed in order to speak out about your experience? II. PROJECTIVE EXERCISE Now I'm going to ask you to imagine that you can say anything you want to the person or persons involved or responsible for the incident or situation you've described to me and that they cannot do anything about what you say. Tell me now what you'd like to say, what you've always wanted to say to him or her. CLOSURE Now that the interview is over, what were reactions to it? How did you feel and how do you feel now? Is there anything that I didn't ask that I should have? 223 APPENDIX D LIST OF AVAILABLE VOICE SYSTEMS Voice Systems Which voice systems does your organization have in place: Formal grievance procedure Suggestion system Employee-management meetings Counseling service - EAP Ombudsperson Non-management task forces Question and answer program Survey feedback Open-door policy In your opinion, which is most effective for encouraging employees to speak up when they have experienced injustice and are discontented and why? 226 APPENDIX F DEMOGRAPHIC INFORMATION 1. Gender 2. Age 3. Ethnicity 4. Number of jobs you have held since leaving high school: • part-time • full-time 5. Number of different organizations you have worked for since leaving high school? 6. How long have you worked for the organization you are currently with? years months 7. Education level: not completed Grade 12 completed Grade 12 College diploma University degree: B . A . or equivalent M . A . or equivalent 228 APPENDIX G PARTICIPANT PSEUDONYMS AND JOB T I T L E S Case# Pseudonym Job Title l Rahim Contract Administrator 2 Gwen Employment Counsellor 3 Consuela Software Engineer 4 Jennifer Records Analyst 5 Faith Technical Writer 6 Suzanne Project Engineer 7 Brenda Office Assistant 8 Mark Executive Director 9 Eva Office Clerk 10 Karl Project Engineer 11 Shezadi Office Clerk 12 Irene Tour Director 13 Victor Sales Agent 14 Joanne Counsellor 15 Lisa Manager 16 Margaret Senior Programmer 17 John Engineer 18 Dan Facility Engineer 19 Vera Health Officer 20 Patrick Field Executive 21 Ashley Project Coordinator 22 Shaun Merchandiser/Sales Agent 23 Robert Social Worker 24 Nancy Registration Officer 25 Catriona Office Manager 26 Catriona Office Administrator 27 Linda Station Manager 27 Kathleen Assistant Curator 29 Denise Sales Assistant 30 Joe General Manager 31 Bill Architect 32 Chien-chu Engineer 33 Andrew District Manager 229 APPENDIX H INCIDENCE OF CATEGORIES OF ORGANIZATIONAL INJUSTICE Type Of Data Category Classification Singular Conjoint Archival (30 cases total) 1982 (10 cases) 21 8 D 8 P 4 S 2 cases 21 8 cases 4 DP 4 DPS 1987 (10 cases) 51 8 D 8 D 2 cases 2 / 8 cases 5 DP 3 /DP 1992 (10 cases) 21 7 D TP 2S 3 cases 1 / 2S 7 cases 6 DP 1 /DP T O T A L 91 23 D 23 P 6S 7 cases 5/ 2S 23 cases 15 DP 4 IDP 4 DPS Interview (33 cases) 16/ 14 D 17 P 17 S 15 cases 8 / 2P 5S 18 cases 4 DP 3 /S 1 IDP 6 DPS 1 /PS 3 IDPS Archival + Interview 22 cases 41 cases G R A N D T O T A L (63 cases) 25/ 37 D 40 P 23 S 13/ 2P IS 20 DP 3/S 5 IDP 10 DPS 1/PS 3 IDPS Categories of Organizational Injustice I = Interactional D = Distributive P = Procedural S = Systemic 230 A P P E N D I X I T R A I N I N G M A N U A L DECISION GUIDELINES FOR CODING CASES BY CATEGORIES OF ORGANIZATIONAL INJUSTICE Sections of The Training Manual: A. CATEGORIES OF ORGANIZATIONAL INJUSTICE 1. Interactional Injustice 2. Distributive Injustice 3. Procedural Injustice 4. Systemic Injustice B. SINGULAR VS. CONJOINT CLASSIFICATIONS Can a Category be Assigned? Singular Conjoint Unclassified Element Other 'Unclassifieds' or Exclusions C. GENERAL CODING PROTOCOL Transcripts Possible Coding Patterns Key Sections of Transcripts Viewpoint What to Do if You Find a Case Difficult to Code D. SPECIFIC CODING INSTRUCTIONS Steps Do's and Don'ts E. APPENDIX Interview Protocol Coding Forms (15 copies) Debriefing Form TRAINING MANUAL FOR CLASSIFYING CASES BY CATEGORIES OF ORGANIZATIONAL INJUSTICE Goal I have developed a typology for classifying participants' experiences (or cases) of workplace injustice into four categories: interactional, procedural, distributive and systemic injustice. Your job is to code, using these categories, a subset of the cases which I have coded. The purpose of this training is to achieve consistency (i.e., a high level of agreement) for between the three coders (Bonnie, Jay, myself) regarding the assignments of categories to cases. A. C A T E G O R I E S O F O R G A N I Z A T I O N A L INJUSTICE 1. Interactional Injustice a) Perceived injustice focused on interpersonal mistreatment from one or two authority figures to whom the participant reported. Authority figures included 'new hires' and 'old hands,' and were paid members of the participant's organization who were perceived as having greater authority than the participant. b) Usually one or two authority figures emerged as 'bad guys' (men or women) whose interactions with participants, regardless of intent, caused significant upset to participants. c) In cases involving smaller organizations, an authority figure's influence was sometimes significant which could make it difficult to separate interactional injustice from other types of injustice, particularly systemic. To distinguish between interactional and systemic injustice, ask yourself what would happen if the authority figure were removed: would injustice perceptions be substantively reduced (or eliminated)? Without his or her influence, would the essence or substance of the case disappear? If so, this is a case of interactional injustice. d) The main issue underlying interactional injustice is 'managerial maltreatment.' Example of Interactional Injustice: "He (the boss) suddenly started to get on my case for every little thing I did. Whatever I did, it wasn't good enough. He told me that he didn't know what was happening to me, that I used to be a 'good' employee but now I was 'bad.' Even if I had changed so much - which I hadn't, I wasn't doing anything differently - the way he supervised me sure didn't help. Once he tore a strip off me for a decision I'd made in front of everyone at our weekly meeting. And it wasn't a bad decision, it just wasn't what he would have done. I just started to feel so stupid and afraid." 232 2. Distributive Injustice a) Perceived injustice focused on resources seen as rewards or punishments (pay, work distribution, promotions, reprimands) believed to be misallocated (i.e., did not get paid enough, given too much work, did not get a promotion to which s/he felt entitled). b) Distributive injustice was separate from (i.e., not including) the procedures by which distributive decisions were made or manner in which they were communicated (although, as will be explained later, both distributive and procedural injustice may occur at the same time). c) Representative issues included pay/work inequity, involuntary reassignment, non-promotion and unjust dismissal. Example of Distributive Injustice: '7 was a casual staff member with the greatest seniority among the casuals. So when the full-time job came up, I applied for it knowing I would get it. But I didn't -I couldn't believe it. They hired someone from the outside, saying I didn't have the skills for the job. That was a bunch of shit. So I challenged it, took it to the union. We eventually won, it was some new manager who didn't know what they were doing who had hired from outside, but it took so long and it was so stressful to get the mess sorted out." 3. Procedural Injustice a) Participants perceived that the procedures by which distributive decisions were made or manner in which they were communicated were unfair. Procedural injustice also included the manner by which organizational practices or policies were implemented. b) Representative examples included procedures behind unfair performance reviews and unfair dismissals rather than the actual outcomes. Example of Procedural Injustice: "Apparently they received a complaint about me. So they did a performance review on me without telling me let alone checking out the information or asking me for input. They called up one or two clients who were unhappy because of things that weren't my fault or responsibility and used that to determine their assessment of me. It was not so much that it (performance review) wasn't written down...but had I been told when I considered taking the position 'we reserve the right to launch a performance review without telling you and we will let you know when we are finished,' I wouldn't have taken the job." 233 4. Systemic Injustice a) Perceived injustice focused on the larger organizational context within which interactions, outcomes and procedures occurred and typically included the contents and/or effects of organization-wide policies, practices or managerial styles. b) Participants' use of "they" rather than "he/she" to describe injustice was one indicator of systemic rather than interactional injustice. c) Offensive or troubling authority figures were typically common throughout the organization and were largely seen as secondary to other issues of injustice. Thus, injustice seemed larger than a given individual in the organization (i.e., if individuals were removed, injustice perceptions would remain). d) Representative issues included: i) organizational inconsistency: when organizations failed to honour policies or commitments; showed favoritism with promotions and discipline ii) organizational abandonment: when the responsibility given to employees was not matched by authority, support, protection and/or organizational accountability iii) discrimination: differential treatment of employees due to gender, race, sexual orientation, age iv) organizational overwork: when organizational practices routinely violated employment standards regarding compensation, conditions of work, etc. Example of Systemic Injustice: "God, they make me so mad. They are so inconsiderate about how they treat people there. Of all the hospitals in the province, theirs has the highest number of LTD (long-term disability) claims for stress-leave. You'd think that that would give them a clue but it doesn't. A while back I could feel myself starting to burn out. So I requested five days off - unpaid -for a holiday. They are so confused about the staffing schedule and my supervisor is so incompetent that up until 10 p.m. of the night before I was leaving, my supervisor was telling me Td have to cut my holiday short. She was wrong, dead wrong about the schedule but everybody had different ideas of what was allowed and what wasn't. My partner started screaming from the other room that she shouldn't be calling so late - we had to get up at 5 am to leave -that this should have been worked out a long time ago. Finally my supervisor said, 'don't worry, I'll sort it out.' They drive people crazy there." 234 B. SINGULAR VS. CONJOINT CLASSIFICATIONS Cases may involve one and only one category (i.e., singular), or two, three or all four categories of injustice (i.e., conjoint). Can a Category be Assigned? To be coded, a given category had to be a fundamental and/or significant feature of the case. To not classify it as a given category meant that either the essential nature or an essential aspect of the participant's experience of injustice would be missing and the case would be fundamentally different or misrepresented without it. Singular Cases reflecting one predominant category of organizational injustice were coded as singular. If content concerning what you considered to be the singular injustice were removed, there would be no substantive discontent. Conjoint Some cases reflected multiple (two or more) categories of injustice which were seen as significant and compelling aspects of the participant's injustice perceptions. They were coded as conjoint. In conjoint cases, if content concerning categories of injustice believed to be conjoint were removed from participants' descriptions, their experience or complaint would be fundamentally different or misrepresented. Unclassified Elements Sometimes elements of additional categories of injustice were present in cases. These elements may be indicative of certain types of injustice, but not sufficiently significant or compelling to warrant classification as a category of injustice (whether singular or conjoint). If you excluded these elements in coding the case, there would be little or no effect on the essence of the participant's experience; their complaint would not be fundamentally different. Other 'Unclassifieds' or Exclusions 1. Responses by organizations to participant's disclosure of discontent. Sometimes organizations exacerbated participants' perceptions of injustice by their response (or non-response) to participants' complaints. You are to focus on the situation prior to participants' requests for support or action. So, for example, if a participant felt frustrated because his complaints about injustice to his organization were ignored, you would not code injustice compounded by the organization's handling of the complaint. You would code the substance of the complaint he brought the organization. 2. Interactional Injustice Interactions with participants' subordinates were excluded from considerations of mistreatment. Interactions with employees in lateral positions could be included as long as there was a reporting relationship between the participant and the employee. In reporting relationships, the participant is accountable to, or is supervised or trained by someone with greater knowledge, expertise or authority. Interactions involving people who were not 235 members of participants' respective organization were excluded on the grounds that no reporting relationship was possible. C. GENERAL CODING PROTOCOL Transcripts I interviewed 32 participants who provided me with 33 cases of workplace injustice (one participant described two separate experiences of injustice from two different organizations). Transcripts were typed verbatim from tapes of these interviews. You will be studying 10 of the transcripts in their entirety, except for possible omissions of introductory and closing comments. Interviews were semi-structured; their format followed the Interview Protocol provided in the Appendix (although not necessarily in linear order). Your task I have coded each of the 33 cases as reflecting one or more of the four categories of perceived injustice: interactional, distributive, procedural and systemic. Your task is to read through 10 cases and apply the coding protocol to yield your judgments of the predominant categories of injustice involved in each case. Possible Coding Patterns There are 3 main possibilities or coding patterns which can occur: 1. the case is singular 2. the case is singular with unclassified elements 3. the case is conjoint 4. the case is conjoint with unclassified elements For both 1. and 2. one and only one category of injustice is coded. For both 3. and 4. two or more categories of injustice are coded. To code the case: a) Start by matching the nature of the complaints or issues to the category or categories described in Section A of this Training Manual. b) Then ask yourself: Are the complaints or issues compelling and significant parts of the injustice experience described by the participant? If these portions were removed, would it fundamentally alter the case? If only one issue or aspect of the case meets the 'compelling and significant' criterion, it is a singular case of injustice. If there are two or more issues or complaints which meet the 'compelling and significant' criterion, then it is a conjoint case. c) However, you must also ask yourself: 236 Are there indications of additional categories of injustice present? If these were removed, what would be the effect on the case? Remember that if there would be little to no effect on the substance of the participant's complaint or issue (i.e., the complaint would not be fundamentally different), these are considered 'unclassified elements' which are not coded. If these elements appear to be a substantive part of the participant's complaint or issue, then it may be 'codable' - go back to a) and b). Key Sections of the Transcript Make sure that you pay extra close attention to participants' responses to these key questions in the interview: 1. Tell me what happened. usually near BEGINNING of interview 2. What is the fundamental nature or essence of the injustice? Usually near MIDDLE 3. Projective exercise section (participants were asked to imagine that they could speak with impunity to the person or persons responsible).... usually near E N D In many cases, responses to these key portions of the interview are not contained in one place within interviews or in the same places across interviews. So you must be ready to be an 'injustice detective' to find the relevant portions of the interview and discover the data. Viewpoint It is very important for you to remember that, in your analyses, you try to understand each case from the participant's point of view, not your own. That is, although you may disagree with something the participant or I say, you are to code participants' experiences of injustice, not your own opinions or judgments about the participant, organization, events, responses, behaviours, etc. For example, if you think an organization should be able to fire pregnant employees, you must suspend this opinion for now and focus on the fact that the participant says that she considered it unfair. What to Do if You Find a Case Difficult to Code The process of coding participants' experiences of injustice is inherently subjective. Subjectivity can be a problem if it leads to biases which reduce the consistency or level of agreement between us as coders. You may find that some cases are harder than others; some may seem especially hard. Here's why and what you can do: 1. Non-Linear or Implicit' Information: Participants did not always respond to questions directly or in a linear way. When information is not explicit, you must look for 'clues' to carefully infer what the participant meant or said. 2. Ambiguous Communication: You will notice variations in participants' verbal ability, conciseness, directness, etc. Try to follow 'the story line' even when the story line seems 237 ambiguous or hard to follow. (The story line refers to the main sequence of events around which characters revolve - it's the 'guts' of the story.) 3. Ambiguous Categories: Some cases may not clearly and easily lend themselves to classification because possible category assignments seem ambiguous. For example, there could be 'hints' of interactional injustice in a case but not enough to code it as a case of interactional injustice. (These hints are called 'unclassified elements' and are discussed in the next section). Remember to ask yourself 'what is going on here, what are the big complaints?' according to the participant. D. SPECIFIC CODING INSTRUCTIONS 1. Code one case at a time. Read through the transcript before coding. Get an overall sense of what happened, what the participants' major complaint(s) were. Note on a separate piece of paper the pages and/or sections of the transcript you feel are key. 2. Using Coding Forms (provided in the Appendix), assign category(ies). Make sure you indicate on the Form where requested your grounds for assigning each category. 3. Leave that case and go on to something else (another case, take your dog for a walk) 4. Go back to the coded case. Test your initial judgment - does it hold up? How would you respond to a challenge to your judgment - how easily and convincingly could you argue that it truly is a case of systemic injustice, for example? If necessary, change your initial coding of the case. Retain only those categories which you feel you can soundly defend. REMEMBER DO NOT • Ask A N Y O N E else (friends, family members, Karen, Jay, Bonnie) for help • Code each case in absolute, not relative terms - in other words, code each case on the basis of its own merits, not compared to other cases DO • Take your time and make reasoned judgments which you can defend • Fi l l out a Coding Form for each case • Complete the Debriefing Sheet when you have finished coding all cases Save specific questions for the Coding Conference Call (date and time to be agreed upon at the training session August 29, 1997). 238 Coder's Name: Participant Code #:. CODING FORM C A T E G O R Y O F INJUSTICE Y E S O R NO R A T I O N A L E F O R Y E S 1. Interactional Injustice 2. Distributive Injustice 3. Procedural Injustice 4. Systemic Injustice S U M M A R Y : Singular (Y/N): Conjoint (Y/N): If yes, # categories: 239 DEBRIEFING F O R M The following questions require brief answers only: 1. Coding a) what was the easiest aspect(s) to coding the cases? b) what was the hardest aspect(s)? 2. Training a) what was the most effective part(s) of the training? b) what was the least effective? c) how would you recommend the training be changed (if at all) 3. Additional Comments: Thank you for your efforts APPENDIX J INTER-RATER RELIABILITY My ratings Rater 1 Interactional Systemic Distributive Procedural Not Present Total Interactional 5 1 6 Systemic 6 6 Distributive 4 3 7 Procedural 5 2 7 Not Present 14 14 Total 5 6 4 5 20 40 K i = .80 My ratings Rater 2 Interactional Systemic Distributive Procedural Not Present Total Interactional 4 2 6 Systemic 5 1 6 Distributive 4 2 6 Procedural 5 1 6 Not Present 16 16 Total 4 5 4 5 22 40 K 2 = .79 ' Agreement between myself and each of two raters was determined for 30% of the interview cases (10 of 33 cases). Cohen's Kappa (K) was used as the measure of agreement, corrected for chance, between the two raters (Cohen, 1960; Fisher & Van Belle; 1993). 241 APPENDIX J (cont.) INTER-RATER RELIABILITY Rater 1 Rater 2 Interactional Systemic Distributive Procedural Not Present Total Interactional 5 1 6 Systemic 5 1 6 Distributive 6 1 7 Procedural 6 1 7 Not Present 14 14 Total 5 5 6 6 18 40 K 3 = .87 242 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0088786/manifest

Comment

Related Items