UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Job satisfaction in child welfare : a study of line social workers Gorrie, Ernest David 1990

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

Item Metadata

Download

Media
831-UBC_1990_A5 G67.pdf [ 9.98MB ]
Metadata
JSON: 831-1.0098003.json
JSON-LD: 831-1.0098003-ld.json
RDF/XML (Pretty): 831-1.0098003-rdf.xml
RDF/JSON: 831-1.0098003-rdf.json
Turtle: 831-1.0098003-turtle.txt
N-Triples: 831-1.0098003-rdf-ntriples.txt
Original Record: 831-1.0098003-source.json
Full Text
831-1.0098003-fulltext.txt
Citation
831-1.0098003.ris

Full Text

JOB SATISFACTION IN CHILD WELFARE: A STUDY OF LINE SOCIAL WORKERS By ERNEST DAVID GORRIE B.A., The University of British Columbia, 1976 B.S.W., The University of British Columbia, 1984 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SOCIAL WORK  in THE FACULTY OF GRADUATE STUDIES SCHOOL OF SOCIAL WORK  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA June 1990 e  Ernest David Gome, 1990  In  presenting this  degree  at the  thesis  in  University of  partial  fulfilment  of  of  department  this thesis for or  by  his  or  requirements  British Columbia, I agree that the  freely available for reference and study. I further copying  the  representatives.  an advanced  Library shall make it  agree that permission for extensive  scholarly purposes may be her  for  It  is  granted  by the  understood  that  head of copying  my or  publication of this thesis for financial gain shall not be allowed without my written permission.  Department  of  SOC/frC  W/gRK  The University of British Columbia Vancouver, Canada  Date  DE-6 (2/88)  JU/Ut Xi,  ii  ABSTRACT  The general issue of defining job satisfaction and the adverse implications of low job satisfaction are discussed. This is followed by an application of the research to the specific problem of job satisfaction among child welfare line social workers in a provincial government agency.  The needs satisfaction and values satisfaction model of job satisfaction theory are described and a rationale is provided for the selection of the needs satisfaction model for this research. A variety of research instruments are reviewed, including the Job Satisfaction Survey Questionnaire, the Job Descriptive Index, the Quality of Work Life Survey, and the Professional Satisfaction Inventory. A new quantitative measures questionnaire is introduced for use specifically among social workers.  It was hypothesized that job-specific variables will be stronger correlates of a facet-free measure of job satisfaction than will variables not specifically related to social work. It was hypothesized that specific differences will exist between the importance attributed to variables between less experienced and more experienced workers. It was further hypothesized that specific differences will exist between the satisfaction with variables as reported by less experienced workers and more experienced workers. Finally, hypotheses were presented regarding specific differences in job related correlates of job satisfaction, between less experience and more experienced workers.  iii The research proceded by way of questionnaire among child welfare line social workers in the Lower Mainland area of British Columbia. Extensive efforts were made to ensure confidentiality while allowing the opportunity for followup research. The methodology of this confidentiality plan is explained. A sample of 60 social workers resulted in 49 questionnaires being returned.  Support was found for the hypothesis that job specific variables were better correlates of job satisfaction than were variables not specific to social work. There was neither confirmation nor disconfirmation for the hypothesis that there would be differences in the importance of variables between less experienced and more experienced workers. Satisfaction was higher for those variables under the control or influence at the local level than for variables which were controlled centrally. The few significant differences between less experienced and more experienced workers in satisfaction with variables were explainable by objective influences rather than subjective experiences. Only one variable, getting a sense of accomplishment from the job, was a significant correlate for both less experienced and more experienced workers. Less experienced workers also demonstrated correlations between job satisfaction and satisfaction with professional identification, while more experienced workers demonstrated correlations between job satisfaction and control of their work.  Recommendations are made for the government which employs social workers, the agency which administers programs, supervisors of social workers, and the union which bargains on behalf of social workers.  iv Table of Contents  Page  ABSTRACT  ii  LIST OF TABLES  xi  LIST OF FIGURES  xiii  ACKNOWLEDGEMENT  xiv  CHAPTER 1  1  Background and Problem Area  1  The Problem Area - What Is Job S a t i s f a c t i o n ? Why Job S a t i s f a c t i o n Effects Possible  Research?  1 3  of Low Job S a t i s f a c t i o n  C o r r e l a t e s of Job S a t i s f a c t i o n  3 7  S o c i e t a l Awareness of the Problem  11  Problematics  12  Summary  13  CHAPTER 2  15  I n t e r a c t i o n of P r a c t i c e and Research  15  The S e t t i n g f o r The F i e l d P r a c t i c e  15  P r e p a r a t i o n f o r Research  21  D i s s e m i n a t i o n of R e s u l t s  22  Successes and L i m i t a t i o n s of C o l l a b o r a t i v e Process Summary  CHAPTER 3 The Research Problem  the 24 25  27 27  V  T h e o r e t i c a l Overview  27  Needs S a t i s f a c t i o n Model  27  Value S a t i s f a c t i o n Model  29  Job S a t i s f a c t i o n  30  vs D i s s a t i s f a c t i o n  Questions f o r Research  .  .  32  Conceptual Model  35  Knowledge B u i l d i n g F u n c t i o n  38  Formal Hypotheses  40  R e l a t i o n s h i p To Other Research  41  L i m i t a t i o n s of The Research  43  CHAPTER 4  45  Research Design  45  R e s e a r c h e r ' s C o n t r o l Over Phenomena S t u d i e d .  45  Sampling Design  45  Timing of the Data C o l l e c t i o n  46  Choice of M e t h o d o l o g i c a l O r i e n t a t i o n  . . . .  Data C o l l e c t i o n  47 48  Confidentiality Provisions  51  The Q u e s t i o n n a i r e  54  The Job S a t i s f a c t i o n  Survey Q u e s t i o n n a i r e .  .  54  The Job D e s c r i p t i v e Index (JDI)  55  The Q u a l i t y of Employment Survey  56  The P r o f e s s i o n a l  Satisfaction  The S o c i a l Work Job Questionnaire  Inventory  . . .  57  Satisfaction 57  vi Validity,  Reliability,  And P r e - T e s t i n g  61  Data A n a l y s i s  63  Data A n a l y s i s Methods  64  Job S a t i s f a c t i o n  Variables  64  Inferences To Be Drawn From The Data  66  Ethical  68  Issues In The Research  i  CHAPTER 5  70  Research F i n d i n g s  70  Introduction  70  Descriptive  72  Data  F (Importance) V a r i a b l e s  72  S (Satisfaction)  76  D (Discrepancy)  Variables Variables  C o r r e l a t e s Of O v e r a l l Job S a t i s f a c t i o n S (Satisfaction) D (Discrepancy)  Variables Variables  Between Group D i f f e r e n c e s  77 78 78 81 82  F (Importance) V a r i a b l e s  82  S (Satisfaction) Variables  82  C o r r e l a t e s Of O v e r a l l Job S a t i s f a c t i o n  84  S (Satisfaction) Variables  84  D (Discrepancy)  86  Hypothesis Summary  Testing  Variables  88 92  vii CHAPTER 6  93  C o n c l u s i o n s and Recommendations  93  Introduction  93  Comparisons With Other Research  94  Facet Free Job S a t i s f a c t i o n  94  C o r r e l a t e s Of Job S a t i s f a c t i o n  97  Recommendations  101  Recommendations To The Government  101  General Recommendations  102  Workload And Hours  102  RECOMMENDATION G . l  105  Vacation  106  RECOMMENDATION G.2  106  Access To Resources  107  RECOMMENDATION G.3  107  Promotional O p p o r t u n i t i e s  108  RECOMMENDATION G.4  108  Recommendations S p e c i f i c To Less Experienced Workers  108  P r o f e s s i o n a l Development  109  RECOMMENDATION G.5  110  Promotional O p p o r t u n i t i e s  Ill  Workload And Hours  Ill  RECOMMENDATION G.6  Ill  Recommendations S p e c i f i c To More Experienced Workers  Ill  viii Workload And Access To Resources  112  RECOMMENDATION G.7  112  Recommendations To The Agency  112  General Recommendations  113  Workload  113  RECOMMENDATION A . l  115  RECOMMENDATION A . 2  116  RECOMMENDATION A . 3  116  Professional Recognition  116  RECOMMENDATION A . 4  119  R e l a t i o n s h i p With S u p e r v i s o r  119  RECOMMENDATION A . 5  119  RECOMMENDATION A . 6  120  Agency Respect  120  For Employees  RECOMMENDATION A . 8  121  Recommendations S p e c i f i c To Less Experienced Workers RECOMMENDATION A . 9  121 122  Recommendations S p e c i f i c To More Experienced Workers Recommendations To S u p e r v i s o r s  122 123  General Recommendations  123  C l a r i t y Of Job R e s p o n s i b i l i t i e s  123  RECOMMENDATION S . l  124  Autonomy And C o n s u l t a t i o n  124  RECOMMENDATION S.2  125  ix S u p e r v i s o r y Respect For Employees  125  RECOMMENDATION S.3  125  Hours Worked  126  RECOMMENDATION S.4  126  Recommendations S p e c i f i c To Less Experienced Workers  126  Recommendations S p e c i f i c To More Experienced Workers  127  RECOMMENDATION S.5  127  Recommendations To The Union  127  Pay And V a c a t i o n  129  RECOMMENDATION U . l  129  Agency Respect For Employees  129  RECOMMENDATION U.2  130  R e c o g n i t i o n Of S o c i a l Work As A P r o f e s s i o n  .  RECOMMENDATION U.3  130 131  Recommendations Regarding A c h i e v i n g A Sense  Of AccomplisSment  131  RECOMMENDATION 0.1 Recommendations For Further Research  133 . . . .  133  BIBLIOGRAPHY  135  APPENDIX A  138  APPENDIX B  156  APPENDIX C  158  APPENDIX D  165  APPENDIX E  169  X  APPENDIX F  1  7  6  APPENDIX G  1  7  7  APPENDIX H  1  9  3  APPENDIX I  2  0  1  APPENDIX J  2  1  6  APPENDIX K  2  1  7  APPENDIX L  2  1  9  xi L i s t of Tables  Page  Table 4.1  Advantages Of Interviews vs Q u e s t i o n n a i r e s  Table 4.2  Disadvantages  Of Interviews  .  vs  Questionnaires  Table 4.3  50  Examples of D i r e c t i o n a l vs N o n - D i r e c t i o n a l Questions  Table 5.1  59  Descriptive  S t a t i s t i c s Regarding G e t t i n g A  Sense Of Accomplishment  Table 5.2  72  Importance Of G e t t i n g A Sense Of Accomplishment From The Job  Table 5.3  73  C o r r e l a t i o n s Between SOVER And Other Satisfaction  V a r i a b l e s (p <= 0.01)  Table 5.4  Discrepancy V a r i a b l e C o r r e l a t i o n s With SOVER  Table 5.5  T - T e s t s Of S a t i s f a c t i o n  C o r r e l a t i o n Of SOVER With Variables  79  81  V a r i a b l e s With  TOTGOV As Grouping V a r i a b l e  Table 5.6  49  83  Satisfaction  (TOTGOV <= 3 Years)  85  xii Table 5.7  C o r r e l a t i o n Of SOVER With S a t i s f a c t i o n (TOTGOV > 3 Years)  Variables  Table 5.8  C o r r e l a t i o n Of SOVER With Variables  Table 5.9  Table 6.1  Discrepancy  (TOTGOV <= 3 Years)  C o r r e l a t i o n Of SOVER With Variables  86  87  Discrepancy  (TOTGOV > 3 Years)  87  Comparison Of Job S a t i s f a c t i o n  Scores  Between Barber And Current Research  Table 6.2  Comparison Of Job S a t i s f a c t i o n  95  . . . .  Scores  Between V a r i o u s Research R e s u l t s And Current 96  Research R e s u l t s  Table 6.3  Table 6.4  S a t i s f a c t i o n With I n d i v i d u a l  V a r i a b l e s As  Significant  Correlates  Of Job S a t i s f a c t i o n  Discrepancy  Scores Of I n d i v i d u a l  As S i g n i f i c a n t C o r r e l a t e s  Of Job  T o t a l M i n i s t r y F u l l Time E q u i v a l e n t Positions  99  Variables  100  Satisfaction  Table 6 . 5  .  Staff 105  xiii List of Figures Chart 2.1  Chart of Organisation of The Ministry of Social Services and Housing  Map 2.1  Page  19  Map of Ministry of Social Services and Housing Region C  20  xiv ACKNOWLEDGEMENT  This research would not have been possible without the continuing support of the Ministry of Social Services and Housing. My thanks go to the Personnel Department of the ministry for its support for professional development through the Professional Social Worker Education Program. I am particularly grateful to my many co-workers over the past several years of part-time study. They not only assisted me in developing my work, they also adjusted to my repeated absences. I hope this research will contribute to improving their job satisfaction.  My thanks to my wife, Glenna Jones, for putting up with the many missed dinners, working weekends, and my years of being distracted. Finally, my thanks to my parents, Peter and Martha Gorrie, who have consistently loved me for who I am, but were delighted when I became more than what I had been.  1 CHAPTER 1 BACKGROUND AND PROBLEM AREA  The Problem Area - What Is Job Satisfaction? People in North America who are employed full-time spend extended lengths of time in their employment. The full-time work week is considered to be between 35 and 40 hours per week, although many people spend more time than that at work. And the effects of this one-third of their days carries over into other parts of their lives. People spend time talking about work, preparing for work, travelling to and from work, and sometimes dreaming about work. Work is always in their lives. In fact, for some people work is their life. But what is this part of their lives like? How satisfied are people with their jobs? What do they say is important to them in their jobs? What are they satisfied with in their jobs? What really is important for them in experiencing job satisfaction? What would need to be different for their job satisfaction to be better?  A major consideration in deciding whether to study a topic is to determine what it is one is proposing to study, i.e. what is the definition of the topic. Here, the topic of job satisfaction is found to be wanting. Few people have defined it, even when researching it. Portigal, using the terms job satisfaction and work satisfaction interchangeably, avoided the definition by saying, "Because it is unclear at present precisely what approach to work satisfaction will ultimately prove most fruitful, it does not seem useful to propose a detailed definition of the concept" (pg. 8).  2 Smith et al, developers of the Job Descriptive Index, define job satisfaction variously as, "the feelings a worker has about his job" (pg. 6), "persistent feelings toward discriminable aspects of the job situation" (pg. 37), and "an affective response to distinguishable aspect of the job, evaluated in relation to appropriate frames of reference" (pg. 87).  The Merriam-Webster Dictionary defines job as;  "1: a piece of work, 2: something that has to be done, 3: a regular remunerative position"  and satisfaction as;  "2: contentment, gratification".  Presumably job satisfaction is contentment or gratification which one receives through a regular remunerative position.  Why then do researchers have difficulty defining job satisfaction? This difficulty may be in part due to differences in theoretical frameworks. Although they differ in some aspects, the frameworks have similarities in other aspects. This shifting of agreement and disagreement results in difficulty in confirming any one definition. These differing orientations will be discussed in Chapter 3 - The Research Problem.  3  For the purpose of this research job satisfaction is defined as the cumulative effect of elements included in an employment situation which, based on individual differences are experienced differently by workers, resulting in reports of varying experiences of satisfaction. This research deals with this phenomenon among a limited group of employees: child welfare line social workers employed by the British Columbia Ministry of Social Services and Housing.  Why Job Satisfaction Research? Why is job satisfaction among child welfare line social workers an issue worthy of research?  The field of social work is often believed to be staffed by people who are "not in it for the money, they're in it for the job satisfaction". It is often said that social workers are persons who seek personal fulfilment by performing altruistic acts. Yet those employed in social work often experience variations in their levels of job satisfaction. Social workers working in the same agency may have very different experiences of job satisfaction.  Effects of Low Job Satisfaction Low job satisfaction can result in a number of effects in different types of employment, and in different employment climates. Kermish and Kushin investigated high employee turnover among social workers and supervisors in a northern California metropolitan county social welfare agency to determine the reasons for staff separation. They found that most had left because of dissatisfaction with their job. They also found that 54% of those interviewed  4  by the researchers "might have remained, were there some modification of agency atmosphere and/or policy" (pg. 137).  These workers did not necessarily leave human service work. In fact, Kermish and Kushin determined that 71% of the separated workers subsequently returned to social work and that an additional 15% entered the related service profession of teaching. They were not leaving human services, they were leaving a particular human services agency.  Herrick et al researched social workers who actually left, or never entered, social work practice. They found that of those who were employed at the time of the research 55.7% gave criticisms of social work employment as their primary reason for leaving vs only 30.2% who gave attractions of another job as their primary reason for leaving.  Among Canadian workers in a variety of fields of work, Burstein et al found that,  "Those who are least satisfied with their jobs are more likely to change employment, and to be absent from work. This same group is also less likely to work overtime, to classify themselves as conscientious, or to feel a sense of commitment, to either their jobs or their employers" (pg. 25).  5 Barber found a statistically significant negative correlation between job satisfaction and absenteeism among human service workers. Barber surveyed 3,935 employees in a range of job classifications in a state social insurance agency and received 2,521 returned questionnaires. He found that, "dissatisfied workers reported significantly more days absent than satisfied workers" and that "this relationship was particularly strong for workers who missed 15 or more days" (pg. 31). Barber further found that dissatisfied workers were more likely to be thinking of changing jobs (p < 0.01) and were less likely to believe they would still be working for the same agency in three years time (p < 0.01).  But job satisfaction is a concern for more than employers. It has direct implications for individual workers. Jayaratne et al (1986) reported that social workers who scored low on job satisfaction were likely to be more anxious, more depressed, more irritated, have lower self-esteem, and have more health problems than those who scored higher on job satisfaction. These workers reported more dissatisfaction with their marriages and they were more likely to report their spouses as being irritable, anxious, and depressed.  We have seen evidence that one effect of job dissatisfaction is high staff turn-over. This has obvious implications for social work administration in the areas of costs for recruitment, expenses related to new employee training, and lower efficiency of work done by inexperienced employees. High turn-over has adverse implications for quality of service to clients. Clients may be deprived of the work of experienced social workers if an agency is characterized by high social worker turn over. Children in the care of a child welfare agency  6  may experience multiple social workers rather than the continuity and stability of a single social worker for extended periods of time.  Low job satisfaction has implications for an agency even when high staff turn-over is not apparent. Low job satisfaction combined with low staff turnover may result if there are limited alternative employment opportunities. This scenario results in low productivity among existing employees, and may result in the establishment of a negative culture in the work environment. This low productivity may be difficult to document due to the difficulty of defining productivity in child welfare practice. It was documented in the Barber study with regards to work absences, however. This environment may then be difficult to modify when the problem becomes recognised.  Clients may also receive lesser quality of service from social workers who experience low job satisfaction. We have seen that these workers are more likely to be exhibiting physical health, mental health, and relationship difficulties in their own lives.  Low job satisfaction has a long term impact on the attractiveness of social work as a profession as compared to other helping professions. Low job satisfaction among social workers may result in more competent practitioners entering other fields of human service as was demonstrated in the Kermish and Kushin study.  7 Possible Correlates of Job Satisfaction The literature to date tentatively identifies a number of important factors which correlate with job satisfaction to varying degrees. It also identifies some of the differences which exist between job classifications.  Kermish and Kushin (1069) interviewed a sample of social workers who left their jobs to determine reasons why they left. Many of them had completed "exit reports" for their employers. Unfortunately many of these reports were found to be so vague that they were of no value in determining why people had left. The major complaints cited in the interviews with Kermish and Kushin were:  "1)  overwhelming job demands;  2)  poor atmosphere and morale within the agency;  3)  inability to be of real help to the client;  4)  poor supervision;  5)  little respect, encouragement, and support for the worker by agency administration; and  6)  little opportunity to use one's own initiative and to be creative".  8  Job satisfaction research specific to human service workers can be found in American research in the mid 1980s. Barber summarized the findings of a number of researchers who examined job satisfaction and identified 11 elements which may be included in determining job satisfaction in a range of job situations. Factors which were job related included;  i)  the nature of the work,  ii)  having a sense of achievement,  iii)  having appropriate responsibility,  iv)  getting recognition,  v)  having opportunity for advancement,  vi)  pay.  vii)  job security, and  viii)  working conditions.  Factors which were related to human agents within the job included;  i)  technical competence and relationship qualities of supervisors or supervision,  ii)  mutual supportiveness, common work attitudes and values of co-workers, and  iii)  the company or organization's personnel policies and expectations of employees.  Barber found that among differing job classifications within a single agency, different variables correlated with job satisfaction. Of the 19  9 variables he studied, only "getting a feeling of achievement" was important for the five job groupings of i) direct service workers, ii) supervisors, iii) managers, iv) technical/support staff, and v) clerical staff. And only two factors, "autonomy" and "technical aspects of supervision" were not significant for any classification. Within the population of employees of a social service agency there was a wide variation in the constellation of factors which would best predict job satisfaction.  McNeely, Feyerherm, and Johnson investigated job satisfaction within a single agency over time. They reported that within that agency, over a period of two years during and subsequent to a re-organization of the service delivery model, there were substantial changes in which variables significantly correlated with overall job satisfaction. Ten variables in 1979 yielded an R Square of 0.364 on a multiple regression while ten variables in 1981 yielded an R Square of 0.690 on a multiple regression. Yet only three variables were significant contributors to the regression on both occasions. Those variables were:  1)  the job being dull and monotonous (negative correlation);  2)  clear job expectations from the supervisor (positive correlation) and;  3)  feeling very much underpaid (negative correlation).  Fourteen other variables contributed to the matrices, yet seven of them were present only on the first occasion and seven were present only on the second occasion.  10  Burstein et al had a representative sample of employed Canadians rate the importance of 34 job characteristics, first for the importance which they ascribed to each characteristic, then for how nearly that characteristic was found in their job to obtain a job satisfaction rating for each characteristic for each respondent.  Burstein et al also found that most Canadians surveyed reported job satisfaction. In fact, 46% described themselves as "very satisfied", 42% were "somewhat satisfied", 9% "not too satisfied" and 2% "not at all satisfied".  Early research into job satisfaction in Canada was non-specific in type of employment. Burstein et al found that in general, Canadian workers reported themselves as being satisfied with their jobs, yet provided no correlations between specific job satisfaction variables and the overall job satisfaction.  Job satisfaction among social workers then can be seen to have important implications for social workers individually, for clients, for employers of social workers, and for the profession of social work. We have also seen that research into job satisfaction results in different outcomes depending on the nature of the job being studied and the nature of the agency organization and structure.  11  Societal Awareness of the Problem Job satisfaction among social workers has not been extensively researched. As a result, it is difficult to ascertain the extent of societal awareness of this issue. It does not have immediate direct impact on members of the public. Literature shows no research into public perceptions of job satisfaction amongst child welfare line social workers specifically or amongst social workers in general, so it is not possible to determine societal awareness of the problem at this time. It would seem reasonable to hypothesize though that awareness would be greatest among those members of the public who have the greatest contact with child welfare line social workers. These would include adult clients of child welfare services; caregivers, such as foster parents and resource staff; collateral professionals such as teachers, doctors, lawyers, and school counsellors; and to some extent, older children who receive child welfare services. The concept of job satisfaction may not be included in their frames of reference however. They may have different perceptions of what is important for job satisfaction in their own fields and not relate difficulties within child welfare social work to job satisfaction among the social workers. Additionally, some of the effects of varying levels of job satisfaction and dissatisfaction may be the result of other factors, so the level of job satisfaction may not be perceived as a cause of desirable or undesirable effects.  Finally, many people may view job satisfaction as an employer-employee issue, to be dealt with solely within that context. It may not be viewed as a social problem, rather as an individual or intra-organization problem, not appropriate for intervention by those not party to the employer-employee  12  relationship. Just as abuse within families was once viewed as a personal matter not appropriate for legally mandated intervention, it may be that job satisfaction is viewed as a personnel matter, to be addressed solely by the parties.  Problematics Perhaps the greatest problem regarding job satisfaction among child welfare line social workers is a wariness regarding discovering the existence of a problem. It would be difficult to research correlates of job satisfaction without establishing levels of job satisfaction. If one were to determine that job satisfaction was low, involved parties would then have to decide whether to ignore an acknowledged problem or to take steps to rectify the problem.  The agency in which the research took place was very open to exploring this issue. Authorization for research within the agency requires approval from the deputy-minister, the highest non-elected official within the government ministry. Mr. Dick Butler, the deputy-minister at the time of the research was not only willing to have the research take place, he actively encouraged the research.  A second possible problem with research of this nature is that the affected parties may be tempted to influence the development of the research in a way designed to affect the outcome and recommendations. Major possible influences in this research were the employing agency, the workers themselves, and their union. The workers involved were members of the B. C. Government Employees' Union. The researcher is pleased that neither the employing  13 agency, nor the union made any efforts to influence the development of the research tool, the results of the research, or the interpretation of the results. This may be due in part to the researcher's considerable involvement within both organizations, but it may be due to the interests of both parties in the development of a knowledge base which is agreed upon as being unbiased.  A final problem was the absence of an appropriate research instrument. A review of existing instruments for job satisfaction research indicated that none were appropriate for the study of job satisfaction among child welfare line social workers in the Ministry of Social Services and Housing. Accordingly, it was necessary for the researcher to develop a new questionnaire for use in this research. That questionnaire is included here as Appendix A. Discussion of the need for the development of the research instrument is included in Chapter 4.  Summary This chapter has explored job satisfaction as an issue in the human services and has identified the issue of job satisfaction among child welfare line social workers as being the specific object of this research. The literature was reviewed to demonstrate some effects of varying levels of job satisfaction among workers in general and among some groups of human services employees specifically. We have seen how low levels of job satisfaction have adverse impacts on individuals, on agencies, and on clients receiving social work services. This chapter briefly discussed some issues regarding societal awareness of the problem and drew parallels with the earlier perception of child abuse being defined as not a social problem. It closed with  14 an overview of the potential problems involved in this kind of research including: the possible reluctance of an employer to acknowledge the possibility of a problem; the possibility of improper outside influences on the researcher; and the absence of an appropriate research instrument.  15 CHAPTER 2 INTERACTION OF PRACTICE AND RESEARCH  The Setting For The Field Practice The British Columbia Ministry of Social Services and Housing (M.S.S.H.) is the sole agency providing legislatively mandated child welfare services in British Columbia. Unlike some other jurisdictions in which legislatively mandated child welfare services are provided by small, locally organized and locally administered agencies, M.S,S.H. is a large, provincially organized, and centrally administered agency operated as a ministry within the provincial government. It's total staffing in 1989-90 was 4,424 full-time equivalent positions. This included staff for all programs listed below, as well as centralized divisions such as research, personnel, etc. The ministry provides a range of services in addition to child protective services.  The ministry is responsible for the administration of all provincial programs which provide income assistance under the Guaranteed Available Income for Need (G.A.I.N.) Act. This includes income support for clients who are temporarily awaiting benefits from other agencies (e.g. federal Unemployment Insurance Commission, Workers Compensation Board), clients who are unable to secure employment due to meeting defined unemployability criteria, unemployed clients who are not eligible for benefits under other employer or government programs, and employed clients whose total monthly income is less than that which a client in receipt of G.A.I.N. benefits would receive. This division of M.S.S.H., known as the Income Assistance Division,  16 operated with a budget of $905,400,000 in the fiscal year 1987-88, the year nearest to the research for which data is available.  The ministry is also responsible for providing services to mentally handicapped residents in British Columbia. These services include residential placements, services to mentally handicapped adults living in the community, and support services to families of mentally handicapped children and adults. Budgets for this division of M.S.S.H. were subsumed within budgets for other divisions in 1987-88.  The ministry is responsible for the development of publicly supported housing initiatives in British Columbia, through the B.C. Housing Management Corporation. This division of M.S.S.H. operated on a budget of $16,300,000 in 1987-88.  Finally, the ministry is responsible for child protective services under the Family and Child Service Act (F.S.C.A.), Adoption Services under the Adoption Act, and has some non-protective child welfare responsibilities under the Family Relations Act. This division of M.S.S.H. consisted of approximately 800 field social work positions, was responsible for 5,662 children being admitted into care, conducted 26,283 child welfare investigations, and had a budget of $136,100,000 (including social work staffing for both child welfare and services for mentally handicapped adults) in 1987-88. This is the division within which the research took place, although research was limited to child welfare social workers.  17 Region C, was allocated approximately 97 social worker positions, of which approximately 60 were allocated to child welfare district offices. The remainder of approximately 37 social worker positions were assigned to services for mentally handicapped people and resource development and liaison. There were 683 children-in-care of the ministry on March 31,1988 within the boundaries of the region studied.  Most policies within M.S.S.H. are provincial in scope, and are established by a variety of centralized authorities. Government Personnel Services Division (G.P.S.D.) is responsible for personnel policies effecting all provincial government ministries, while M.S.S.H. Personnel Services is responsible for ministry specific personnel issues. Most employees, including all of the participants in the research are represented in collective bargaining by the British Columbia Government Employees Union (BCGEU). The BCGEU also represents most other employees of the provincial government.  The office of the Superintendent of Family and Child Service (the Superintendent) is responsible, along with the Family and Children's Services Division, for development of child welfare policies. The Superintendent's office is also responsible for the Inspection and Standards Unit, which monitors general standards of practice and investigates allegations of improper practice. The Superintendent's position is not independent from the ministry, rather it is an assistant-deputy minister position within the ministry.  Family and Children's Services Division is also responsible for the implementation of policy. At the time of the research, it was responsible for  18 consultative services to the field, and policy interpretation services including contact with direct practice social workers.  Chart 2.1 provides an administrative description of the ministry.  The Ministry of Social Services and Housing Region C includes the area bounded by the Bumaby/Vancouver border to the west, Mission to the east, the Fraser River to the south, and the sparsely populated mountain area north of the Fraser River to the north. It includes urban areas (e.g. Burnaby, New Westminster) as well as semi-rural areas (e.g. Maple Ridge, Mission). Map 2.1 provides a graphical representation of the area. Region C employs social workers with child welfare experience ranging from a few months to many years.  Within Region C, area CI is the designation for child welfare services in the eastern portion and area C4 is the designation for child welfare in the western portion. Within each of these areas services are delivered from district offices. At the time of the research, services dealing with the development, funding, negotiation of contracts, liaison and monitoring of child welfare resources were performed by specialized teams known as Resource Teams. These teams were not included in the research. District offices which dealt with the range of direct individual and family client contact were the only offices included in the research. Approximately 60 workers located at 10 district offices work as child welfare line social workers within the two areas. The numbers vary as absent workers are not always replaced and additional  Province ol British Columbia  Ministry ot Social Services and Housing  • ASSOCIATE SUPERINTENDENT -INSPECTION* STANDARDS  JANUARY 16,1989  —1 WOODLANDS  REGIONS A, 8, Ct . F,G,H.J,K.4i, REGIONAL DIRECTORS  .1  OS  AREA MANAGERS  U  fjkca.  MCOUEASSBTANCE -HEALTH SERVICES • FAMILY MAINTENANCE • EMPLOYMENT INITIATIVES CORPORATE SERVICES EMERGENCY SOOAL  ADHM SERVICES DIV. STAFF TRAINING A LIBRARY FINANCIAL SERVICES DW. FINANCIAL PLANNING BV. CAP ACCOUNTING OFFICE SYSTEMS SERVICES DIV. AUDTT SERVICES DIV. PERSONNEL SERVICES DIV.  BRITISH COLUMBIA HOUSING MCMT. COMMISSION HOUSING DIVISION SENIORS PROGRAMS  FEDERAL/ PROVINCIAL AGREEMENTS •OEINSTrrUTIONAUZATION PROJECT  1  DISTRICT SUPERVISORS -LA. • EJ.P.  -F.AC.S. • M.R. -INTEG. SERVS.  OnGSVOL*  AUDIT TEAM  21 workers are sometimes allocated to offices experiencing exceptionally high work demands.  Preparation for Research The researcher had over 10 years employment in child welfare with M.S.S.H. at the time of the research. He was therefore not only a researcher, but also a source of information about the topic. Additionally, he had conducted similar research in 1987 regarding the issue of factors which graduating B.S.W. students use in considering employment.  Preparation within the agency for the research began in 1986 with the initiation of research into factors which graduating B.S.W. students consider in selecting employment. This preparation included development of an instrument for that research. The development of the instrument made extensive use of focused discussions with line staff within the agency, supervisory staff, and practicum students in the agency. This established the issue of examining the issue of staff recruitment within the structure of a research base.  Subsequent to the completion of that research, results were sent to the deputy-minister of M.S.S.H., the Superintendent of Family and Child Service, M.S.S.H. Personnel Services, as well as some regional managers and district supervisors. The research included a number of recommendations, some of which were included in a subsequent re-organization of the ministry, and others of which eventually were found in subsequent plans for improving recruitment and retention strategies. This research and the dissemination of the results fostered recognition of the benefit of research into the related  22 area of job satisfaction. It is believed as well that the research provided some legitimation of the researcher in the minds of key decision-makers within the agency.  The next stage in the development of the research within the agency was the re-definition of recruitment and retention as an issue related to job satisfaction. This was undertaken in the autumn of 1988 with the submission of a draft of the research proposal to local management personnel, the Superintendent, and to the ministry research division for feedback. This was followed in January 1989 with the submission of the revised draft of the research proposal to the deputy-minister. Thus agency management personnel were kept advised of the development of the research proposal throughout all stages.  Line staff were not directly involved in the development of the topic or of the research instrument during this phase, although they had been involved in the development of the instrument in the earlier research. The content of the questionnaire was developed principally from earlier instruments as will be described in Chapter 4. This involved some fine tuning of the specific wording to ensure clarity of the questions, and inclusion of two additional variables.  Dissemination of Results Several methods have been developed for dissemination of the results of the research considering the parties who might be influenced by the results and recommendations and the possible broader interest in the results.  23  Preliminary results of the research were presented as a lecture and discussion at a ministry conference in Vancouver in January 1990. The researcher plans to advise all regions within the ministry of his availability to provide presentations of the full results after the completion of the thesis. This will provide the opportunity for dissemination of the research to line staff in the effected ministry.  The results will also be summarized and presented to the ministry's committee on staff recruitment and retention. This committee has been responsible for the development of initiatives to enhance employee recruitment and retention with particular emphasis on social workers. The results of the research will have particular importance for this committee. Additionally, a summary of the research will be forwarded to the head of the ministry's provincial personnel services division to ensure it is received at the highest levels within that division. As some of the recommendations flowing from the research will address practice issues, a copy of the summary will also be forwarded to the Superintendent of Family and Child Service. A full copy of the thesis will also be provided to the ministry library for access by all ministry staff.  The research will be brought to the attention of the B.C.G.E.U. through two routes. The summary will be forwarded to the provincial chairperson of Component 6, the component of the union which includes all ministry social workers. Some of the results and recommendations may have considerations within the context of the labour-management relationship. Some of the results may also be important for the union in considering which issues are important  24  to this group of its membership in the context of collective bargaining. An additional copy of the summary will be provided to the Research Department of the union for their information.  Copies of the summary will be sent to each district office which participated in the research to provide some tangible result to the participants. The summary will include the information that a copy of the thesis has been submitted to the ministry library.  The researcher will offer to provide a presentation at the University of British Columbia - School of Social Work Annual Research Symposium at the next possible date.  Finally, in order to promulgate the results into the broader social work community, articles will be submitted to the provincial social work publication, "Perspectives" and the national social work journal, "The Social Worker".  Successes and Limitations of the Collaborative Process The principal success of the collaborative process was that the research took place in an unfettered manner within an agency which had reason to be concerned about the outcome. It is believed that this is in part due to the extended length of time taken in developing the research plan, and the credibility established by the researcher over that time. The development of the research over a period of years and the initiation of the research outside of the ministry established to the agency that the intent of the research was not to develop some short-term goal for the purposes of the union. Rather,  25 the researcher's intent was clearly to address a long term concern of importance to both parties. Similarly, the union did not perceive the researcher to be someone hired by the agency to minimize a problem. He was perceived as a person who recognized the importance of the issue to union members and as having legitimacy in that regard.  The researcher's multiple roles in the collaborative process did have some drawbacks however. The researcher did not request access to management documents which might have been useful out of concern that those same documents might eventually come into the possession of the union. This possibility might have raised questions about the trustworthiness of the researcher. Examples of this data include staff turnover rates, absenteeism rates, and other confidential research which the ministry may have already conducted.  Similarly, a researcher who did not have the multiple roles might have been able to engage in more extensive collaboration with the union without risking having his independence from this party being challenged.  Summary This chapter has identified the agency in which the research took place and described its organization. It has described the role of the researcher within the agency and how the research plan was developed with the involvement of key players in the setting. A description was provided of a number of plans for dissemination of the results of the research to key players within the setting, to the research participants, and to the broader social work  26  community. Finally, some of the limitations of the collaborative process were described, with reference to the role of the researcher within the setting.  27  CHAPTER 3 THE RESEARCH PROBLEM  Theoretical Overview Just as job satisfaction has no clearly agreed upon definition, it has no clearly agreed upon dimensions. Two models of job satisfaction dominate research however, the needs satisfaction model and the value satisfaction model. The first is based upon perceived universal human needs, the other based upon perceived individual human values.  Needs Satisfaction Model The needs satisfaction model has been the predominant model is job satisfaction research. This model is based on Maslow's conceptualization of a hierarchy of human needs. At the first level, Maslow postulated that the most basic human needs are physiological needs such as those for food, water, etc. When these needs are satisfied, needs for safety, such as security, order, and protection become a more pressing priority. The higher levels of need are for social needs (e.g. acceptance, friendship, etc.), esteem needs (status, prestige, acknowledgement), and self-actualization needs (personal fulfilment, growth, etc.). As each level of need is satisfied, the next level of need comes to predominate. Lower levels of need are never extinguished however, and if the needs of a lower level are not met, they quickly come to predominate as the principle goal of behaviour. These needs were conceptualized as being stable and universal in nature.  28 The needs satisfaction model of job satisfaction presupposes that fulfilment of Maslow's hierarchy of needs is an important objective of work, and the important component in job satisfaction. Job satisfaction was seen to be a function of how nearly the components of a job contributed to fulfilment of all of the needs in Maslow's hierarchy. In this model, jobs which provided physiological, safety, social, esteem, and self-actualization would be the most satisfying, while those which failed to provide the lower levels would be the least satisfying.  While this model provides a useful framework, it nevertheless has some considerable limitations. Hopkins (pg. 20) summarizes the work of Salancik and Pfeffer in commenting, "The most important problem with the model arises from its dual assumptions that needs are relatively stable across individuals and across time and that jobs have fixed characteristics". The needs satisfaction model fails to take into account variables such as individual differences in growth needs and individual differences in sources of satisfaction for growth needs. Some individuals may obtain physiological and safety need through their work, while they obtain social and esteem needs through family and social networks unrelated to their job, and selfactualization needs though individual pursuits. Such a person may experience high job satisfaction in a job which meets only physiological and safety needs, while the needs satisfaction model would predict that such a job would result in low job satisfaction.  Similarly, a strict needs satisfaction model would fail to explain any job satisfaction among people in jobs which provided low satisfaction of  29 physiological or safety needs. Such jobs might include high risk fields such as fire-fighters, police, and military jobs. Yet employers in such fields of work typically do not experience significant difficulty in staff recruitment and retention.  The needs satisfaction model however, did provide a basis for identifying specific job elements for measuring overall job satisfaction however. It allowed for the development of need-specific questions to be asked in addition to non-specific questions regarding overall job satisfaction. Later developments in the needs satisfaction model recognized that individual needs might be variable and refinements included the concept of how important a given variable was to the individual respondent.  Value Satisfaction Model The value satisfaction model was proposed by Locke as an alternative to the needs satisfaction model. The value satisfaction model postulates that job satisfaction is a function of how nearly the explicit or implicit values of a job match the values of the individual. The nearer the match between job and individual, the greater the job satisfaction experienced. This model has the advantage of explaining varying levels of job satisfaction among people in the same occupation and work setting who objectively have the same human needs. The job satisfaction difference is explained by differing individual values resulting in a difference in the values satisfaction experienced.  This model of job satisfaction assumes inherent values in all jobs which are assumed to have level of importance to workers in that job. Although it  30  is not argued here that some jobs are absolutely value free, it appears evident that the values in some jobs are more explicit and intrinsic than in other jobs. Additionally, it appears that the inherent importance of the values associated with some jobs is greater than the inherent importance of the values in other jobs. We would expect religious clergy and ethicists to be more aware of the underlying values of their job and consider them to be more important aspects of the job than perhaps television weather broadcasters. If this is the case, it would seem that the values satisfaction model would be more useful in measuring job satisfaction among religious clergy and ethicists than among television weather broadcasters. Thus a conflict between the values of a clergyman and his employer would arguably result in a greater effect than a values conflict between a weatherman and his employer.  Job Satisfaction vs Dissatisfaction Many instruments designed to measure job satisfaction assume a range of responses from low satisfaction to high satisfaction, e.g. "All in all, how satisfied you would say you are with your job — very satisfied, somewhat satisfied, not too satisfied, or not at all satisfied?" (Quality of Employment Survey).  Herzberg recognized a difference between job satisfaction and job dissatisfaction and hypothesized that different factors influenced job satisfaction and job dissatisfaction. He provided a description of what he termed "motivators" and "hygiene" factors. Motivators are factors which influence job satisfaction, while hygiene factors influence job dissatisfaction. He identified motivators as being;  31  i)  getting a sense of achievement;  ii)  getting recognition for work performance;  iii)  being satisfied with the nature of the work;  iv)  having clearly defined responsibilities;  v)  having opportunities for advancement; and  vi)  having opportunities for personal growth.  Hygiene comprised 10 factors, the most important of which were;  i)  company policy and administration;  ii)  quality of supervision;  iii)  relationship with the supervisor;  iv)  working conditions;  v)  salary; and  vi)  relationships with peers.  Herzberg believed that changes in hygiene factors would result in changes in levels of job dissatisfaction but would have little impact on reported job satisfaction. Similarly he believed that changes in the motivation factors would result in changes in job satisfaction, but have little effect on levels of job dissatisfaction.  32 Questions for Research The guiding question for this research was, "How can overall job satisfaction among child protection line social workers be enhanced?". Given the context of the agency being studied, the employee group being studied, and the theoretical framework of previous research, a number of questions were selected for research, as follows:  1.  What is the level of job satisfaction or dissatisfaction within the group?  2.  What variables best correlate with differences in overall job satisfaction?  3.  What variables do respondents report as being important to them in their jobs?  4.  Are the variables reported in #2 and #3 the same or different?  5.  Are the reports of what is important and what correlates with overall job satisfaction constant with varying levels of work experience, or do they change?  The first question was selected in order to establish a basis for further questions. It was important for statistical purposes in order to establish some measurement against which other variables would be tested for co-variation. The method chosen for asking the question was important too for theoretical reasons. The method selected was to ask respondents to rate their satisfaction based on a scale with end points of "Very Dissatisfied" and "Very Satisfied". This reflects a recognition of Herzberg's work in defining job satisfaction  33  existing on a continuum that does not end at no satisfaction, but extends it into the range of actual job dissatisfaction.  The second question was selected in order to establish which specific elements of the job are the most important correlates of overall job satisfaction. Again, respondents would have the opportunity to report dissatisfaction as well as levels of satisfaction. This research question is the fundamental question flowing out of the guiding question. A number of variables were selected from previous research and additional variables were developed for this research. It was believed that although some elements of job satisfaction are transferrable across jobs, some elements would be specific to different jobs. Several variables were created based on elements perceived to be important to the workers in the agency which were not fairly represented in other studies. Attention was given to the inclusion of values satisfaction variables which related to agency recognition of social work as a profession, and agency support for professional development as well as the more usual needs satisfaction variables.  The third question recognizes that individual variables may be perceived as more or less important to different individuals. This was seen to have value in understanding what social workers perceive as important in employment separate from what satisfaction they experience in their jobs. This perception may be important in recruitment of social workers. In the recruitment process it is valuable to make explicit those job characteristics which are perceived as important, regardless of their correlation with job satisfaction.  34  The fourth question was seen as designed to provide information about the usefulness of developing plans for improving job satisfaction among the study group based on their reports of what is important to them. If findings were that reports of what was important to the group were very similar to the actual correlates of job satisfaction, it would support the argument that future changes in reports of what is important to the group might be indicative of what was important for improving job satisfaction at that time. If however there were few similarities between what was reported as being important and what actually correlated with job satisfaction there would be support for the argument that job satisfaction was not a function of meeting reported needs, rather it was a function of meeting some actual yet unreported needs.  The fifth issue relates to the assumption of constancy in needs as presented in strict needs satisfaction models of job satisfaction and the related issue of recruitment and retention as an issue of concern to employers. If needs remain constant then correlates of overall job satisfaction should remain constant for people in the same job with varying amounts of experience. If needs vary then correlates of overall job satisfaction should vary as well. This would have important implications for the issues of recruitment and retention of employees.  If the correlates of overall job satisfaction remain constant then recruitment and retention strategies can be unitary. If the correlates vary however, differing strategies would be required for these two goals. It may be necessary to modify jobs and benefits for employees with varying levels of  35 experience. Similarly, if reports of what is important in a job remain constant, recruitment strategies can be based on the promotion of a limited set of factors regarding the job. If reports of what is important vary as a function of years of experience, recruitment strategies may need to emphasize different aspects of the job in order to attract employees with varying levels of experience.  Note that the issue to be studied is not the tracing of the changes with a fixed sample over time. Rather a sample of social workers with varying lengths of experience will be selected for study at the same time. This crosssectional approach will test for overall job satisfaction and specific correlates of job satisfaction given the same external circumstances. A longitudinal study would better allow for qualitative analysis of the change process with a fixed group but would increase the impact of the variable of changes in the social policy environment in which social workers work.  Conceptual Model The conceptual model used in this research can be seen to be a combination of the models outlined above. The principal model used was the needs satisfaction model, although, as has been noted above, elements of the values satisfaction model were included. The model also recognized the continuation of satisfaction into the realm of dissatisfaction as identified by Herzberg.  The needs satisfaction model was selected as the predominant model for theoretical and practical reasons. First, the needs satisfaction model has a  36 well-developed base on which to build research. It has been the predominant model in job satisfaction research, hence has developed a number of measurement instruments on which to base further research. Despite the limitations of earlier conceptualizations noted above, the needs satisfaction model has reflected the recognition that individual needs vary and must be accounted for in research. Following from this, research based on the needs satisfaction model more readily allows for comparisons between research reports. Although complete comparisons are not possible where tools have been modified, there is greater opportunity for limited comparison between studies based on the same theoretical model than between model which use different conceptualizations while purporting to measure the same phenomena.  The 27 employment situation variables to be used have been drawn from a number of sources including Barber, and Jayaratne and Chess (1984,1986). The variables are listed in Appendix B. The variable of pay identified by both Barber and Jayaratne and Chess (1984) has been sub-divided into separate variables of PAY (amount of pay) and FRINGE (employment benefits). Supervision has been addressed as SUPER (quality of supervision) and CONSULT (access to consultation), and a variety of working conditions have been made explicit as PHYSIC (physical comfort), HOURS (number of hours worked), ACCESS (to the worksite), and FUNDS (access to funds and resources to achieve work goals). Many other variables are easily identifiable as being drawn from Barber or Jayaratne and Chess (1984), such as PROMO (opportunity for promotion), JOBSEC (job security), and COWORK (competence of coworkers). The variables PAPER (amount of paperwork), SPECIAL (opportunity for specialization), IMAGE (public image of agency), PRODEV (agency support  37  for professional development), VAC (amount of vacation time), and PROFESS (agency recognition of social work as a profession) were generated by staff of the Ministry of Social Services and Housing and were used in Gorrie. They were included here to allow for comparison between the results of that study and the current research. The variables SUPRESP (supervisor's respect for employees) and AGERESP (agency's respect for employees) were added for this research due to recommendations from ministry staff.  The needs model was important for practice reasons as well. A goal of the research was to make recommendations as to how job satisfaction among child welfare line social workers could be improved. This would presumably entail the modification of the job experience in some way. The needs satisfaction model is more disposed toward analysis of modifiable aspects of the job than is the value satisfaction model.  Recommendations flowing from research based on the value satisfaction model would tend to address the gap between the value base of the individual and the value base of the agency in order to improve that match, or the selection of employees or jobs which better meet the value satisfaction criteria. Although the value satisfaction model may be useful in addressing individual job satisfaction issues, e.g. in individual job counselling, it is not suited to organizational job satisfaction modification unless both employers and employees are willing to examine and modify their value bases. It was not believed that this criteria was met in the situation being studied.  38  The needs satisfaction model was selected then because it provided the strongest research base, provided the opportunity for some comparison of results, and provided a conceptualization which lent itself to recommendations which would be feasible to implement. The sole possible exceptions to the needs satisfaction model use were the variables of agency recognition of social work as a profession and agency support for professional development. The first of these was recognized as a value satisfaction variable (although it could be argued as an esteem needs variable) while the second was recognized as being possibly a value satisfaction variable (although it could be argued as a self-actualization/growth needs variable). These were included because although they were seen as value satisfaction variables, they were also seen as being easily modifiable by an employer, hence amenable to being included in recommendations.  Knowledge Building Function This research was designed both to provide practice recommendations to parties in the agency, and to build on the knowledge base of previous research. The perceived limitations to the existing knowledge base were seen as:  i)  the absence of Canadian research in the field,  ii)  the absence of research specific to child welfare social workers,  iii)  the absence of an instrument designed particularly for child welfare social workers, and  39 iv)  the absence of research which included length of experience with an employer as a control variable.  The research then was designed to include variables perceived as being specifically important to child welfare social workers as well as variables selected from existing research instruments which were more general in nature. It was also designed to include the variable of length of related work experience in the agency as a control variable. It was believed that development of the instrument could prove useful to other researchers and that inclusion of the control variable of length of experience with the agency might reveal a difference in the importance of variables. These differences might have been overlooked as being significant correlates of overall job satisfaction within the larger group, but could emerge as correlates for the groups based on length of experience with the specific agency. This might be true as well for variables which measure the reported importance of variables, as well as those which report correlations with overall job satisfaction.  The primary value of the research is to the employer where the research takes place. The results provide the employer an overview of overall job satisfaction among employees working in line practice positions. The research also provides information on what factors the employer should address to maximize job satisfaction among child welfare social workers. It provides information on how to maximize employee recruitment by modifying those employment situation factors which are most important to new employees. It also provides information on how to maximize employee retention by modifying  40  those employment situation factors which are most important to more experienced employees.  Some of the employment situation factors may be immediately modifiable by the employer and may be able to be addressed directly. Other factors, such as vacations, pay, or benefits may be the subject of collective bargaining or may be indirectly modifiable. These would be longer range areas for change.  The research may also provide the basis for comparative work, within and beyond British Columbia, in developing profiles of factors which are more or less important to social workers employed in various fields of social work practice with varying levels of experience.  Formal Hypotheses This research postulated six formal hypotheses:  i)  respondents will report job-specific variables as being more important to them than job variables not specific to social work;  ii)  less experienced workers will rate professional identification variables as more important (than will more experienced workers) and more experienced workers will rate tangible benefit variables are rated as more important (than will less experienced workers);  iii)  respondents will report greater satisfaction with variables  41  over which they have control or influence than variables which are outside of their control or influence; iv)  less experienced workers will report less satisfaction with variables related to professional identification (than will more experienced workers) and more experienced workers will report less satisfaction variables related to tangible benefits (than will less experienced workers);  v)  satisfaction with job-specific variables will correlate significantly with overall job satisfaction;  vi)  variables based on professional identification will correlate with facet-free job satisfaction for less experienced workers while variables based on tangible job benefits will correlate with facet-free job satisfaction for more experienced workers.  These hypotheses will be tested using an questionnaire developed for the research, based on existing research and questions specifically designed for the target group.  Relationship To Other Research Basing the research instrument on instruments used in previous research allows some comparison of the results with other studies. Most other research limits responses to questions regarding satisfaction levels to "Very Satisfied", "Somewhat Satisfied", "Not Too Satisfied" and "Not At All Satisfied". This does not allow responses which express dissatisfaction. The elimination of the latter two responses and the inclusion of possible responses of "Neither  42 Satisfied Nor Dissatisfied", "Somewhat Dissatisfied", and "Very Dissatisfied" limit that comparability however. It would appear that given that the first two possible responses are identical across instruments however, that there should be comparability in that range. The difference in instrument construction should permit respondents who previously would have indicated little or no satisfaction to now respond in the range of the three new responses. We would anticipate therefore some comparability in the higher satisfaction levels between this study and other studies.  Predominant in the studies which would allow comparability are the job satisfaction section of the Quality of Employment Survey (1979), Jayaratne and Chess (1984), and Barber. Opportunities for comparison of correlation between specific satisfaction factors and overall job satisfaction are more limited, in part due to the difference in measurement and in part due to the addition of variables in the instrument used. Comparison with the Quality of Employment Survey results would be of little interest given its generic nature. McNeely et al do not provide correlates based on specific occupations; instead they limit their reporting to correlates for all employees of the agency. Barber does report results for a number of specific variables, however these reflect all professional employees of an agency, rather than just child welfare social workers, or even all social work positions.  There is opportunity for some comparison with other research in the field of expressed importance of job factors within a similar occupational group. Gorrie reported on the importance of similar factors among graduating B.S.W. students within the same geographic area. This will allow some comparison in  43 the importance of those graduating from a school of social work, and those currently employed in child welfare social work.  Limitations Of The Research In addition to the fact that the research is confined to child welfare line social workers, it is further limited by its geographic scope, comparability with other research as discussed above, and limitations in available statistical analysis due to small sample size.  Perhaps the most important limitation for interpretation of the results is the defined limitation of the population. Barber has reported the difference in correlates of overall job satisfaction among different job categories within a single agency. McNeely et al have reported differences in correlates of overall job satisfaction within the same agency under different models of service delivery. Jayaratne and Chess (1984) have reported differences in correlates of overall job satisfaction among practitioners in different fields of social work. Quinn and Staines have reported differences in overall job satisfaction among national populations over time. There is considerable evidence that aspects of job satisfaction are variable over time, among job categories, among fields of practice within a job category, and between service delivery models. The results of this research therefore should be generalized with extreme caution.  In addition to the above noted limitations, it may be significant that the research was conducted within a single geographic region of the agency. Although the region selected includes urban and semi-rural areas, it may not be fully representative of either type of area specifically. It may not be  44  specifically representative of child welfare social workers who practice at a greater distance from large urban centres. It may not be specifically representative of social workers who work in an urban core.  A preferred route for testing the generalizability or lack thereof of the results would be to administer the instrument throughout all of the agency's offices. This would include urban core offices with six social workers covering a few square kilometres through to isolated rural offices which are responsible for thousands of square kilometres with only two line social workers. This administration might reveal important differences in job satisfaction correlates between urban and rural practice.  Despite a high response rate of approximately 80% the small population size precluded the use of advanced statistical analysis procedures. It was not possible to apply factor analysis which requires a minimum of 10 to 20 cases per variable. Factor analysis allows the identification of groups of variables which tend to co-vary and provides an opportunity for interpretation of these results to hypothesize underlying common factors. Instead, correlations were performed which allow the identification of individual variables which significantly correlated with overall job satisfaction. Although this has the limitation of not statistically identifying underlying factors, it allows the prescription of recommendations based on covariation with specific variables. It should be noted however that the independent variables will covary as well. Therefore the effect of interventions which change the score of one independent variable will be effected by the covariates of that independent variable.  45 CHAPTER 4 RESEARCH DESIGN  Researcher's Control Over Phenomena Studied The research conducted was based on a naturalistic model. Clearly an experimental laboratory model would have been inappropriate for an issue of job satisfaction. The alternative of using a field experimental model would be interesting and potentially very useful. It would require resources, authority, and finances far beyond those available to the researcher however. A field experimental model would presumably involve pre-testing two samples of a population for levels of job satisfaction, implementing strategies designed to enhance job satisfaction within the experimental sample and not within the control sample, and post-testing both sample groups. Furthermore, a field experimental model would have been inappropriate at this time. The research at this stage is designed to establish what job variables correlate with job satisfaction among child protection line social workers within a given agency at a given time. Once this is established, it may be appropriate to design a field experimental model which would allow for testing of the results of the correlational study.  Sampling Design The unit of domain in this study was the individual social worker. As noted earlier, the Ministry of Social Services and Housing employs social workers in the fields of child protection, services for mentally handicapped, and resource development and maintenance for each of these two areas. It also employs non-social workers in the fields of income security and child  46 maintenance payment enforcement programs. Within this context, the group selected for study was homogeneous in that it was restricted to line child protection social workers. Within that category of employees however, the group was heterogenous, in that there was no selection based on gender, marital status, age, or any other demographic variable. The probability of inclusion in the research was based on the factors of availability at work during the time period of the research and self selection.  Timing of the Data Collection The intent of the research was to collect data at a time which was likely to represent a normal state of affairs within the current context of the target group, i.e. there was no design element which required the data collection to take place before or after an event in order to study the effect of the event. Thus data collection was at an undifferentiated time. It is recognised that many large organisations are constantly in some type of change, either localized or involving greater parts of the system. This research was undifferentiated in that it was not designed to measure changes in variables over time or as a function of some event.  Some attention was paid to the timing of the event with regards to events external to the agency however. The period of availability of the data collection instrument varied slightly, but generally was for one week during the third week of June 1989. This period did not include any religious or cultural events likely to be significant to the target group which might result in undue bias in the sample. It was also prior to the end of the public school year, so would not likely have resulted in the exclusion of workers with  47 children who may be more likely to be absent due to vacation during the school break.  Choice of Methodological Orientation The methodological orientation selected was quantitative data collection and analysis. There was an attempt to incorporate a qualitative element by the inclusion of a section for open-ended comments as Part 6 of the questionnaire used. This proved to be little used by respondents and the few comments made have not been included in the analysis.  The quantitative methodology was chosen for a number of reasons. First, there is a history of quantitative methodology in this research area. Previous research tools include the quantitative Index of Job Satisfaction, the Job Descriptive Index, and the Job Satisfaction Survey Questionnaire. Second, many of the possible variables regarding job satisfaction have already been at least partly identified. Qualitative methodologies are typically superior in proposing variables for studying of a problem, however that task has already been accomplished in this field. Finally, quantitative methodology is more useful in convincing large agencies of the need for action, change or remediation. Qualitative research can prove effective in some situations, particularly where research on a topic benefits from "humanization" of the issue. It was believed that in the large government ministry involved, there was more benefit in quantifying the extent of the problem and in producing results which lend themselves to data manipulation and analysis.  48 Data Collection Interviews and questionnaires were explored as the most likely methods for data collection. The major advantages and disadvantages of interviews and mailed questionnaires are presented in Tables 4.1 and 4.2 below (modified from Bailey).  Most of the disadvantages of mailed questionnaires were not important in this research. The variables to be included for measurement were predetermined and as the research was quantitative, capture of non-verbal responses was less important. Response rate was expected to be high due to the importance of the issue for possible respondents, and the respondents were presumed literate. Research was to take place in a relatively controlled situation and time frame, and there were no critical considerations regarding the ordering of the questions. There were no important factors regarding a specific date on which the data was to be collected. The advantages of lower costs, a compressed time line, anonymity, convenience for respondents, and standardization of questions were considered to be very important.  49 Table 4.1 Advantages Of Interviews vs Questionnaires ADVANTAGES OF INTERVIEWS  DISADVANTAGES OF QUESTIONNAIRES  1. Flexibility of the interview, modifications can be made when appropriate,  1. Loss of flexibility, everything is pre-determined,  2. Typically high response rate among those contacted, literacy not a problem,  2. Typically a much lower response rate, literacy may be a problem,  3. Non-verbal behaviour observable,  3. Verbal responses only,  4. Controlled environment possible,  4. Little control over the research environment,  5. Interviewer controls the order of questions,  5. No control over the order in which questions are presented,  6. Interviewer can record spontaneous answers,  6. Responses can be filtered or edited prior to returning the questionnaire,  7. Respondent cannot consult other people or records prior to responding, which provides confidence in who the response really came from,  7. Responses may be unduly influenced by other individuals or sources of information,  8. Allows the opportunity to ensure the completeness of answers (probing),  8. No ability to probe or clarify responses,  9. Clear information on when the responses were generated,  9. No control over date of responses,  10. Allows the development of a more complex questionnaire to be administered by skilled questioners.  10. Simpler questionnaire usually required.  50 Table 4.2 Disadvantages Of Interviews vs Questionnaires DISADVANTAGES OF INTERVIEWS ADVANTAGES OF QUESTIONNAIRES 1. High cost,  1. Much lower cost per response,  2. Extended time needed for each interview, transcription, etc.,  2. Time line from starting data collection to receiving data is typically much shorter,  3. Possible interview bias, (perception by respondent of interviewer bias, influence of interviewer's social influence on respondent, or misunderstanding of response)  3. Less personal contact reduces a source of real or perceived interviewer bias,  4. No opportunity for respondent to consult records or reflect on the issues,  4. Respondent able to give extensive thought to questions before responding,  5. Possible inconvenience to respondent (e.g. scheduling of interview),  5. Respondent can complete the questionnaire at a convenient time, location, etc.,  6. Face to face contact reduces both real and perceived anonymity,  6. Real and perceived anonymity easier to achieve,  7. Reduced standardization of wording of questions may introduce differences in responses based on questions rather than other variables,  7. Identical questions presented to all respondents reduces the number of variables influencing responses,  8. Reduced accessibility to respondents due to geographic location,  8. Geography need not be a significant limit on sample size or selection.  51 Confidentiality Provisions Confidentiality can be an important consideration when conducting research within an employment setting. It was even more important in this situation, as will be described below, to the extent that anonymity would have been preferable. Anonymity carries with it a price tag however, in that it eliminates the opportunity for the researcher to contact the same respondents at a later date in order to conduct follow-up research.  This research required more stringent assurance of confidentiality than usual due to the role of the researcher. The researcher was known as a longtime social worker within the region being sampled, as a labour union activist, and a person who is interested in being in a supervisory position. All of these factors emphasized the need for anonymity for respondents. It was reasonable to believe that some potential respondents would be reluctant to participate fully and honestly or might be influenced to respond with what they believed to be the "right" labour union answers, the "right" answers when answering a possible future supervisor, or the "right" answers when answering a co-worker. It would be extremely difficult to determine the effect of the interviewer in the research. Even if there were no strong effect, the validity of the results could be challenged on these grounds. On the other hand, the researcher wanted to keep open the option of resampling the respondents at a later date to determine what changes had taken place within the same sample group.  A complex method was developed therefore to maximize confidentiality, while at the same time providing the researcher the opportunity to contact the  52 participants. The administrative methodology may prove useful to other researchers, so it is included here.  First, several documents were created as can be found in Appendix C. Packages consisting of one copy of documents ii) through v) were randomly placed into sealed envelopes, along with one copy of document i). The envelopes were then randomly divided into two bundles and delivered to each of the two area offices within the target region. One staff member from each district office then selected one sealed envelope of package for use at their office. The documents were as follows:  i)  Instructions To Be Read To Participants,  ii)  Introduction and Instruction Sheet,  iii)  Dear Participant letter,  iv)  Control Data Sheet,  v)  Questionnaire (Appendix A).  The "Instructions" sheet was to be read to all social workers at their own worksites at a regularly scheduled staff meeting. No additional meetings were scheduled for participation in the research. Workers were provided time within the context of their usual meetings to participate in the research without taking additional time from their work days or private lives. This was considered an important factor in ensuring a high response rate. Social workers had already allocated that time, so use of the time to complete the research questionnaire was less likely to be seen as taking something away  53  from them. The reading of the "Instructions" also provided visible confirmation of the support of the agency for the research.  The "Introduction" sheet provided potential respondents with an explanation of the security system which had been built into the data collection methodology. Each sealed package contained a pre-numbered Control Data Sheet, along with a pre-numbered questionnaire. Participants completed the Control Data Sheet by writing their name and current address. They also provided the name and address of another person who would likely be able to contact them at a later date in the event they are no longer at that address at the time of subsequent research. They then proceeded to complete the questionnaire. They then forwarded only the Control Data Sheet to a lawyer contracted by the researcher to receive these documents. They returned only the questionnaire to the researcher for data purposes. The lawyer is prohibited by contract from releasing the Control Data Sheet and the contents or the information thereon to any person with the sole possible exception of the university research ethics committee.  The "Dear Participant" letter was provided as further confirmation of the legal arrangement between the researcher and the lawyer.  In the event that further questionnaire research is conducted with the same sample group the researcher will forward the research tool to the lawyer with instructions as to which Control Data Sheet numbers are to be included in the subsequent research. This may include all or a subset of participants in the current research. The lawyer will then contact the participants either  54 through their main address provided on the Control Data Sheet or through the alternative address which they provided.  The Questionnaire Several questionnaires exist for researching issues of job satisfaction. However, after reviewing several questionnaires the researcher chose to design an instrument specifically for this research. The questionnaires reviewed included the Job Satisfaction Survey Questionnaire, as used by Burstein et all; the Job Descriptive Index, as developed by Smith; the Quality of Employment Survey (1977 edition), developed by Quinn et al; and the Professional Satisfaction Inventory, used by Jayaratne and Chess (1984,1986). The advantages and disadvantages of each is discussed briefly below.  The Job Satisfaction Survey Questionnaire This instrument was developed for Canadian research from the American developed Survey of Working Conditions, also known as the Quality of Employment Survey. It provided the advantage of allowing comparability between the results of the current research with other Canadian research. It was also relatively brief. Although there were 80 questions, 34 were repeated twice, once regarding the importance of a variable and once regarding its presence in the respondents current job.  This tool included an important disadvantage for the current research however. The questionnaire was designed, "for administration to diverse occupational groups" (Burstein, pg. 15.). This makes it useful for broadly based  55  research but does not include variables which may be applicable for specialized employment.  It had the additional disadvantage of not directly questioning respondents regarding their job satisfaction i.e. job satisfaction was presumed based on a priori judgements regarding what was satisfying. For example, one variable in the section regarding job satisfaction provides the statement, "I am given a lot of chances to make friends" and the respondents rate their job on a four point scale between "Not At All True" and "Very True". A response of "Very True" is considered more satisfying than a response of "Not At All True". This may be true for some people, but it may be that some people are more satisfied by a job which does not include that aspect. Two people may respond with the same answer, but mean completely different things by their answers.  The Job Descriptive Index (JDI) This tool has considerable advantages in ease of administration, applicability to people with limited competency in written English, and the availability of normative scores. It is a five page, 72 item checklist with limited responses of Y(es), N(o), or ? (unsure). Each page is devoted to eliciting one global job satisfaction factor in relation to Work, Supervision, Pay, Promotions, or Co-Workers. Most job descriptive items are one to three words. For example, in the section dealing with Supervision some items (here provided with the response indicating a satisfying job) are:  56  Asks my advice (Y) Up-to-date (Y) Quick tempered (N)  The JDI can be seen to share the disadvantages of the Job Satisfaction Survey Questionnaire for the purposes of this research. It is appropriate for administration in a wide range of jobs where specific data regarding the variables of that job are not needed, but is not useful in identifying variables of importance to a limited group. It can also be seen to contain elements of the a priori issue identified above.  The Quality of Employment Survey This tool has the advantages of having been used, in several revised forms, in repeated research based on nationally stratified sample throughout the contiguous American states. This makes it is possible to compare results with national and occupational norms in a nearby country. This advantage is limited by some of the cultural differences which arguably might have an important effect on job satisfaction or correlates of job satisfaction. Examples of this are the different rates of labour union membership, the differences in health care systems, and the differences in the context and forms of discrimination between the two countries.  A further limitation of the Quality of Employment Survey was that it was designed to be administered as a structured interview, rather than as a questionnaire. As already discussed, a questionnaire methodology was deemed most appropriate for this research. Additionally, like many other studies the  57 Quality of Employment Survey was designed to, "be meaningful to all workers in a national survey". (Quinn and Staines pg. 206). Quinn and Staines complained that a problem with other measures of job satisfaction was that they were "developed on relatively homogeneous populations of workers and their wordings were therefore too occupation specific", (ibid. pg. 206) Yet this limitation for Quinn and Staines was precisely the type of job specific tool which was desirable for the current research.  Additionally, the Quality of Employment Survey consists of 91 pages of data collection, including the interviewer's observations, hardly a small piece of work for either the interviewer or the respondent.  The Professional Satisfaction Inventory This tool was developed to study relationships between job satisfaction and job stress among social workers. The job satisfaction section of it, part 27, was based on the Job Satisfaction Survey Questionnaire described above. It did however make attempts to identify issues of particular concern to social workers. It also shared some of the problems of the Job Satisfaction Survey Questionnaire in that it was based on a priori questions.  The Social Work Job Satisfaction Questionnaire The questionnaire used was developed from the research instrument developed by the researcher in "Employment Preferences of Graduating B.S.W. Students" (Gorrie). That instrument had been based in part on the Job Satisfaction Survey Questionnaire, interviews with recent graduates from a B.S.W. program, and current students in a B.S.W. program. The questionnaire  58 was designed to allow comparability between the responses of graduating B.S.W. students and practising social workers in identifying the importance of specific variables in employment. The questionnaire can be found in Appendix A.  Part 1 of the questionnaire identified demographic variables for use as possible control variable in analysis.  Part 2 of the questionnaire identified a number of employment situation variables and asked the individual to rate how important each of the variables was to them. Some of these variables have been taken from Gorrie's 1987 questionnaire.  Part 3 again identified the variables and asked respondents to rate their level of satisfaction with that variable in their current employment situation. Included in this section was the question, "All in all, how satisfied would you say you are with your job?" from Quinn and Shepard. Note that the problem of a priori questions has been avoided in the questions dealing with job satisfaction. Each of the questions is "non-directional" in nature. That is, they do not present items as being desirable or undesirable. Rather than having respondents agree or disagree with the statement, they rated how satisfied they were with each of the dimensions of their job, i.e. "How satisfied or dissatisfied are you with each of the following (aspects of your employment situation)". Table 4.3 provides a comparison of the "directional" wordings found in the Job Satisfaction Questionnaire" and the "non-directional" wordings for the same items in the Social Work Job Satisfaction Questionnaire.  59 Table 4.3 Examples of Directional vs Non-Directional Questions Directional Question  Non-Directional Question  k) the pay is good  51. The amount you are paid.  o) my supervisor is competent in doing (his/her) job  52. The quality of supervision you receive.  p) my responsibilities are clearly defined  Sll. How clearly defined your job responsibilities are.  g) I am not asked to do excessive amounts of work  S15. Your caseload/workload size.  f) I receive enough help and equipment to get the job done  S18. The access you have to funds and resources to achieve client related job goals.  The full list of variables can be found in Appendix B. Variables which use the prefix F (e.g. FPAY - importance of being well paid) indicate an employment preference variable while variables which use the prefix S (e.g. SPAY - satisfaction with amount paid) indicate a job satisfaction variable.  In addition to the variables which were developed from the literature, the variables of PAPER (amount of paperwork), SPECIAL (opportunity for practice specialization), IMAGE (the public image of the agency), PRODEV (agency support for professional development and ongoing training), VAC (amount of vacation time), and PROFESS (agency recognition of social work as a profession) were added in 1987 through consultation with B.S.W. students, recent B.S.W. graduates, and agency staff. The variables SUPRESP (supervisor treats employees with respect), and AGERESP (agency treats employees with respect) were added in 1989 due to recommendations from agency staff.  60 The variables were all rated using a 10 point interval scale. The F scale included ordinal-like identifiers of "Very Important", "Somewhat Important", "Moderately Important", "Quite Important", and "Very Important" and the satisfaction scale included ordinal-like identifiers of "Very Dissatisfied", "Somewhat Dissatisfied", "Neither Satisfied Nor Dissatisfied", "Somewhat Satisfied", and "Very Satisfied" to assist respondents in understanding the meaning of their responses.  Part 4 dealt with the issue of desire to remain in the current employment situation. This was done by asking respondents their preference for employment in a number of employment situations, including their current situation.  Part 5 dealt with the issue of intent to change jobs in the same manner. Respondents were given the same selection of employment situations, and asked to indicate where they realistically expected to be employed in two years time. Their current employment situation was one of the possible selections.  The method applied in sections four and five was used to determine which alternative employment situations were preferred by child protection social workers. If certain employment situations appeared to be heavily favoured, analysis of those employment situations could offer some clues as to other employment situation factors which may be important.  Part 6 provided an opportunity for additional feedback, observations, or  61 reports about the job satisfaction of the respondents. As noted above, it did not generate many responses, so it has not been included in the analysis.  Validity, Reliability, And Pre-Testing Monette et al define validity as, "the accuracy of a measure" (pg. 96) and reliability as, "a measure's ability to yield consistent results each time it is applied" (pg. 99). They advise that methods of establishing validity include content validity, criterion validity, and construct validity. They further advise that reliability can be established using test-retest procedures, multiple forms, and split-half approaches.  Establishing validity of a measure is more difficult when the measure assesses one variable in order to infer the measurement of some other variable. For example, measuring a person's age, may be used to infer a measurement of the person's shoe size. This measurement might have good validity from birth to early adulthood. A measure of a person's shoe size to infer a measurement of the person's weight would likely be much less valid. Establishing validity is less difficult when the measure is being used to directly assess the variable being measured. For example, a measurement of nationality of passport held is likely to be a highly valid measure of nationality.  Measures of personal opinion or belief, as in this research, likely fall between these extremes. Although they may measure a belief or opinion directly, clarity of wording of the question may be of considerable importance to avoid improperly inferring a secondary measurement. For example, the  62 researcher previously included in a questionnaire a variable which asked how important it was to have a union represent the respondent in contract negotiations. The intent of the question was to measure the importance of a collective bargaining structure. The question however, proved to be unusable for such a measure since it inferred that a union was the only method of collective bargaining. Within the particular political jurisdiction involved the option of a "professional association" as a collective bargaining agent was a possibility. Therefore, by exclusively identifying "union" with contract negotiations, the question limited the respondents' range of answers and this undermined the validity of the question.  Content validity was the method used in establishing the validity of the current instrument. Content validity can be ascertained either by the weaker individual method or the stronger jury opinion method. In the jury opinion method, the logical connection between the question posed and the underlying variable supposedly being measured is developed by the researcher. This logical connection is then confirmed or disconfirmed by others who are presumed to have worthy opinions regarding the matter. These others may be other academics, practitioners, or members of the population on which the instrument will be used. This is often included as part of the pretesting procedure in developing an instrument.  The questionnaire used was based on Gorrie's 1987 questionnaire. The validity of the questions in that instrument were established using the jury method of content validity. The instrument was pre-tested extensively, including 14 revisions of the wording of questions, the addition and deletion of  63 questions, and modifications to the measurement scale for responses. The pretesting was conducted using current B.S.W. students, recent graduates, and experienced child welfare line social workers. The current instrument added the satisfaction variables by only slightly modifying the wording of the questions.  The reliability of the measures has not been established through any formal procedures. Monette et al (1986) state that, "In general, a valid measure is a reliable one. So if we were certain of the validity of a measure, we would not need to concern ourselves with its reliability." (pg. 99). It is not argued here that the instrument has complete formal validity or complete reliability. Indeed, further testing of the instrument, through a test-retest procedure to establish its reliability would be a desirable continuation of the research. It is believed that the extensive pre-testing of the instrument both in its 1987 form and its current form argues for some confidence in its validity and its reliability.  Data Analysis The data generated were of two types; nominal (GENDER, ASSOC, ATTEMPT, BELIEVE, and the six variables identifying educational level), and interval (AGE, TOTEX, TOTGOV, HRSWRK, HRSREAS, DEPEND, and the importance variables, satisfaction variables, employment preference, and employment expectation variables). Data analyzed will be presented in tables and graphs along with descriptive paragraphs.  64 The questionnaires were pre-coded so little coding was required subsequent to data collection. There were a few instances of a respondent circling the ordinal-like descriptive. The decision was made to include those responses and code them as being half way between the two interval rating on either side.  Data Analysis Methods Job Satisfaction Variables The first aspect of data analysis consisted of obtaining descriptive data regarding the variables to observe any noticeably deviant variables.  Particular attention in the descriptives was paid to the variable SOVER (overall job satisfaction) which provided a facet-free measurement of job satisfaction. A facet-free measure is one which is independent of the job specific measures in contrast to measures of job satisfaction which are calculated from measures of satisfaction with individual job variables. This information was of limited value in itself in describing the actual satisfaction levels in that it provided a mathematical mean rather than a report of the types of experiences. For example, if in Case A half of the respondents reported SOVER (facet-free job satisfaction) of 1 and half reported SOVER of 10, and in Case B half of the respondents reported SOVER of 5.0 and half reported SOVER of 6.0 the means would be 5.5 for both cases. The job satisfaction experiences of the groups involved would be very different however. The value of the descriptive data regarding SOVER (facet-free job satisfaction) was that it provided confirmation that the distribution fell within normal parameters, and therefore was amenable to further parametric tests.  65  To interpret the mean, one needs to know the shape of the distribution, the range, and the standard deviation. The best descriptive device would include these measures as well as a graph of the distribution.  The alternative of recoding the importance and satisfaction variables into the ordinal-like classifications provided a more useful description of the job satisfaction of the research sample so they were produced as well to describe the sample.  Correlations between facet-free job satisfaction (SOVER) and the balance of the satisfaction variables were performed to determine which satisfaction variables best correlated to overall job satisfaction.  Finally, TOTGOV (years of experience with the agency) was recoded to create two groups, those with 3 year experience or less and those with more than three years experience. This provided groups of approximately equal size, and appeared to be useful in terms of "less experienced" workers vs "more experienced" workers. T-tests were then performed on all satisfaction variables using these two groups to determine first, if the means for facet-free job satisfaction (SOVER) were significantly different for the two groups and second, if means were significantly different for other satisfaction variables.  66 Inferences To Be Drawn From The Data Referring back to the section Data Analysis Methods, this section will address the inferences that may be drawn from the data analysis performed.  The description of the data was undertaken to allow for assessment of unusual results if they were found to exist. These apparently unusual results would then be viewed more closely to determine if they were questionable. Examples of this would be a satisfaction variable which had a mean value substantially higher or lower than the means for the other satisfaction variables or a variable displaying a disproportionate number of missing data responses.  Recoding of the importance and satisfaction variables to correspond with the ordinal-like classifications allowed closer comparison of the data with other surveys which provided fewer possible responses. This allowed inferences to be drawn regarding job satisfaction in the target group with results in other places or with other job groupings where similar variables existed. It also allowed for a more manageable display of data for ease of understanding of the results. This allows for easier inference of which F variables are considered most important to the respondents. It also identifies which satisfaction variables should be given closer scrutiny as importance variables which are rated as more important deserve greater attention.  T-tests allow the researcher to infer whether some other variable (the grouping variable) significantly differentiates between means. If there were significant difference between means of many variables one could conclude the  67 data deserved treatment as having come from distinct groups and that further analysis should proceed differentially based on the identified groups. For example, if it were found that T-tests indicated significant differences in means of many variables based on GENDER it would indicate that the data should be analyzed on the basis of GENDER.  Significant correlations between variables allows the inference that changes in one variable will be reflected in changes in the other variable. Significant correlations do not prove causality. Causality is proven experimentally by demonstration that an intervention or event resulted in a given change in a variable. This typically requires pre-tests, post-tests, and control and experimental groups. Much social science experimental research is not possible because of ethical considerations or difficulties with experimental methodology. Correlations are commonly accepted as providing inferential indication of a causal relationship in social science research. Significant correlations between facet-free job satisfaction (SOVER) and other satisfaction variables in this research allowed the argument that interventions which affected another satisfaction variable would result in changes in the facet-free measure of job satisfaction (SOVER). Each of the satisfaction variables was considered a possible element of or contributing factor toward facet-free job satisfaction (SOVER). It was further believed that as the other satisfaction variables were more amenable to direct intervention than SOVER, this provided an appropriate framework for intervention.  T-tests of the satisfaction variables using years of experience with the agency (TOTGOV) as the grouping variable, if they indicated significant  68  differences, would allow the inference that less experienced workers report significantly different satisfaction in those dimensions than do workers with greater experience. If T-tests indicated many variables with significant differences it would argue for differential analysis.  Ethical Issues In The Research Issues regarding confidentiality and anonymity of the research have been discussed above. There were additional considerations for the researcher, however. The first issue was a conflict between maximizing the usefulness of the results and maximizing confidentiality or anonymity. The researcher would have preferred to have had a method of identifying questionnaires by office of origin. This would have allowed the writing of individual office reports with specific recommendations for supervisory personnel. This identification would have invalidated all other attempts at confidentiality however. District offices typically have five to seven social workers so it would have been easy to identify individuals based on demographics and office. Additionally, the success of the implementation of the research plan depended on the cooperation and support of district office supervisory personnel. The researcher wanted to avoid any concern on the part of supervisors that results of the satisfaction variables would have any adverse effect on them individually. Thus the decision was made to maximize confidentiality and accept the loss of ability to provide feedback to individual offices.  The researcher had another concern which was dealt with by instructions to the participants. This was to avoid any implication that a decision not to participate in the research was improper. For that reason the Introduction  69  and Instruction sheet asked potential respondents who chose not to participate to return the documents without any identification or marks, as if they had completed the questionnaire. This allowed potential respondents to not participate, but not be identified as not having participated. Additionally, both the instructions read to potential participants and the Introduction and Instruction Sheet which potential participants received reiterated that they were under no obligation to participate.  70 CHAPTER 5 RESEARCH FINDINGS  Introduction This chapter will begin by providing an overview of the demographics of the respondents. It will then review the results of the data analysis plan laid out in the previous chapter. It will be of interest to readers who wish information about the outcomes of specific analysis procedures. Readers whose main concern is in the practice implications of these results will find Chapter 6, Implications and Recommendations, of more interest. This chapter will report the results of the data analysis in the context of the following five goals, to:  i)  establish what specific items respondents report as being important to them regarding job satisfaction;  ii)  establish their satisfaction level with each of the items;  iii)  establish in which items there are the greatest discrepancies between importance and satisfaction;  iv)  establish which specific variables best correlate with overall job satisfaction, and;  v)  establish whether there are differences in the ratings of the above between respondents with less experience and those with more experience.  71 Description Of The Sample Approximately 60 line social child welfare social workers were potentially in the sample. There were 49 questionnaires returned for a response rate of approximately 81.6%.  Appendix D provides graphs of the demographic data collected. Most of the respondents were aged 25 to 44 (85.4%), and females comprised over two thirds of respondents (68.8%). The number of economic dependents ranged from zero to 4, with more than half (53.1%) of the respondents reporting no more than one other person being "wholly or significantly dependant" on their income. One third of respondents report voluntary membership in the B.C. Association of Social Workers, or an equivalent professional association for social workers.  More than one-half of respondents (53.1%) had three years experience or less with the agency and more than one-half (55.1%) had a total of four years or less child welfare experience with any employer. The employer had not significantly expanded programs during the previous three years, which might have explained the large number of workers with little experience.  Almost one-half of respondents (46.9%) reported having made "real efforts during the past 12 months to find other employment". There was no significant difference on this variable between respondents with less than three years experience and those more than three years experience (Chi Square = 0.02881, p = 0.8652). Over one third of respondents (34.7%) did not believe they would, "be working in this agency two years from now".  72  In general then, respondents were lacking extensive experience in child welfare. They were fairly young and had few economic dependents. Most were female. Many had made efforts to change employers during the previous year. A sizable minority did not believe they would be working in the agency in two years.  Descriptive Data F (Importance) Variables Appendix E provides the listings of the descriptive data for the F variables. All of the descriptives reflect distributions within acceptable definitions of the normal curve except for the variable FACCOMP (getting a sense of accomplishment from the job) which is reproduced in Table 5.1.  Table 5.1 Descriptive Statistics Regarding Getting A Sense Of Accomplishment Variable FACCOMP Mean Kurtosis Skewness Range Maximum  Importance of Getting A Sense of Accomplishment  8.510 10.344 -2.564 9.000 10.00  Valid Observations - 49  StdDev S.E.Kurt S.E.Skew Minimum  1.566 .668 .340 1.00  Missing Observations - 0  We can see that the kurtosis of this variable is 10.344, indicating that the distribution is much more narrow and taller than a normal distribution. Table 5.2 and Graph 5.1 make this more understandable. We can see that fully 93.9% of respondents report that getting a sense of accomplishment from the  73 job is either Quite Important or Very Important. This information provides two implications. First that FACCOMP (getting a sense of accomplishment from the job) is not appropriate for use in further analysis which requires that variables conform to the normal curve. Second the high mean and high kurtosis indicate that FACCOMP is consistently an important variable for the sample.  Table 5.2 Importance Of Getting A Sense Of Accomplishment From The Job Value Label  Value  NI SI MI QI VI  1.00 2.00 3.00 4.00 5.00 TOTAL  Percent  Valid Percent  Cum Percent  1 0 2 18 28  2.0 0.0 4.1 36.7 57.1  2.0 0.0 4.1 36.7 57.1  2.0 2.0 6.1 42.9 100.0  49  100.0  100.0  Frequency  NI: (Not Important); SI: (Somewhat Important); MI: (Moderately Important); QI: Quite Important; VI: (Very Important)  74 Graph 5.1 Importance Of Getting A Sense Of Accomplishment From The Job Nl 1.00^1 SI 2.0010 MI 3.00 BBS 2 QI 4.00 ^^^gggggggggggggggggggggggggg 28 VI 5.00! Valid Cases  49  Missing Cases  :28 0  Nl: (Not Important); SI: (Somewhat Important); MI: (Moderately Important); QI: Quite Important; VI: (Very Important)  Appendix F provides a listing of the means and standard deviations of the F variables listed in order of means. The first noticeable observation about this list is the apparent discrepancy between the two highest rated variables, FSUPRESP (importance of supervisor treating employees with respect) and FAGERESP (importance of agency treating employees with respect), and the lowest rated variable, FSTATUS (importance of having a position with high status within the agency). Intuitively, one might have believed these three variables to be measuring substantially the same phenomena, that of the level of importance given to respondents within the agency. FSUPRESP and FAGERESP would relate to the respect given by those specific parties, while FSTATUS would be a more general measure of the recognition given to the position.  One hypothesis for the discrepancy between these variables is that FSUPRESP and FAGERESP measure the phenomena as it relates to specific parties who are very important to respondents, while FSTATUS measures the phenomena as it relates to parties whose opinions are less important to the  75  respondents. If this was the case, it would appear that respondents are very concerned about respect from organizational superiors, but not concerned about respect from others within the agency.  An alternative hypothesis is that child welfare line social workers view respect and status as very different phenomena. This would be consistent with a definition of status as being preferred treatment contrasted with respect being defined more as a fundamental issue of human dignity.  Finally, it may be that while child welfare line social workers view some of the elements of status as being important, they also view some of the elements as being either unimportant, or undesirable. This would explain the rejection of FSTATUS with the retention of FSUPRESP and FAGERESP as important variables.  Appendix G shows graphs of the importance given to the F variables after recoding into the five ordinal-like terms. It shows that most of the F variables are rated by most respondents as being moderately to very important. This was expected, as the variables were selected to represent the issues most likely to be of importance with regards to job satisfaction. Noticeable differences to this are the variables FPAPER (importance of having little job related paperwork), FIMAGE (importance of working for an agency which has a positive public image), FACCESS (having easy access to the workplace), and FSTATUS (having a position with high status within the agency). Each of these variables displays a more nearly flat distribution, reflecting a wider distribution of importance scores for these variables.  76  S (Satisfaction) Variables Appendices H and I provide tables and graphs of the distributions of the satisfaction variables. The tables confirm the appropriateness of the data for inclusion in further statistical analysis. The graphs demonstrate a tendency for satisfaction scores to be more widely distributed than importance scores were. This too is not unexpected. While on average respondents objectively experienced similar situations, these experiences were interpreted individually resulting in reports of different levels of satisfaction, satisfaction variable scores tend to be more nearly rated around the mid-point as well while F variables tended toward the extreme scores. Notable differences to the trend of more centralized distributions in these graphs are SSUPRESP (satisfaction with having a supervisor who treats employees with respect) which is loaded toward the high end of scores and SPRODEV (satisfaction with agency support for professional development), SAGERESP (satisfaction with having an agency which treats employees with respect), and SSWKSIZE (satisfaction with workload size), which are loaded toward the low end of the scores.  We can see from the descriptive report and graphs that there is also a much wider distribution in the means of the reports of satisfaction levels. Respondents report high levels of satisfaction with respect from supervisors and quality of supervision (SSUPRESP, M=7.74; SSUPER, M=6.87), and coworkers (SCOWORK, M=7.07), while reporting low levels of satisfaction with issues relating to agency support for professional development, agency respect for employees, and the public image of the agency, (SPRODEV, M=3.44;  77  SAGERESP, M=3.52; SIMAGE, M=3.54), workload size and amount of paperwork, (SWKSIZE, M=3.68; SPAPER, M=3.99), and access to funds for work purposes and for pay (SFUNDS, M=3.80; SPAY, M=4.28).  D (Discrepancy) Variables These two sets of descriptive statistics provide an overview of what is the respondents report as important and what they report as their satisfaction levels with the variables. An additional set of variables was computed to provide discrepancy variables. These variables are the difference between the importance score and the satisfaction score of each variable (FVARIABLE - SVARIABLE = DVARIABLE). Variables which had a low importance score but high satisfaction score would result in a minus discrepancy score (e.g. 1 - 10 = -9), while variables which had a high importance score and a low satisfaction score would result in a high discrepancy score (e.g. 10 - 1 = 9). The higher the score of a discrepancy variable, the more discrepancy there is between the importance of the variable and the satisfaction with the variable.  Appendix J provides the descriptive statistics summary for the discrepancy variables. This shows that the variable scored most important by respondents, DSUPRESP (supervisory respect for employees) had a low discrepancy score of 1.38. The variable score second most important, DAGERESP, (agency respect for employees) had the highest discrepancy score of 5.30. A review of the 10 variables rated most important reveals that they  78 included the six most discrepant variables: DAGERESP DPRODEV DWKSIZE DFUNDS DPROFESS DVAC  (agency respect for employees) (support for professional development) (worksize) (access to funds and resources) (recog. of social work profession) (amount of vacation time)  M M M M M M  5.30 5.26 4.97 4.53 3.77 3.72  The variables with the least discrepancy exhibited very low discrepancy: DACCESS DSTATUS DCOWORK DCLEAR DSUPER DJOBSEC  (access to the workplace) (status of position within the agency) (having competent co-workers) (clearly defined responsibilities) (quality of supervision) (having job security)  M M M M M M  0.11 0.46 1.02 1.21 1.28 1.32  Correlates Of Overall Job Satisfaction S (Satisfaction) Variables The analysis provided above has shown how respondents report the importance of different variables in their jobs, how satisfied they are with those variables, and what discrepancies exist between those two sets of variables. They do not report what connections there are between satisfaction with specific elements of their jobs and their report of levels of overall job satisfaction. Correlation tests between overall job satisfaction and the other satisfaction variables were performed to determine this.  Table 5.3 shows the correlations between the facet-free measure of job satisfaction SOVER and those satisfaction variables which were found to be significant at the 0.01 level.  79 Table 5.3 Correlations Between SOVER And Other Satisfaction Variables (p < = 0.01) SACCOMP SCONSULT SWKSIZE  SVAC  .6000 ( 47) P= .000  .4949 ( 47) P= .000  .4719 ( 47) P= .000  .4566 ( 47) P= .001  SPAPER  SSTATUS  SINFLU  SAUTO  .4512 ( 47) P= .001  .4348 ( 47) P= .001  .4055 ( 47) P= .002  .3964 ( 47) P= .003  SPAY  SSUPRESP SPROFESS SCLEAR  .3652 ( 47) P= .006  .3604 ( 47) P= .006  .3601 ( 47) P= .006  SPROMO  SFUNDS  SHOURS  .3519 ( 47) P= .008  .3475 ( 47) P= .008  .3407 ( 47) P= .010  .3570 ( 47) P= .007  Table 5.3 reports which satisfaction variables best correlate with overall job satisfaction among the respondents. As discussed in Chapter 4 these correlates, while not having the formal predictive strength of an experimental research design, are predictive in nature and logically causal in nature. In addition to the importance of variables and the satisfaction regarding variables, we now have a third set of data relating to overall job satisfaction. The data relates simply to the influence of satisfaction with variables on overall job satisfaction.  80 Fifteen of the 27 satisfaction variables are significant predictors of overall job satisfaction. The small sample size prohibits the use of factor analysis to suggest underlying common factors in the selection of 15 variables and the multiple colinearities shown in Appendix K argue against the appropriateness of regression analysis to predict the influence of each variable on overall job satisfaction.  Observation, however shows that six variables deal with the content of work itself; getting a sense of accomplishment from the job (SACCOMP), access to consultation (SCONSULT), workload size (SWKSIZE), amount of job related paperwork (SPAPER), having clearly defined job responsibilities (SCLEAR), and having access to funds and resources to achieve client related job goals (SFUNDS). Five variables deal with the role of the respondents within the agency: having a position with high status within the agency (SSTATUS), being able to influence decisions of the agency (SINFLU), having decision making autonomy (SAUTO), having a supervisor who treats employees with respect (SSUPRESP), and agency recognition of social work as a profession (SPROFESS). Four variables deal with the employer-employee exchange; amount of vacation time (SVAC), amount of pay (SPAY), opportunities for promotion or advancement (SPROMO), and number of hours worked per week (SHOURS).  81 D (Discrepancy') Variables Finally in the analysis of undifferentiated overall correlations, the facetfree measure of job satisfaction, SOVER, was correlated with the discrepancy variables. The results of the correlations significant at 0.01 level are presented in Table 5.4. This table provides a report of the influence of the discrepancy values on overall job satisfaction. A smaller number of discrepancy variables than satisfaction variables are significant correlates of SOVER. Nine of the variables continue to be represented and the variable DAGERESP is added. Table 5.4 Discrepancy Variable Correlations With SOVER Correlations:  DPAPER  DHOURS  DACCOMP DPROMO  SOVER  -.3806 ( 49) P= .003  -.4337 ( 49) P= .001  -.5779 ( 49) P= .000  -.3537 ( 49) P= .006  DWKSIZE  DFUNDS  DVAC  DSTATUS  -.4830 ( 49) P=.000  -.4183 ( 48) P=.002  -.3775 ( 49) P=.004  -.4146 ( 49) P=.002  DPROFESS DAGERESP -.4014 ( 49) P= .002  -.3321 ( 49) P= .010  These four types of data analysis will be used in Chapter Six to develop recommendations for each of five parties to the employment situation. These parties are the individual, the supervisor, the agency, the governmentemployer, and the union. Briefly however, we have seen that the importance  82 scores given to variables by respondents, the correlates of overall job satisfaction, and the discrepancy scores calculated are not identical. This will have implications for the recommendations to the various parties.  Between Group Differences F (Importance) Variables The variable TOTGOV was recoded into two levels, those with three years or less experience (23 cases) and those with more than three years experience (26 cases). T-tests were performed to determine what differences if any existed between the two groups. The results are reported in Appendix L. We can see that respondents with less experience reported nine variables as being significantly more important than as reported by workers with greater experience. These results should be interpreted with caution however, as less experienced respondents reported every variable except FPAY (importance of pay) as being more important than more experienced workers reported. These statistically significant differences may be a function of the overall lower expectation scores of more experienced workers rather than a true difference in these particular variables.  S (Satisfaction) Variables T-tests were performed to determine if differences existed in the satisfaction reported by less experienced and more experienced workers. Three satisfaction variables were found to have significant differences at the 0.05 probability level. These results are shown in Table 5.5.  83 Table 5.5 T-Tests Of Satisfaction Variables With TOTGOV As Grouping Variable TOTGOV fYears Of Experience With The Agency) t-test for: SJOBSEC (Satisfaction With Job Security)  Group 1 Group 2  Number of Cases  Mean  Standard Deviation  Standard Error  26 23  5.3269 6.9130  2.240 1.857  .439 .387  Pooled Variance Estimate F 2-Tail Value Prob.  t Degrees of 2-Tail Value Freedom Prob.  1.46 .376  -2.68  47  .010  t-test for: SCLEAR (Satisfaction With Clear Job Responsibilities) Number of Cases Group 1 Group 2  Mean  Standard Deviation  Standard Error  5.2115 6.3478  1.576 1.496  .309 .312  Pooled Variance Estimate t Degrees of 2-Tail Value Freedom Prob.  F 2-Tail Value Prob.  -2.58  1.11 .810  47  .013  t-test for: SAGERESP (Satisfaction With Agency Respect for Employees)  Group 1 Group 2  Number of Cases 26 23  Mean  Standard Deviation  Standard Error  3.0192 4.0870  1.676 1.998  .329 .417  Pooled Variance Estimate F 2-Tail Value Prob. 1.42 .395  t Degrees of 2-Tail Value Freedom Prob. -2.03  47  .048  84  Less experienced workers reported lower means on SJOBSEC (satisfaction with job security), SCLEAR (satisfaction with having clearly defined job responsibilities), and SAGERESP (satisfaction with agency respect for employees). These differences appear to be based at least in part on objective factors rather than subjective experience. Less experienced workers have less job security than more experienced workers. Many less experienced workers would be classified as auxiliary workers so are subject to lay-off at any time. This lay-off and recall may be one factor underlying the low score for SAGERESP as well. Similarly, less experienced workers may well have less clear understandings of the job responsibilities resulting in less satisfaction on this variable.  Correlates Of Overall Job Satisfaction S (Satisfaction) Variables Table 5.6 shows the satisfaction variables which significantly correlate with overall job satisfaction for less experienced workers. We see that the number of significant correlates has dropped from the 15 shown in Table 5.3 to five variables. These variables appear to focus on issues of professional recognition and growth along with the issue of getting a sense of accomplishment from work.  85 Table 5.6 Correlation Of SOVER With Satisfaction Variables (TOTGOV < = 3 Years) SOVER (Job Satisfaction; TOTGOV (Years of Experience With The Agency) Corr:  SSTATUS  SPROMO  SACCOMP SPROFESS  SPRODEV  SOVER  .6674 ( 26) p= .000  .6491 ( 26) P= .000  .5917 (26) P= .001  .5376 ( 26) P= .002  .5483 ( 26) P= .002  Table 5.7 shows the satisfaction variables which significantly correlate with overall job satisfaction for more experienced workers. Again, the number of correlates has been reduced, this time to seven variables. These variables appear to focus on issues of control of work (SWKSIZE: satisfaction with workload size; SPAPER: satisfaction with amount of job related paperwork; SFUNDS: satisfaction with access to funds and resources; SINFLU: satisfaction with ability to influence decisions of the agency; and SCONSULT: satisfaction with access to consultation), getting a sense of accomplishment (SACCOMP), and personalized respect (SSUPRESP: satisfaction with having a supervisor who treats employees with respect).  86 Table 5.7 Correlation Of SOVER With Satisfaction Variables (TOTGOV > 3 Years) SOVER (Job Satisfaction; TOTGOV (Years Of Experience With The Agency) Correlations:  SCONSULT SACCOMP SPAPER  SINFLU  SOVER  .6388 ( 21) P= .001  .6113 ( 21) P= .002  .6040 ( 21) P= .002  .5479 ( 21) P= .005  SFUNDS  SWKSIZE  SSUPRESP  .5441 ( 21) P= .005  .5323 ( 21) P= .006  .5146 ( 21) P= .008  Only one satisfaction variable, SACCOMP (satisfaction with getting a sense of accomplishment from the job), significantly correlates to overall job satisfaction for both less experienced and more experienced workers.  D (Discrepancy) Variables Table 5.8 displays the correlation between overall job satisfaction and discrepancy variables for less experienced workers. We see that six discrepancy variables correlate with overall job satisfaction at the 0.01 level of significance. Again, the issue of getting a sense of accomplishment (DACCOMP) is important, issues of professional recognition and growth are prominent (DPRODEV: agency support for professional development; DPROMO: opportunities for promotion or advancement; and DPROFESS: agency recognition of social work as a profession), and this time issues of workload (DHOURS: number of hours worked per week; and DWKSIZE: workload size) appear.  87  Table 5.8 Correlation Of SOVER With Discrepancy Variables (TOTGOV < = 3 Years) SOVER (Job Satisfaction; TOTGOV (Years Of Experience With The Agency) Correlations:  DACCOMP DPRODEV DPROMO  DHOURS  SOVER  -.6166 ( 26) P=.000  -.5513 ( 26) P=.002  -.5951 ( 26) P=.001  -.5944 ( 26) P=.001  DPROFESS DWKSIZE -.5258 ( 26) P= .003  -.4964 ( 26) P= .005  Table 5.9 reports the correlations between facet-free job satisfaction and discrepancy variables for more experienced workers. Getting a sense of accomplishment (DACCOMP) is again significant along with having resources to do the job (DFUNDS and DCONSULT). The issue of paperwork (DPAPER) nears significance at the 0.01 level (p = 0.011).  Table 5.9 Correlation Of SOVER With Discrepancy Variables (TOTGOV > 3 Years) SOVER (Job Satisfaction; TOTGOV (Years Of Experience With The Agency) Correlations:  DFUNDS  DCONSULT DACCOMP DPAPER  SOVER  -.5871 ( 21) P= .003  -.5765 ( 21) P= .003  -.5016 ( 21) P= .010  -.4966 ( 21) P= .011  88  Hypothesis Testing The six formal hypotheses stated in Chapter 3 were:  i)  respondents will report job-specific variables as being more important to them than job variables not specific to social work;  ii)  less experienced workers will rate professional identification variables as more important (than will more experienced workers) and more experienced workers will rate tangible benefit variables are rated as more important (than will less experienced workers);  iii)  respondents will report greater satisfaction with variables over which they have control or influence than variables which are outside of their control or influence;  iv)  less experienced workers will report less satisfaction with variables related to professional identification (than will more experienced workers) and more experienced workers will report less satisfaction variables related to tangible benefits (than will less experienced workers);  v)  satisfaction with job-specific variables will correlate significantly with overall job satisfaction;  vi)  variables based on professional identification will correlate with facet-free job satisfaction for less experienced workers while variables based on tangible job  89 benefits will correlate with facet-free job satisfaction for more experienced workers.  The first hypothesis was confirmed. Respondents reported some variables as being very important to them, with Means as high as 9.12, while reporting others with Means as low as 5.49. Factor analysis to measure underlying factors was not possible due to the sample size, however some principles appeared evident. Respect was highly important with having an employer and a supervisor who treat employees with respect being the highest rated variables overall. Opportunity for professional growth was represented by the high importance scores for support for professional development and access to consultation. Job demand issues were highly rated in the form of having a manageable worksize and getting a sense of accomplishment. Variables not specific to social work such as ease of access to the workplace and amount of paperwork were rated less important.  The second hypothesis was neither confirmed nor disconfirmed. Less experienced workers scored nine variables significantly higher than did more experienced workers. These variables tended to be variables related to professional issues as well. Less experienced workers scored 26 of the 27 variables as more important than more experienced workers however. It is unclear whether the apparently significant differences in these variables are an artifact of the discrimination with which the groups rate variables, rather than a true difference in importance.  90 The third hypothesis was confirmed. Mean satisfaction scores ranged from a high of 7.74 for having a supervisor who treats employees with respect to a low of 3.44 for the level of agency support for professional development. Highest satisfaction was reported for variables which dealt with issues most immediate to the worksite while greatest dissatisfaction was reported for variables which dealt with issues more strongly effected by influences at higher levels of the agency.  The fourth hypothesis was not confirmed. There were significant differences in the satisfaction with three variables between workers with lesser and greater experience. These differences addressed issues of job security, clarity of job responsibility, and agency respect for employees. It is believed that these were the result of a difference in objective experience rather than a difference in how the same experience was interpreted by less experienced and more experienced workers. These variables did not principally relate to satisfaction with professional issues.  The fifth hypothesis was confirmed. Satisfaction with fifteen of the variables was found to significantly correlate with overall job satisfaction at the 0.01 significance level. Again the small sample size did not allow for factor analysis, however several issues appeared to be represented. Issues related to the nature of the work and accomplishing the work, relationships within the work setting, and the nature of the employee-employer exchange were represented. Less job specific variables such as access to the worksite and physical comfort were not significant correlates of the facet-free measure of job satisfaction.  91  Variables were calculated to determine the discrepancy between the importance of variables and the satisfaction with those variables. These new variables were then correlated with overall job satisfaction. Ten variables were found to be significantly correlated. These appeared to represent issues of the nature of the work and accomplishing the work, relationships within the work setting, and the nature of the employee-employer exchange.  The sixth hypothesis was partially confirmed. Satisfaction with only one variable was found to significantly correlate with overall job satisfaction among less experienced and more experienced workers. Satisfaction with getting a sense of accomplishment from the job was significant at the 0.001 level for less experienced workers and at the 0.002 level for more experienced workers. Satisfaction with four other variables significantly correlated for less experienced workers. These variables represented issues of professional recognition and growth. Satisfaction with six additional variables correlated significantly for more experienced workers. These variables represented issues of control over work and personalized respect.  The discrepancy score for getting a sense of accomplishment was the only significant discrepancy correlate for both less experienced and more experienced workers. Five other discrepancy variables representing issues of professional recognition and growth and workload were significant correlates for less experienced workers. Two additional variables relating to having needed resources to do the job were significantly correlated and the issue of the amount of paperwork was significant at the 0.011 level.  92  Summary Three of the six hypotheses were confirmed. The hypothesis regarding significant differences between less experienced and more experienced workers in the satisfaction reported with professional identification was not confirmed. The hypothesis that less experienced workers would report professional identification variables as more important than more experienced workers would report those variables was neither confirmed nor disconfirmed. The hypothesis regarding variables which correlate with facet-free job satisfaction for less experienced workers and more experience workers was partially confirmed.  Several correlates with overall job satisfaction were found for less experienced and more experienced workers. In general, overall job satisfaction correlated with issues related to the content of the work itself, the role of the workers within the agency, and the employer-employee relationship. Overall job satisfaction for less experienced workers correlated with issues dealing with professional recognition and growth. Overall job satisfaction for more experienced workers correlated with issues dealing with workload size, control of their work, and personalized respect.  93 CHAPTER 6 CONCLUSIONS AND RECOMMENDATIONS  Introduction This chapter will begin by making comparisons between the results of this research and other related research. This will include a comparison of the measurement of job satisfaction with those outcomes found in other research. It will also review the correlates of job satisfaction found in related research to those found in this study. The chapter will then make recommendations to each of four organizational parties to the employment situation of the respondents. These parties are the provincial government, the agency (the Ministry of Social Services and Housing), the local supervisor, and the union (the B. C. Government Employees' Union). A distinction is made between the government and the agency due to the organizational structure outlined in Chapter 2.  The government is responsible for negotiated collective agreements, overall personnel matters, and government policies and priorities. The agency, which is responsible for the implementation of personnel matters within the parameters provided by the Government Personnel Services Division (G.P.S.D.), is also responsible for the development and implementation of social work practice policies and making recommendations to the government regarding overall policies and priorities.  Recommendations will not be made regarding individual social workers. Broad recommendations for application by individuals are not appropriate. The  94 wide range of individual scores in both the importance of variables and the satisfaction with variables demonstrate this. Furthermore, Chapter 3 explained the usefulness of the values satisfaction model as a preferred method for individual decision making and the needs satisfaction model as the preferred method for organizational decision making. The chapter will conclude with recommendations for further research.  Comparisons With Other Research Facet Free Job Satisfaction This variable was measured using the widely used question, "All in all, how satisfied would you say you are with your job?". Previous research has typically allowed only responses indicating a level of satisfaction, rather than including a measure of dissatisfaction. Some comparisons are possible however.  Barber used the same wording as above in measuring this variable. Table 6.1 compares the responses he received with the responses in the current research.  95 Table 6.1 Comparison Of Job Satisfaction Scores Between Barber And Current Research Barber (1986) Completely Satisfying  Fairly Satisfying  Not Very Satisfying  Not At All Satisfying  7%  51%  30%  12%  Current Research Very Satisfied  Somewhat Satisfied  Neither Satisfied Nor Dissatisfied  Somewhat Dissatisfied  Very Dissatisfied  10.2%  42.9%  10.2%  28.6%  8.2%  We see that satisfaction at the higher levels appears to be comparable, with Barber reporting 58% with higher levels of satisfaction compared with 53.1% in the current research. Yet very few respondents (12%) in Barber report very low levels of satisfaction. This is contrasted with the current research in which 36.8% of respondents report actual job dissatisfaction. The discrepancy is made up in the difference in the more neutral responses of Barber's "Not Very Satisfying" (30%) and the current research's "Neither Satisfied Nor Dissatisfied" (10.2%).  Jayaratne and Chess (1984) reported job satisfaction among a combined sample of child welfare, family service, and community mental health social workers, Quinn and Staines reported job satisfaction among a representative sample of employed Americans, and Burstein reported job satisfaction among a representative sample of employed Canadians. All used the question, "All in  96 all, how satisfied would you say you are with your job?". Their results are presented in Table 6.2 and compared with the current results. Jayaratne and Chess (1984) reported no significant difference between job satisfaction among child welfare social workers and the other practice fields. They also did not report responses for the lower levels of satisfaction.  Table 6.2 Comparison Of Job Satisfaction Scores Between Various Research Results And Current Research Results Jayaratne and Chess (1984) Very Satisfied  Somewhat Satisfied  Not Too Satisfied  Not At All Satisfied  35.2%  49.1%  N/A  N/A  Quinn and Staines (1979) Very Satisfied  Somewhat Satisfied  Not Too Satisfied  Not At All Satisfied  46.7%  41.7%  8.9%  2.7%  Very Satisfied  Somewhat Satisfied  Not Too Satisfied  Not At All Satisfied  46%  42%  9%  2%  Burstein etal (1975)  Current Research Very Satisfied  Somewhat Satisfied  Neither Satisfied Nor Dissatisfied  Somewhat Dissatisfied  Very Dissatisfied  10.2%  42.9 %  10.2%  28.6%  8.2%  97  These tables indicate that job satisfaction among M.S.S.H. child welfare line social workers is substantially lower than generally reported by employed workers in Canada and the United States. They also indicate that there are substantial differences between job satisfaction among M.S.S.H. child welfare line social workers and social workers in other agencies for which there is literature available. In comparison with Barber, it appears that M.S.S.H. child welfare line social workers experience greater job dissatisfaction. In comparison with Jayaratne and Chess (1984), it appears that M.S.S.H. child welfare line social workers report substantially lower levels of job satisfaction.  Correlates Of Job Satisfaction Barber reported the single best correlate of job satisfaction for direct service workers in a social service agency as getting a sense of achievement from the job. The next most significant correlates were negative correlations with the job not being interesting and with having too heavy a work load.  Jayaratne and Chess (1984) reported the single significant correlating factor with job satisfaction for child welfare social workers was the opportunity for promotion.  McNeely et al reported only three variables as being consistent correlates of job satisfaction among employees of a social service agency. These were negative correlations with having a dull and monotonous job and with feeling very much underpaid, and a positive correlation with having clear job expectations from the supervisor.  98  Table 6.3 shows which satisfaction variables were significant correlates of job satisfaction in the current research and identifies which of these were found in other studies. Table 6.4 provides the same information for discrepancy variables. Satisfaction variables which were found to be significant correlates in other research are included for comparison, although they are not truly discrepancy variables.  99 Table 6.3 Satisfaction With Individual Variables As Significant Correlates Of Job Satisfaction Variable  Overall  < = 3 Years  > 3 Years  Other  SSUPRESP SPRODEV SWKSIZE SCONSULT SACCOMP SFUNDS SPROFESS SVAC SPAY SAUTO SPROMO SINFLU SHOURS SCLEAR SPAPER SSTATUS  YES NO YES YES YES YES YES YES YES YES YES YES YES YES YES YES  NO YES NO NO YES NO YES NO NO NO YES NO NO NO NO YES  YES NO YES YES YES YES NO NO NO NO NO YES NO NO YES NO  N/A N/A YES N/A YES N/A N/A N/A YES N/A YES N/A N/A YES N/A N/A  SSUPRESP (Satisfaction with respect with which supervisor treats employees); SPRODEV (Satisfaction with agency support for professional development); SWKSIZE (Satisfaction with workload size); SCONSULT (Satisfaction with access to consultation); SACCOMP (Satisfaction with getting a sense of accomplishment); SFUNDS (Satisfaction with having sufficient funds/resources to achieve goals); SPROFESS (Satisfaction with recognition of social work as a profession); SVAC (Satisfaction with amount of vacation time); SPAY (Satisfaction with amount of pay); SAUTO (Satisfaction with decision making autonomy); SPROMO (Satisfaction with opportunities for promotion/advancement); SINFLU (Satisfaction with ability to influence agency decisions); SHOURS (Satisfaction with number of hours worked per week); SCLEAR (Satisfaction with clarity of job responsibilities); SPAPER (Satisfaction with amount of job required paperwork); SSTATUS (Satisfaction with status of position within the agency) (Note: N/A reflects no specific comparable variable was found in other research.)  100 Table 6.4 Discrepancy Scores Of Individual Variables As Significant Correlates Of Job Satisfaction Variable  Overall  < = 3 Years  > 3 Years  Other  DAGERESP DPRODEV DWKSIZE DCONSULT DACCOMP DFUNDS DPROFESS DVAC DPROMO DHOURS DPAPER DSTATUS  YES NO YES NO YES YES YES YES YES YES YES YES  NO YES YES NO YES NO YES NO YES YES NO NO  NO NO NO YES YES YES NO NO NO NO YES NO  N/A N/A YES N/A YES N/A N/A N/A YES N/A N/A N/A  DAGERESP (Discrepancy between importance of and satisfaction with agency respect for employees); DPRODEV (Discrepancy between importance of and satisfaction with agency support for professional development); DWKSIZE (Discrepancy between importance of and satisfaction with workload size); DCONSULT (Discrepancy between importance of and satisfaction with access to consultation); DACCOMP (Discrepancy between importance of and satisfaction with getting a sense of accomplishment from the job); DFUNDS (Discrepancy between importance of and satisfaction with availability of funds and resources to achieve goals); DPROFESS (Discrepancy between importance of and satisfaction with agency recognition of social work as a profession); DVAC (Discrepancy between importance of and satisfaction with amount of vacation time); DPROMO (Discrepancy between importance of and satisfaction with opportunities for promotion or advancement); DHOURS (Discrepancy between importance of and satisfaction with number of hours worked per week); DPAPER (Discrepancy between importance of and satisfaction with amount of job required paperwork); DSTATUS (Discrepancy between importance of and satisfaction with status given job position within the agency) (Note: N/A reflects no specific comparable variable was found in other research.)  101  Only one variable is consistently represented in overall analysis within M.S.S.H., at both levels of experience with M.S.S.H., and appears in some other literature regarding job satisfaction among employees of social service agencies. This is the variable of getting a sense of accomplishment from the job. This absence of a number of consistent correlates emphasizes the importance of developing plans for improving job satisfaction for specific target groups of employees. The target groups should be defined as clearly as possible, taking into account years of experience with the agency. Similar findings were reported by Barber with regards to cross-classification recommendations.  Recommendations The single most consistent correlate of job satisfaction found in this research has been getting a sense of accomplishment from the job. Recommendations regarding this variable will be made following the specific recommendations to the parties involved.  Recommendations To The Government Recommendations to the government will be based on the government's role in relation to employees rather than its role with regards to clients. Recommendations regarding agency operations will be addressed in the sections dealing with recommendations to the agency.  102 General Recommendations This section will deal with recommendations regarding variables which correlate with job satisfaction without reference to years of service as a control variable. These variables are found in the column labelled, "Overall" in Table 6.3 and Table 6.4.  The provincial government, through political decisions or through its agent G.P.S.D., has principal responsibility for five of the nine variables common to the Overall columns of Table 6.3 and Table 6.4. The variables in common are WKSIZE (workload size), FUNDS (access to funds and resources), VAC (amount of annual vacation), PROMO (opportunities for promotion), and HOURS (number of hours worked each week). The most important issues here are those of workload size and relief from it (WKSIZE, HOURS, and VAC), opportunity for promotion (PROMO), and resources to perform the job (FUNDS).  Workload And Hours Only 20.4% of social workers reported being somewhat or very satisfied at the number of hours they actually worked per week. Their contract requires a 35 hour work week, yet they reported working a mean of 39.42 hours per week, or almost 13% more than the hours required. The minimum hours reported was 35 hours per week and the maximum was 52.5 hours per week. Almost two-thirds (63.3%) reported working more than 35 hours per week on average.  103 Social workers did not report that they wanted reduced hours of work. The mean number of hours they believed was reasonable for a child welfare line social worker to work was 35.1 hours per week and 73.5% reported 35 hours per week as the reasonable work week. Social workers did not report dissatisfaction with the number of hours they are contracted to work, they reported dissatisfaction with the number of hours they actually work each week. Yet satisfaction with the number of hours worked was a significant correlate of job satisfaction.  Why would social workers be working more hours per week than they are contracted to work, while being dissatisfied with working those hours? Perhaps the answer lies in a second variable at the control of the government. Social workers reported a large discrepancy between the importance of workload size and their satisfaction with workload size. This variable was also an important correlate of job satisfaction. High workloads may result in social workers working longer hours than they believe is reasonable and for which they have been contracted. This heavy workload and long hours results in lower job satisfaction.  Workload size can be adjusted using three principal methods. The volume of work to be performed can be reduced, the efficiency with which work is performed can be improved, or the number of workers to perform the work can be increased. Increased efficiency is primarily a matter for the agency, so it will be addressed in that section.  104 Child welfare social work is fundamentally a demand driven type of work. Child welfare social workers do not typically seek out work; clients, courts, or communities demand their work. Governments may legislate acceptable child care standards for families, but these are defined by the larger society in large part. When government is also the employer, political requirements to avoid embarrassment may also create work demand. It is unlikely then that reduced volume of work will occur.  An increase in staffing appears to be the most feasible action available to the government for improving satisfaction with workload. Numerous influences have increased the workload on child welfare line social workers during the past 15 years. Social workers in M.S.S.H. increased in specialization during the several years prior to the research. This has reduced the number of social work positions in child welfare services as workers were transferred to specialized teams dealing with services for mentally handicapped people living in the community. Cases were transferred from child welfare social workers to the specialized teams. However, many social workers believe that the number of social work positions transferred was disproportionately high compared to the transfer of workload.  Additionally, ministry staffing decreased significantly during the years prior to the research. Although data is not available specifically for child welfare social work positions, the number of staff allocated to the ministry has decreased between 1980-81 and 1987-88 while ministry field offices have taken on greater responsibility for the increased number of de-institutionalized  105 mentally handicapped people. Table 6.5 shows the decrease in the total staff allocation for the ministry between 1980-81 and 1987-88. It is apparent that the low satisfaction which child welfare line social workers report is not simply subjective; it is a reflection of actual high workloads.  Table 6.5 Total Ministry Full Time Equivalent Staff Positions (includes all job classifications in all locations') 1982- 83 1983- 84 1984- 85 1985- 86 1986- 87 1987- 88  5988 5443 4928 4463 4473 4466 (Source: B.C.G.E.U.)  RECOMMENDATION G.l The government should review both its expectations of child welfare line social workers and its staffing levels for those positions. Some combination of decisions to either reduce work demands or increase the number of workers to fulfil the demands should be made. Recognizing the importance of public pressure on government to maintain high standards, special attention should be give to increasing staffing rather than decreasing standards. A decision to decrease stated standards may not result in decreased workloads as community demands on the agency may remain at the same level. Additionally, a reduction in standards may produce increased incidents which come to public attention, resulting in increased community demands on child welfare social workers.  106  Vacation Vacation time is a related but separate issue in that while an increase in vacation time does not offset a high workload, it does offer temporary relief from it. Satisfaction with amount of vacation time is significantly correlated with job satisfaction. The discrepancy between the importance and satisfaction scores on this variable was the sixth highest among the 27 variables.  RECOMMENDATION G.2 The government should increase the length of annual vacation time for child welfare line social workers. Implementation of this recommendation would be subject to the collective bargaining process so would require negotiation with the B.C.G.E.U. The government should recognize that this goal may not be easily achievable. The B.C.G.E.U. negotiates collective agreements for employees in a wide range of classifications and so has interests beyond those of members in one field of practice within one job classification. Differential benefits do exist, but they are not common.  This recommendation may be less important than recommendation G.l. If the high discrepancy score for vacation time is because vacation is seen as a relief from high workloads, adoption of Recommendation G.l may reduce the discrepancy score for VAC (amount of vacation time) by both reducing its importance and increasing satisfaction with it.  107  Access To Resources Just as vacation time is separate from but related to workload, so is the issue of access to funds and resources to achieve client related job goals. Access to sufficient resources allows work to be performed with greater efficiency. Respondents report a high discrepancy between the importance with which they view having access to funds and resources and their satisfaction with access to funds and resources. Satisfaction and discrepancy scores of these variables are also significant correlates of job satisfaction.  Much of the work of child welfare social workers in this agency relates to access to resources rather than direct access to funds. Resources are contracted or provided directly by the agency based on funds provided by government however. It appears that the issue of concern to respondents in this variable is access to resources funded by government. Satisfaction with access to resources can be improved by increasing the number and variety of resources; increasing the services of each resource, thereby reducing waiting time; and by increasing the efficiency of social workers seeking access to available resources.  RECOMMENDATION G.3 Government should increase resource funding to ensure that social workers have access to an adequate range of resources of sufficient quality to allow them to achieve client related job goals. Resources which may require additional funding could include counselling services, treatment services, and residential care services including foster homes and group living facilities.  108 Utilization of the increased resource funds should be based on local needs as identified by social workers. Additional specialized resources may require identification at the inter-regional or provincial level.  Promotional Opportunities Opportunity for promotion was a significant correlate of job satisfaction. Respondents overall ranked it 16th in terms of importance and 11th in terms of discrepancy scores. It appears then that in general terms this is a less important overall variable for respondents. Although there is potential benefit in increasing opportunities for promotion within the agency, greater benefit will likely come from improvements in other areas.  RECOMMENDATION G.4 Improved opportunities for promotion within the agency should be a lower priority than making other improvements at this time. It may be useful to review this variable at a later date if job satisfaction increases.  Recommendations Specific To Less Experienced Workers The above recommendations were based on the overall correlates of job satisfaction. It may be that the government has particular interest in improving job satisfaction among workers with less experience. Over one-half of respondents had three years or less experience with the employer. This implies that there is much movement out of child welfare line social work positions in the first years after taking employment. It may be that these employees advance rapidly or relocate geographically or to other government agencies. It is also possible that they leave government employment. In any  109 case, there is a high representation of less experienced workers in these difficult positions. Improved job satisfaction among these workers may improve staff retention in these line positions. It may also enhance staff recruitment for these positions as the positions come to be known as being characterized by high job satisfaction.  Four variables are represented in both Table 6.3 and Table 6.4 as correlates of job satisfaction for less experienced workers, PRODEV (agency support for professional development), ACCOMP (getting a sense of accomplishment from the job), PROFESS (agency recognition of social work as a profession), and PROMO (opportunity for promotion or advancement). PRODEV and PROMO are both subject to influence by the government. Additionally, workload issues as represented by DWKSIZE (workload size) and DHOURS (number of hours worked each week) are significant correlates of job satisfaction as is SSTATUS (status of the position within the agency).  Professional Development Agency support for professional development is in large part enhanced by access to time and funds to undertake professional development and restricted by virtue of workload. All social workers who are regular employees of the provincial government are guaranteed 10 days per year of professional development leave, through their collective agreement. This is a substantial amount of professional development leave which few social workers use fully. Social workers report this to be an important variable, report low satisfaction with this variable in their jobs, yet do not use the professional development  110 time available to them, The reasons for this apparent contradiction may lie in two factors.  First, there is little financial support for expenses related to professional development. It appears that while social workers have time available for professional development, they have no support from the government to pursue it. Second, it may be that while social workers are guaranteed paid leave time to participate in professional development, they believe that their workload is such that they are not able to take that time. The government makes no provision for workload coverage for staff who are on paid leave to undertake professional development under that article of the collective agreement. While it may be that social workers have formal access to professional development, the barriers of financial cost and heavy workloads undermine that access.  RECOMMENDATION G.5 Government should recognize the importance of professional development both as a method for improving service and as a method for improving job satisfaction. This recognition should include a commitment of funds to support professional development separate from staff training. It should also include provision of replacement staff on a one-to-one basis for social workers who are absent for purposes of professional development.  Ill  Promotional Opportunities This variable was identified above as being less important as a correlate of job satisfaction overall. It is however, a significant correlate of job satisfaction for less experienced social workers. A recommendation regarding this variable for less experienced workers is included in the agency recommendations.  Workload And Hours Recommendation G.l above addresses these variables as overall correlates of job satisfaction. It is emphasised here that these variables are significant correlates of job satisfaction for less experienced workers as well.  RECOMMENDATION G.6 See recommendation G.l. This recommendation is of importance specifically to less experienced workers.  Recommendations Specific To More Experienced Workers The four variables CONSULT (access to consultation), ACCOMP (getting a sense of accomplishment from the job), FUNDS (access to funds and resources), and PAPER (amount of job related paperwork) are present in both Table 6.3 and Table 6.4 for more experienced workers. The additional variables SUPRESP (supervisor's respect for employees), WKSIZE (workload size), and INFLU (ability to influence decisions of the agency) are present in Table 6.3 alone.  112  Workload And Access To Resources Recommendations G.l and G.3 above addressed the variables FUNDS and WKSIZE. Recommendations regarding CONSULT, PAPER, SUPRESP, and INFLU will be directed to the agency. It is emphasized here that satisfaction regarding workload issues is of specific importance to more experienced workers.  RECOMMENDATION G.7 Recommendation G.l is restated as being of importance specifically for more experienced workers.  Recommendations To The Agency Recommendations in this section deal with matters under the control or significant influence of the management of the Ministry of Social Services and Housing. No effort has been made to direct the recommendations to specific levels of management as it is recognized that implementation may take place at a variety of levels.  The recommendations in the first section deal with the overall correlates of job satisfaction without using years of experience as a control variable. Recommendations for specific groups follow the general recommendations.  113 General Recommendations The four variables WKSIZE (workload size), PROFESS (agency recognition of social work as a profession), HOURS (number of hours worked each week), PAPER (amount of job related paperwork), and STATUS (status of the position within the agency) which appear in both Table 6.3 and Table 6.4 as overall correlates of job satisfaction are either principally the domain of the agency, or the agency is a strong influence on the principal player for those variables. The variable STATUS will not be addressed here due to its low importance score and low discrepancy score. The agency is similarly important with regards to the variables SSUPRESP (supervisor's respect of employees), SCONSULT (access to consultation), SAUTO (decision making autonomy), SINFLU (ability to influence the agency), SCLEAR (having clearly defined job responsibilities), and DAGERESP (agency respect for employees). Although the agency is not directly responsible for the variable PROMO (opportunities for promotion or advancement), it will be addressed here as well.  Workload The agency is a partial player in the matter of workload. The agency is responsible for adminstration of various pieces of legislation in a demand driven work. It is constrained by legislation, government, and community demands in its ability to define what work is to be done. It is constrained by government in its ability to provide staff to meet those demands. It may also be partially limited in its ability to restrict work to only the more urgent and perhaps less rewarding aspects of the job by its need to recruit and retain employees as that may increase staff attrition and decrease agency marketability. The limited ability to restrict work and the limited ability to  114 increase staffing implies that the agency's most likely method for success in reducing workloads is in increasing efficiency of work.  Child welfare social worker time is consumed by four major work areas; direct client contact, collateral agency contact, court work, and documentation. Almost 90% of respondents reported being "Very Dissatisfied" (22.4%), "Somewhat Dissatisfied" (40.4%), or "Neither Satisfied Nor Dissatisfied" (26.5%) with the amount of job required paperwork. This would appear to be an ideal area in which to improve efficiency.  Client and agency contact are fundamental elements of social work, and may be the reason why many people enter social work. Increased efficiency in client related areas which resulted in decreased quantity or quality of client contact, may reduce social worker effectiveness and job satisfaction.  The agency has some influence over the number of hours its employees work however. Respondents report that they work on average almost 13% more time than they are contracted to work. They report high discrepancy between how important it is for them to average no more than 35 hours per week and how satisfied they are with the number of hours they actually work. The mean number of hours per week they believe is reasonable for child protection line social workers to work is 35.1. Most report working more than 35 hours per week. This data, while perhaps documented here for the first time, is not new information. It is common understanding among social workers, supervisors, and management personnel that social workers work more  115 than 35 hours per week. Apparently no one takes steps to change the behaviour.  Social workers may see this as a way to try to better serve their clients or to fulfil excessive job requirements of the agency. Managers may view it as an acceptable method for increasing efficiency. Supervisors may view it as a combination of these. We have seen however that there is a negative correlation between job satisfaction and staff turnover. We know that staff turnover results in reduced efficiency, increased costs, and decreased quality of service. We know that the respondents report dissatisfaction with the number of hours they work per week. We therefore can deduce that requiring or allowing social workers to work more than 35 hours per week will result in increased staff turnover, reduced efficiency, increased costs, and decreased quality of service.  RECOMMENDATION A.1 The agency should explore all methods for increasing efficiency, with particular emphasis on paperwork requirements. Paperwork should be eliminated where possible, reduced where it is not possible to eliminate it, and performed by non-social work staff whenever there is not a professional necessity or legal requirement that the paperwork be performed by a social worker. Elimination and reduction may include reductions in the number of forms required to complete a task. Transfer of non-social work tasks may require an increase in the number of administrative support staff or the hiring of case aides. The increase in efficiency and the lower pay level for these staff should offset the costs however.  116  RECOMMENDATION A.2 The agency should explicitly advise social workers that they are not to work more than 35 hours per week, averaged over a two week period, unless specifically authorized to do so. It is recognized that social workers may be required to work more than 35 hours per week at times due to emergency situations. This should be closely monitored however.  RECOMMENDATION A.3 It is recognized that social workers may continue to feel obliged to work more than 35 hours per week in order to fulfil explicit or implicit unrealistic job expectations. The agency should explicitly direct social workers to consult with their supervisor prior to working any more than an average of 35 hours per week over a two week period. The supervisor, in conjunction with management personnel, should direct the social worker on whether or not to work the additional hours. This use of approved overtime should not be used to combat a chronic problem of overwork however. Continuous use of overtime is not a cost-effective method for dealing with workload problems.  Professional Recognition Agency recognition of social work as a profession and satisfaction with the status of the position within the agency address issues of professional recognition. Social workers report that having a position with high status within the agency is the least important of the 27 variables provided. They also report very little discrepancy between the importance of this variable and  117 their satisfaction with it. There appears to be little benefit from the agency directing efforts to this area.  Social workers report a much higher level of importance to the variable of agency recognition of social work as a profession and report a high discrepancy score for this variable. It appears then that the issue is recognition of the profession, rather than individual recognition. There appears to be some conflict between this and the 22.4% membership rate in a professional association by the respondents. This may be partially explained in several ways.  First, only 55.1% of respondents reported having a B.S.W. or M.S.W. degree. Under the current constitution and by-laws of the B.C.A.S.W. and the Board of Registration for Social Workers, almost one-half of the respondents would not be eligible for registration as a social worker or for full membership in the professional association. They may have strong feelings about the need for recognition of social work as a profession, but they are not able to participate in the professional body.  A second possible explanation is that respondents view recognition of social work as a profession as being a combined effort. That is, they see individual participation or behaviour as being directly related to agency participation or behaviour. If this was the case, they would view personal behaviour as being dependent on agency behaviour. Agency recognition of social work as a profession would be followed by personal behaviour which would reflect support for social work as a profession.  118  Third, it may be that respondents do not perceive agency recognition of social work as a profession to be comparable to recognition of or participation in a professional association. If this were the case, agency recognition of social work as a profession would not be correlated with individual membership in or support for a professional association.  More likely, some combination of the above is in effect. Of respondents eligible for membership in the B.C.A.S.W., 40.7% are members. There is no employment benefit or compensation for membership in the association and membership fees for full-time employed social workers in 1989 exceeded $225 annually. Other respondents who were eligible for membership may support the concept of social work as a profession, but wish to see agency recognition of this prior to making personal commitments. Still others may view agency recognition of social work as a profession and membership in a particular professional association as being unrelated issues. Finally, fully 44.9% would not have been eligible for new full membership in the association at the time of the research.  It is clear however that respondents report low satisfaction with agency recognition as social work as a profession. There is considerable discrepancy between the importance with which they view it and their satisfaction with it. Finally, it is a significant correlate of job satisfaction for less experienced social workers.  119  RECOMMENDATION A.4 The agency should review its level of recognition of social work as a profession. The agency should enter into discussions with the B.C.A.S.W. regarding actions which would enhance the agency's explicit recognition of social work as a profession. These discussions should include the B.C.G.E.U. as the recognized bargaining agent for the social workers. Although this may not be a collective bargaining issue, it is important that any efforts by the agency in this area not be frustrated by a failure to consult with the collective bargaining agent.  Relationship With Supervisor Many of the correlates which appear in Table 6.3 only deal with issues of supervision. SUPRESP (supervisor' respect for employees), CONSULT (access to consultation), CLEAR (having clearly defined job responsibilities), and AUTO (having decision making autonomy) are in large part functions of supervision. They will be addressed in the recommendations for supervisors below.  RECOMMENDATION A.5 Particular attention in the selection, initial training, and ongoing development of supervisors should be given to ensure that supervisors demonstrate respect for employees, clearly define the responsibilities of the line social workers' job, are readily available for consultation, and allow social workers decision-making autonomy regarding case decisions.  120 RECOMMENDATION A.6 The agency should recognize the importance of the above supervisory behaviours and should structure the role of the supervisor to maximize these behaviours. The job expectations of supervisors should not interfere with their ability to perform these functions. This would include avoiding the delegation of management functions or responsibilities from the area manager level to the supervisory level. The supervisor role should be protected as supervisory in nature, rather than managerial in nature.  Agency Respect For Employees Respondents reported AGERESP (working for an agency which treats employees with respect) as being second in importance only to SUPRESP (having a supervisor who treats employees with respect), yet they reported the second lowest satisfaction score for this variable. While 91.8% reported this to be either "Very Important" (75.5%) or "Quite Important" (16.3%), 75.5% reported being "Very Dissatisfied" (34.7%) or "Somewhat Dissatisfied" (40.8%). This significant correlate of job satisfaction scored the highest of the discrepancy scores. There is clearly a great dissatisfaction among social workers regarding the respect they wish from the agency and the respect they experience.  The underlying agency behaviours which lead to this experience of not being respected are not known. As noted earlier, this variable was included in the research at the request of social workers in the agency. It is not reported in other literature, neither does it have a firm theoretical definition. It is clearly an important variable however. A correlation between DAGERESP and other discrepancy variables was performed to observe any strong  121 correlations which might give clues as to which agency behaviours are related to the experience of AGERESP. Nine variables were found to correlate with DAGERESP at the > 0.001 level. These variables were DSUPER (quality of supervision), DMORALE (agency morale), DHOURS (hours worked each week), DCLEAR (having clearly defined responsibilities), DFRINGE (employment fringe benefits), DWKSIZE (workload size), DPRODEV (agency support for professional development), DPROFESS (agency recognition of social work as a profession), and DSUPRESP (having a supervisor who treats employees with respect).  RECOMMENDATION A.8 The agency should undertake a study to determine which agency behaviours are perceived by social workers as being respectful and disrespectful of employees. Particular attention should be given to behaviours related to the variables mentioned above. Other variables not included in this research may contribute to the experience of AGERESP however, so the study should not be limited to these variables.  Recommendations Specific To Less Experienced Workers The agency has important influence regarding three of the variables which best correlate with job satisfaction for less experienced workers. These variables are PROFESS, PRODEV, and STATUS. Each of these have been discussed with regards to general recommendations for the agency and will not be discussed further here.  122 Opportunities for promotion are principally a function of the government in terms of approving the introduction of additional job classifications and the expansions of classification series. The agency can ensure that line workers are aware of possible career paths and provide opportunities for workers to prepare themselves for promotion.  RECOMMENDATION A.9 The agency should ensure that less experienced social workers are aware of possible career paths within the ministry. The agency should develop a program to provide opportunities for workers to develop the knowledge, skills, and personal suitability which the agency seeks in those holding higher positions. The program should be universally accessible to ensure that workers' opportunities for advancement are not restricted by not having access to the program. It should be centrally structured and administered to ensure that all workers have equal access to career opportunities. It should not rely on the skills of the district supervisor, as this would increase the potential for restriction of access or variation in the quality of the program.  Recommendations Specific To More Experienced Workers The agency has important influence regarding supervisory respect for employees (SUPRESP), access to consultation (CONSULT), ability of social workers to influence agency decisions (INFLU), and amount of paperwork (PAPER), which were significant correlates of job satisfaction among more experienced social workers. Each of these have been discussed above and will not be reviewed further here. Recommendations for more experienced workers are contained the in the general recommendations.  123  Recommendations To Supervisors District supervisors in the agency represent agency management at the local level. They also have daily contact with line workers and may be an immediate influence on how individual workers perceive their jobs. The government and the agency create the macro environment for work, while the district supervisor is an important influence on the micro environment of work.  There appears to be little opportunity for substantial improvement in job satisfaction in the agency by changes to supervisory practice. Social workers reported low discrepancy scores regarding the variables over which supervisors have important influence.  General Recommendations The district supervisor has principal influence regarding the four significant correlates of job satisfaction; SUPRESP (supervisory respect for employees), CONSULT (access to consultation), AUTO (decision making autonomy), and CLEAR (clarity of job responsibilities) and significant influence regarding HOURS (hours worked each week).  Clarity Of Job Responsibilities A primary function of supervisory personnel in any employment is to determine the work to be performed. The work performed in child welfare social work typically takes place away from a central workplace, not under immediate supervision. This "determination of work" in social work is primarily one of defining responsibilities. This research shows a significant  124  correlation among child welfare line social workers between satisfaction with the clarity of their job responsibilities and their job satisfaction.  Agency policies do not effectively set local or daily priorities but provide overall direction on how to perform work. This is particularly significant when the work demand does not allow all tasks to be performed. Responsibility for addressing this issue lies with district supervisors.  RECOMMENDATION S.l District supervisors should clearly understand the responsibilities of social workers and the priorities of those responsibilities as established by agency management. Supervisors should emphasize clearly the responsibilities of social workers and establishment of priorities within social work responsibilities. This should include clear instruction on what responsibilities are beyond the domain of line social workers. They should train line social workers to be aware of the responsibilities and priorities of district supervisors as well to reduce the likelihood of unclear responsibility boundaries.  Autonomy And Consultation This research shows a significant correlation between satisfaction with the amount of decision making autonomy social workers have and their job satisfaction. It also shows a significant correlation between access to consultation and job satisfaction. Social workers report relatively little discrepancy between the importance they place on these variables and their satisfaction with the variables. It appears that in general, social workers are  125 satisfied with their decision making autonomy and with the access they have to consultation.  While it appears that in general, change is not needed in this area, there may be considerable individual variations. Dissatisfaction with access to consultation was reported by 28.6% of respondents and a further 16.3% reported being "Neither Satisfied Nor Dissatisfied".  RECOMMENDATION S.2 While no general recommendations are made regarding decision making autonomy or access to consultation, individual supervisors should examine their own practice regarding providing access to consultation. This variable is a significant correlate of job satisfaction, and provides an opportunity for some improvement in job satisfaction among a substantial number of social workers.  Supervisory Respect For Employees There is little opportunity for change in this variable which scored the highest among the satisfaction scores. Only 14.3% of respondents reported any dissatisfaction with this variable, while 55.1% reported being "Very Satisfied".  RECOMMENDATION S.3 No general recommendations are made regarding this variable. Individual supervisors should review their practice however to determine if their behaviour towards employees is experienced as respectful by the employees whom they supervise.  126  Hours Worked The district supervisor has the most immediate impact regarding the number of hours worked by individual social workers. Agency direction should set a clear expectation that social workers not work more than their contracted hours. In the absence of such agency direction, or as a method of ensuring compliance with that direction, supervisors have the opportunity to influence the number of hours which social workers work.  RECOMMENDATION S.4 District supervisors should ensure that social workers report all hours worked. Social workers who work on average more than 35 hours per week should be discouraged from doing so. If this results in the district office being unable to meet the expectations of agency management, this should be brought to the attention of management who have responsibility for modifying standards and work priorities.  Recommendations Specific To Less Experienced Workers None of the variables which significantly correlate with job satisfaction specifically for less experienced workers in this research are under the substantial influence of district supervisors. No recommendations are made in this section.  127 Recommendations Specific To More Experienced Workers The variables SUPRESP and CONSULT are significant correlates of job satisfaction for more experienced workers. Recommendations regarding these variables have been made above and will not be elaborated upon here.  RECOMMENDATION S.5 Recommendations S.2 and S.3 have particular relevance specifically to more experienced social workers.  Recommendations To The Union Collective bargaining agents have different priorities and responsibilities from employers. Employers may be interested in maximizing efficiency, reducing staff turnover, and improving staff recruitment. Unions on the other hand have an interest in making progress toward the stated wishes of their members, within the context of the overall goals and values of the labour movement.  Unions often have little control over a number of variables which are significant correlates of job satisfaction. Unions cannot effectively negotiate access to consultation, amount of paperwork, status in the agency, influence, autonomy, etc. The union representing social workers in this situation of a demand driven work is prohibited by law (Public Service Labour Relations Act s. 13) from negotiating the staff establishment. Additionally, many of the variables in which unions traditionally do have bargaining ability presented little discrepancy (e.g. JOBSEC: job security and BENEFIT: fringe benefits) or were not strong consistent correlates of job satisfaction (PAY: amount paid).  128  Finally, unions also have the difficulty of negotiating collective agreements which maximize the ease of contract administration. A collective agreement which attempted to provide different benefits for each job classification and differential benefits based on years of experience could quickly become unwieldy to administer. Therefore, recommendations to the union will be general in nature, rather than specific for less experienced and more experienced workers.  Despite the limitations outlined above, there are several opportunities for the union to improve job satisfaction for its members. The variables VAC and PAY displayed high discrepancy scores, were significant correlates of job satisfaction either as satisfaction variables or discrepancy variables, and are amenable to union involvement. The variable AGERESP (agency respect for employees) displayed the greatest discrepancy, and while respect is not directly negotiable, some behaviours related to respect may be.  We have seen that negotiation regarding workload issues is difficult due to legislation and the demand driven nature of the work. However, there may be some opportunity for partial addressing of this issue. Finally, although the issue of agency recognition of social work as a profession is not negotiable, its recognition may be influenced by the union.  129 Pay And Vacation The variable of pay is one familiar to labour unions as a bargaining issue. This research demonstrates its importance as a correlate of job satisfaction for child welfare line social workers however. This research also shows that 59.2% of respondents report being either "Very Dissatisfied" (28.6%) or "Somewhat Dissatisfied" (30.6%) with their pay. Similarly, 55.1% of respondents reported being either "Very Dissatisfied" (26.5%) or "Somewhat Dissatisfied" (28.6%) with their annual vacation time.  RECOMMENDATION U.l The union should explore the two contract areas of pay and vacation for negotiating increased benefits for social workers. It has been shown that improvement in these areas would have benefit for the government and the agency in terms of job satisfaction among social workers, so there may be shared interest by these parties to negotiate improvements in these areas.  Agency Respect For Employees As noted above, the specific agency behaviours related to this variable are not well defined. Additionally, the issue of respect is not one usually amenable to negotiation. One agency behaviour which may be negotiable however is the agency's behaviour in situations involving possible discipline of a social worker.  Some collective agreements include a "justice and dignity" article. This article restricts the employer's ability to impose some forms of discipline on an employee until the proposed discipline has been upheld by an independent  130 party. Typically this kind of discipline would involve suspension without pay or termination of employment. Union grievances when this level of discipline takes place often involve protracted arbitration processes, during which the employee is without pay and is at a considerable disadvantage in attempting to secure other employment. A justice and dignity article would require that the employer either maintain the person in employment, perhaps in an alternative capacity, until the disciplinary matter was resolved by arbitration, or suspend the employee with pay until the outcome of the arbitration was determined.  Justice and dignity articles are already contained in collective agreements in British Columbia negotiated between the United Steelworkers of America Local 3376 and the Crown Cork and Seal Company and between Local 2821 and Ball Packaging. Additionally, the B.C.G.E.U. has developed model wording for such an article for use in the public sector.  RECOMMENDATION U.2 It is recommended that the union attempt to negotiate a justice and dignity clause in the collective agreement.  Recognition Of Social Work As A Profession This issue may pose some difficulty for the union in terms of jurisdiction. Recognition of a professional association is sometimes considered to imply that the professional association has a legitimate role in employer-employee relations. This holds problems for unions as they typically perceive themselves to be the sole legitimate representative of the employees in relationship to the employer.  131 The B.C. Association of Social Workers has a history of recognizing unions as the legitimate representatives of social workers with regard to labour-management relations. It perceives itself as being a voice for social work with regards to issues of social policy. There continues to be some hesitancy and caution in the relationship between the B.C.G.E.U. and the B.C.A.S.W. however.  RECOMMENDATION U.3 The union should enter into discussions with the professional association confirming the understanding of the legitimate roles of each body. The union should also recognize potential benefits to its members by increased recognition of the professional association. Increased legitimacy for the professional association could provide a method for improvement of workload conditions of social workers who are members of the union. Union opposition to an initiative by the agency to give greater recognition to social work as a profession might be viewed unfavourably by social workers who are members of the union.  Recommendations Regarding Achieving A Sense Of Accomplishment The variable ACCOMP (getting a sense of accomplishment from the job) was the single variable which was a significant correlate of job satisfaction with and without using years of experience as a control variable and is found in other similar research (Barber). This research does not propose to summarize the literature regarding methods for designing jobs in order to maximize getting a sense of accomplishment. Several principles seem apparent, however.  132  First, there should be maximal agreement among the government, the agency and social workers regarding the goals, objectives, and methods of the work. Employees who fundamentally disagree with the policies and programs of the agency or their employer are unlikely to be able to achieve high job satisfaction. This is a principle of the values satisfaction model of job satisfaction. This agreement may be more difficult to achieve when government is the employer, rather than an independent body being the employer, due to policies being influenced by the political process. In this situation, the agency may act as a buffer to provide some stability in philosophy across political administrations.  Second, there should be maximal clarity regarding the definition of the work. It is difficult to get a sense of accomplishment from doing work which is undefined. Work may be defined in terms of tasks, in terms of outcomes or as a combination of these. There is currently little information in the agency which provides clear definitions of the work of child welfare line social workers.  Third, clearly defined work tasks and goals should be measurable. Clearly defined work which is not measurable will not be useful in increasing the sense of accomplishment from the job. If the work and goals are unmeasurable, social workers are less likely to know whether they have accomplished the goals.  133  Fourth, work should be achievable. It is unlikely that social workers will get a sense of accomplishment from the job if the job is defined in such a way that the goals are not realistically achievable.  RECOMMENDATION O.l The parties identified above should make efforts to increase the congruence between the values base of the work and the professional values of the social workers. The parties should make co-operative efforts to accurately define the work of child welfare line social workers. The definition should include the opportunity for measurement of the work and a definition of acceptable workloads. There should be opportunity for review of workload using the measurement, independent review of workloads, and an agreed upon process for resolution of excessive workload demands.  Recommendations For Further Research This research has identified a number of issues worthy of further research. First, is the issue of the reliability of the measurement instrument. It would be desirable to test the reliability of the individual variables in the instrument using a test-retest methodology.  Second, it would be desirable for research to be conducted regarding the variables which influence worker satisfaction with the variable AGERESP. This variable was included at the recommendation of workers in the agency. Discrepancy in this variable was found to be a significant correlate of job  134  satisfaction, yet it is unclear what phenomena or agency behaviours were being measured.  Third, it would be desirable to retest the sample over time. The methodology used provides an opportunity for this to take place. This resampling may provide some information about the predictive value of the variables ATTEMPT (attempt to find other employment outside the agency) and BELIEVE (belief the respondent will be working in the agency in two years time). It would provide an opportunity to measure whether there have been changes in job satisfaction among the sample over time. It would also provide an opportunity to measure the predictive value of the measure of job satisfaction on termination of employment with the agency.  Fourth, it would be desirable to implement the research on a wider scale. The small sample size in the research did not allow the use of factor analysis to determine underlying job factors. This research also was restricted to a largely urban area. It may be that correlates of job satisfaction are substantially different in rural or isolated work situations. Implementation of the research on a province-wide basis is both desirable and feasible.  BIBLIOGRAPHY Attinson, Z., & Glassberg, E., (1980). After Graduation, What? Employment and Educational Experiences of Graduates of BSW Programs. Journal Of Education For Social Work, 19(1), 5-13. Barber, G., (1986). Correlates Of Job Satisfaction Among Human Service Workers. Administration In Social Work. 10(1), 25-38. Burstein, M., Tienhaara, N., Hewson, P., Warrander, B., (1975). Canadian Work Values (Catalogue No. MP33-6/1975). Ottawa: Information Canada. Goodwin, M., (1972). Correlates of Career Choice. Vancouver: U.B.C. Press. Gorrie, E. (1987). Employment Preferences Of Graduating B.S.W. Students (1987): A Study Of Employment Situation Factors, Employment Preferences, and Employment Expectations In British Columbia. Unpublished. Grunenberg, M. M., (1976). Job Satisfaction - A Reader. New York: Wiley & Sons. Haynes, K. S., (1983). Sexual Differences In Social Work Administrators' Job Satisfaction. Journal Of Social Service Research. 6(3/4), 57-74. Herrick, J., Takagi, C , Coleman, R., & Morgan, L., (1983). Social Workers Who Left The Profession: An Exploratory Study. Journal Of Sociology And Social Welfare. 10(1), 79-94. Herzberg, F., (1968). One More Time: How Do You Motivate Employees? Harvard Business Review. V. 66,53-62. Hopkins, A. H., (1983). Work and Job Satisfaction In The Public Sector. Totowa, New Jersey: Rowman & Allanheld.  136 Jayaratne, S., & Chess, W. A., (1984). Job Satisfaction, Burnout, And Turnover: A National Study. Social Work. September-October 1984, 448453. Jayaratne, S., & Chess, W. A., (1986). Job Satisfaction: A Comparison Of Caseworkers And Administrators. Social Work. March-April 1986,144146. Jayaratne, S., Chess, W. A., & Kunkel, D. A., (1986). Burnout: Its Impact On Child Welfare Workers And Their Spouses. Social Work. January-February 1986, 53-59. Kermish, I. & Kushin, F., (1969). Why High Turnover? Social Work Staff Losses In A County Welfare Department. Public Welfare. April 1969, 134-139. Kovach, K. A., (1977). Organization Size, Job Satisfaction, Absenteeism and Turnover. Washington: University Press of America. Krueger, M. A., (1986). Job Satisfaction For Child And Youth Care Workers. Washington: Child Welfare League of America. Labour Canada, (1973). Measuring The Quality Of Work Life: Proceedings Of A Symposium On Social Indicators Of Working Life. Ottawa: Alan Portigal. Labour Canada, (1973b). QWL Indicators - Prospects And Problems. Ottawa: Richard Walton. Labour Canada, (1981). Quality Of Working Life: QWL In The Federal Public Service. Ottawa: E. Trist & W. Westley. Locke, E. A., (1969). What Is Job Satisfaction? Organizational Behavior and Human Performance. 4:309-336. Maslow, A. H., (1970). Motivation and Personality. New York. Harper & Row.  McNeely, R. L., Feyerherm, W. FL, & Johnson, R. E., (1986). Services Integration And Job Satisfaction Reactions In A Comprehensive Human Resource Agency. Administration In Social Work. 10(1), 39-53. Ministry of Social Services and Housing. Annual Report. Victoria: Queen's Printer for British Columbia. Monette, D. R., Sullivan, T. J., & DeJong, C. R., (1986). Applied Social Research. New York: Holt, Rinehart and Winston. Organization For Economic Co-operation And Development, (1976). Towards The Measurement Of Work Satisfaction. Paris: Alan Portigal. Paul, W. J., and Robertson, K. B., (1970). Job Enrichment and Employee Motivation. London: Gower Press. Quinn, R. P., and Shepard, L. J., (1974). The 1972-73 Quality of Employment Survey, Ann Arbor: Institute for Social Research. University of Michigan. Quinn, R. P., and Staines, G. L., (1979). The 1977 Quality of Employment Survey, Ann Arbor: Institute for Social Research. University of Michigan. Smith, P. C , Kendall, L. M., & Hulin, C. L., (1969). The Measurement of Satisfaction in Work and Retirement. Chicago: Rand, McNally, & Company. Teare, R., (1981). Social Work Practice In A Public Welfare Setting. New York: Praeger.  138 CONTROL #  APPENDIX A  PART 1 DEMOGRAPHIC INFORMATION The purposes.  Dl.  following  questions  What i s your age range?  are  for  statistical  25  years  25 t o 29  years  .(2)  30 t o 34  years  .(3)  35 t o 39  years  .(4)  40  t o 44  years  .(5)  45 t o 49  years  _(6)  50 t o 54  years  -< >  55 t o 59  years  _(8)  Under  Over 59 D2.  Gender  analysis  7  years  _(9)  Male  _(D  Female  (2)  139 CONTROL # D3-7. D3.  D4.  D5.  D6.  D7.  D8.  D9.  Which of the following have you completed? High school graduation  No  (0)  Yes  (1)  No  (0)  Yes  (1)  No  (0)  Yes  (1)  No  (0)  Yes  (1)  No  (0)  Yes  (1)  No  (0)  Yes  (1)  Non-university post-secondary program  Non-social work bachelor l e v e l degree  Non-social work masters l e v e l degree  Bachelor of S o c i a l Work degree  Master of Social Work degree  How many years experience do you have working for any and a l l employer(s) as a c h i l d welfare s o c i a l worker? Years  D10  How many years experience do you have working for the p r o v i n c i a l government (including M . S . S . H . , M . H . R . , D.H.R.) as a c h i l d welfare s o c i a l worker? Years  140 CONTROL #  Dli  Approximately many hours per week do you work in your job as a c h i l d welfare s o c i a l worker? Hours  D12  How many hours per week do you believe is reasonable for a c h i l d protection s o c i a l worker to work? Hours  D13  D14  A r e you c u r r e n t l y , or have you a t any time d u r i n g the past 12 months been a member of a s o c i a l work p r o f e s s i o n a l a s s o c i a t i o n ? ( i . e . B.C.A.S.W. or e q u i v a l e n t i n another jurisdiction) No  (0)  Yes  (1)  Have you made any r e a l e f f o r t s d u r i n g the past 12 months t o f i n d other employment? (e.g. a p p l i e d for a job, a p p l i e d t o a t r a i n i n g program or e d u c a t i o n a l program, gone t o an i n t e r v i e w , spoken with another employer about working there) No  D15  D16  Do you b e l i e v e you w i l l be working i n t h i s agency two years from now?  ( 0)  Yes  (1)  No  (0)  Yes  (1)  How many people ( e x c l u d i n g y o u r s e l f ) a r e wholely or s i g n i f i c a n t l y dependant upon your income? e.g. spouse/partner (whether employed or n o t ) , c h i l d r e n , e t c . Number  141 CONTROL # PART 2 EMPLOYMENT FACTOR IMPORTANCE Social workers s e l e c t employment considering both s p e c i a l i z a t i o n (e.g. working with groups, i n d i v i d u a l s , communities; working with e l d e r l y , f a m i l i e s , c h i l d r e n , etc.) and factors related more to the p a r t i c u l a r employer or employment situation. Within the f i e l d of s o c i a l work and within p r a c t i c e s p e c i a l i z a t i o n s there are numerous employers each offering a d i f f e r e n t s e l e c t i o n of employment f a c t o r s . Factors which may be important to one person may be unimportant to another person. We each view some factors as very important, other factors as unimportant, and others f a l l somewhere in between. How important are the following factors to you regarding your employment? Please rate the importance of each of the following using the following s c a l e : 1 to 2  =  Not Important  3 to 4 =»  Somewhat Important  5 to 6  =  Moderately Important  7 to 8  =•  Quite Important  9 to 10 -  factors  Very important  C i r c l e any number which best unimportant each item i s to you.  expresses  how  important or  Example Factor A 1--NI--2  3--SI--4  5--MI--6  7--QI--8  9--VI--10  3--SI--4  5--MI--6  7--QI--8  9--VI--10  3--SI--4  5--MI--6  7--QI--8  9--VI--10  Factor B 1--NI--2 Factor C 1--NI--2  In t h i s example, Factor A was rated Somewhat Important, Factor B was rated Somewhat important, but s l i g h t l y higher, and Factor C was rated Quite important.  142 CONTROL # NI-Not Important; SI=Somewhat Important; MI-Moderately Important QI=Quite Important; VI=Very Important EMPLOYMENT FACTOR IMPORTANCE Fl.  Being paid w e l l . 1--NI--2  F2.  9--VI--10  3--SI--4  5--MI--6  7--QI--8  9--VI--10  3--SI--4  5--MI--6  7--QI--8  9--VI--10  3--SI--4  5--MI--6  7--QI--8  9--VI--10  3--SI--4  5--MI--6  7--QI--8  9--VI--10  3--SI--4  5--MI--6  7--QI--8  9--VI--10  Having the opportunity for promotion/advancement within the agency.  1--NI--2 F9.  7--QI--8  Getting a sense of accomplishment from the job. 1--NI--2  F8.  5--MI--6  Averaging no more than 35 hours of work per week. 1--NI--2  F7.  3--SI--4  Having a comfortable physical environment i n which to work. 1--NI--2  F6.  9--VI--10  Working for an agency with high s t a f f morale. 1--NI--2  F5.  7*-QI--8  Having a small amount of job-required paperwork. 1--NI--2  F4.  5--MI--6  Having high q u a l i t y supervision. 1--NI--2  F3.  3--SI--4  3--SI--4  5--MI--6  7--QI--8  9--VI--10  Working for an agency which has a p o s i t i v e public image. 1--NI--2  3--SI--4  5--MI--6  7--QI--8  9--VI--10  143 CONTROL # Nl=Not Important; SI=Somewhat Important; Ml=*Moderately Important QI=Quite Important; VI=*Very Important F10.  Having good job s e c u r i t y . 1--NI--2  F'll.  9--VI--10  3--SI--4  5--MI--6  7--QI--8  of  9--VI--10  3--SI--4  5--MI--6  3--SI--4  5--MI--6  7--QI--8  9--VI--10  a medical/dental p l a n , etc. 7--QI--8  9--VI--10  3--SI--4  5--MI--6  7--QI--8  9--VI--10  3--SI--4  5--MI--6  7--QI--8  9--VI--10  Having easy access to the workplace (e.g. a v a i l a b l e parking, convenient p u b l i c t r a n s p o r t a t i o n , e t c . ) .  1--NI--2 F18.  7--QI--8  Having agency support for professional development and ongoing t r a i n i n g .  1--NI--2 F17.  5--MI--6  Having a manageable caseload or workload s i z e . 1--NI--2  F16.  responsibilities.  Having good fringe benefits such as generous sick leave p r o v i s i o n s ,  1--NI--2 F15.  9--VI--10  Having the opportunity for contact with other professionals doing similar work i n other work s e t t i n g s .  1--NI--2 F14.  3--SI--4  7--QI--8  Having the opportunity to develop a s p e c i a l i z e d f i e l d practice.  1--NI--2 F13.  5--MI--6  Having a job with c l e a r l y defined 1--NI--2  F12.  3--SI--4  3--SI--4  5--MI--6  7--QI--8  9--VI--10  Having access to s u f f i c i e n t funds and resources to achieve c l i e n t related job goals.  1--NI--2  3--SI--4  5--MI--6  7--QI--8  9--VI--10  144 CONTROL # Nl=Not Important; Sl=»Somewhat Important; MI=Moderately Important QI=Quite Important; Vl=Very Important F19.  Having l i b e r a l annual vacation time. 1--NI--2  F20.  9--VI--10  3--SI--4  5--MI--6  7--QI--8  9--VI--10  3--SI--4  5--MI--6  7--QI--8  9--VI--10  3--SI--4  5--MI--6  7--QI--8  9--VI--10  3--SI--4  5--MI--6  7--QI--8  9--VI--10  3--SI--4  5--MI--6  7--QI--8  9--VI--10  Working for a supervisor who treats employees with 1--NI--2  F27.  7--QI--8  Being c l e a r l y able to influence decisions of the agency. 1--NI--2  F26.  5--MI--6  Having highly competent co-workers. 1--NI--2  F25.  3--SI--4  Having decision making autonomy within the agency. 1--NI--2  F24.  9--VI--10  Working for an agency which recognizes s o c i a l work as a profession.  1--NI--2 F23.  7--QI--8  Having easy access to high q u a l i t y workload or caseload consultation (e.g. with co-workers, supervisors, or consultants).  1--NI--2 F22.  5--MI--6  Having a job position with high status within the agency. 1--NI--2  F21.  3--SI--4  3--SI--4  5--MI--6  7--QI--8  Working for an agency which treats employees with  1--NI--2  3--SI--4  5--MI--6  7--QI--8  respect. 9--VI--10 respect.  9--VI--10  145 CONTROL # PART 3 EMPLOYMENT FACTOR SATISFACTION Many factors influence how much job s a t i s f a c t i o n people experience. These include what employment s i t u a t i o n i s , i . e . the employment situation f a c t o r s . The employment s i t u a t i o n factors focused on in this research are those elements of the employment s i t u a t i o n which are, or may be modifiable. Each of the following questions i d e n t i f i e s a factor about their employment which s o c i a l workers may find s a t i s f y i n g or dissatisfying. of  Please rate how s a t i s f i e d or d i s s a t i s f i e d you are the following factors using the following s c a l e :  1 to 2  -  Very D i s s a t i s f i e d  3 to 4  =  Somewhat D i s s a t i s f i e d  5 to 6 7 to 8  with each  Neither S a t i s f i e d Nor D i s s a t i s f i e d =  9 to 10 =  Somewhat S a t i s f i e d Very S a t i s f i e d  Circle dissatisfied situation.  any number which best expresses how s a t i s f i e d or you are with that aspect of your employment  Factor A 1--VD- -2  •SD-  5-NSND-6  7- -SS--8  9--VS--10  3--SD--4  5-NSND-6  7--SS--8  9--VS--10  3--SD--4  5-NSND-6  SS- -8  9--VS--10  Factor B 1--VD- -2 Factor C 1- -VD- -2  In this example, the respondent reported being Somewhat S a t i s f i e d with Factor A, Somewhat S a t i s f i e d , but higher, with Factor B, and Neither S a t i s f i e d Nor D i s s a t i s f i e d with Factor C.  146 CONTROL # VD=Very D i s s a t i s f i e d ; SD=Somewhat D i s s a t i s f i e d ; NSND=Neither S a t i s f i e d Nor D i s s a t i s f i e d ; SS=Somewhat S a t i s f i e d ; VS=Very S a t i s f i e d EMPLOYMENT FACTOR SATISFACTION 50.  A l l things considered, how s a t i s f i e d would you say you are with your job?  1--VD--2  3--SD--4  How s a t i s f i e d following? 51.  3--SD--4  5-NSND-6  7--SS--8  9--VS--10  3--SD--4  5-NSND-6  7--SS--8  9--VS--10  3--SD--4  5-NSND-6  7--SS--8  9--VS--10  3--SD--4  5-NSND-6  7--SS--8  9--VS--10  3--SD--4  5-NSND-6  7--SS--8  9--VS--10  The number of hours you a c t u a l l y work per week. 1--VD--2  S7.  with each of the  The physical environment i n which you work. 1--VD--2  56.  you  The l e v e l of s t a f f morale i n the agency. 1--VD--2  55.  are  The amount of your job-required paperwork. 1--VD--2  54.  dissatisfied  9--VS--10  The q u a l i t y of supervision you r e c e i v e . 1--VD--2  53.  or  7--SS--8  The amount you are paid. 1--VD--2  52.  5-NSND-6  3--SD--4  5-NSND-6  7--SS--8  9--VS--10  The sense of accomplishment you get from your job. 1--VD--2  3--SD--4  5-NSND-6  7--SS--8  9--VS--10  147 CONTROL # VD=Very D i s s a t i s f i e d ; SD=»Somewhat D i s s a t i s f i e d ; NSND=Neither S a t i s f i e d Nor D i s s a t i s f i e d ; SS=Somewhat S a t i s f i e d ; VS=Very S a t i s f i e d 58.  Your opportunities for promotion agency.  1--VD--2 59.  3--SD--4  9--VS--10  5-NSND-6  7--SS--8  9--VS--10  5-NSND-6  3--SD--4  5-NSND-6  3--SD--4  5-NSND-6  3--SD--4  3--SD--4  are.  7--SS--8 a  9--VS--10  specialized  7--SS--8 other  7--SS--8  f i e l d of 9--VS--10  professionals 9--VS--10  such as medical/dental plan, sick  Your caseload/workload 1--VD--2  516  3--SD--4  Your fringe benefits p r o v i s i o n s , etc.  1--VD--2 515  7--SS--8  Your opportunities for contact with doing s i m i l a r work i n other s e t t i n g s .  1--VD--2 514  9--VS--10  5-NSND-6  Your opportunities to develop p r a c t i c e within the agency.  1--VD--2 513  3--SD--4  How c l e a r l y defined your job r e s p o n s i b i l i t i e s 1--VD--2  512  7--SS--8  Your job s e c u r i t y . 1--VD--2  511  5-NSND-6  within the  The public image of the agency. 1--VD--2  510  3--SD--4  or advancement  5-NSND-6  7--SS--8  leave  9--VS--10  size. 5-NSND-6  7--SS--8  9--VS--10  The l e v e l of agency support for professional development and ongoing t r a i n i n g .  1--VD--2  3--SD--4  5-NSND-6  7--SS--8  9--VS--10  148  CONTROL # VD=Very D i s s a t i s f i e d ; SD=Somewhat D i s s a t i s f i e d ; NSND=Neither S a t i s f i e d Nor D i s s a t i s f i e d ; SS=Somewhat S a t i s f i e d ; VS=Very S a t i s f i e d 517  The access to your workplace ( e . g . convenient public transportation, e t c . ) .  1--VD--2 518  9--VS--10  3--SD--4  5-NSND-6  7--SS--8  9--VS--10  3--SD--4  3--SD--4  5-NSND-6  7--SS--8  9--VS--10  to high q u a l i t y workload or caseload with co-workers,.. supervisors, or 5-NSND-6  7--SS--8  9--VS--10  3--SD--4  5-NSND-6  7--SS--8  9--VS--10  3--SD--4  5-NSND-6  7--SS--8  9--VS--10  The l e v e l of competence of your co-workers. 1--VD--2  S25  7--SS--8  Your d e c i s i o n making autonomy within the agency. 1--VD--2  524  5-NSND-6  client  The l e v e l of recognition of s o c i a l work as a profession by the agency.  1--VD--2 523  3--SD--4  The access you have consultation (e.g. consultants).  1--VD--2 522  9--VS--10  The status given your job within the agency. 1--VD--2  521  7--SS--8  The amount of your annual vacation time. 1--VD--2  520  5-NSND-6  The access you have to funds and resources to achieve r e l a t e d job goals.  1--VD--2 519  3--SD--4  a v a i l a b l e parking,  3--SD--4  5-NSND-6  7--SS--8  9--VS--10  Your a b i l i t y to influence decisions of the agency. 1--VD--2  3--SD--4  5-NSND-6  7--SS--8  9--VS--10  149 CONTROL # VD=Very D i s s a t i s f i e d ; SD=Somewhat D i s s a t i s f i e d ; NSND=Neither S a t i s f i e d Nor D i s s a t i s f i e d ; SS=Somewhat S a t i s f i e d ; VS=Very S a t i s f i e d 526  The respect with which your supervisor treats  1--VD--2 527  3--SD--4  5-NSND-6  7--SS--8  The respect with which your agency treats  1--VD--2  3--SD--4  5-NSND-6  employees. 9--VS--10  employees.  7--SS--8  9--VS--10  150 CONTROL # PART 4 EMPLOYMENT PREFERENCES  Please c i r c l e you think each of a career s e t t i n g . of working in each  the number which best describes how a t t r a c t i v e the following employment settings would be for i . e . how much would you l i k e to make a career of the following s e t t i n g s .  Please use the following  scale:  1 to 2  =» Very Unattractive  3 to 4  = Somewhat Unattractive  5 to 6  = Neither A t t r a c t i v e Nor Unattractive  7 to 8  - Somewhat A t t r a c t i v e  9 to 10 = Very A t t r a c t i v e  Pi.  C h i l d welfare  1--VU--2 P2.  P3.  5-NANU-6  7--SA--8  9--VA--10  • Non BC  3--SU--4  3--SU--4  5-NANU-6  7--SA--8  9--VA--10  3--SU--4  5-NANU-6  7--SA--8  9--VA--10  3--SU--4  5-NANU-6  7--SA--8  9--VA--10  5-NANU-6  7--SA--8  9--VA--10  7--SA--8  9--VA--10  Hospitals - Non BC  1--VU--2 P7.  9--VA--10  Hospitals - BC  1--VU--2 P6.  7--SA--8  Mental health - Non BC  1--VU--2 P5.  5-NANU-6  Mental health - BC Government (Min. of Health)  1--VU--2 P4.  3--SU--4  C h i l d welfare  1--VU--2  - BC Government (MSSH)  3--SU--4  Non p r o f i t s o c i e t i e s - BC  1--VU--2  3--SU--4  5-NANU-6  151 CONTROL # P8.  Non p r o f i t s o c i e t i e s - Non BC  1--VU--2 P9.  5-NANU-6  7--SA--8  9--VA--10  5-NANU-6  7--SA--8  9--VA--10  Unemployed  1--VU--2 PlO.  3--SU--4  3--SU--4  Any other area of employment or other a c t i v i t y .  (Please Specify) 1--VU--2  3--SU--4  5-NANU-6  7--SA--8  9--VA--10  152 CONTROL # PART 5 EMPLOYMENT EXPECTATIONS Where do from now?  you r e a l i s t i c a l l y  expect to  be working two years  C i r c l e the number which best describes the l i k e l i h o o d of you working in each of the following s i t u a t i o n s two years from now. Please use the following  scale:  1 to 2  => Very U n l i k e l y  3 to 4  = Somewhat U n l i k e l y  5 to 6  - Neither L i k e l y Nor U n l i k e l y  7 to 8  » Somewhat L i k e l y  9 to 10 « Very L i k e l y XI.  C h i l d welfare - BC Government (MSSH)  1--VU--2 X2.  3--SU--4  3--SU--4  5-NLNU-6  7--SL--8  9--VL--10  Health)  5-NLNU-6  7--SL--8  9--VL--10  5-NLNU-6  7--SL--8  9--VL--10  5-NLNU-6  7--SL--8  9--VL--10  5-NLNU-6  7--SL--8  9--VL--10  7--SL--8  9--VL--10  - BC  3--SU--4  Hospitals  1--VU--2 X7.  3--SU--4  Hospitals  1--VU--2 X6.  9--VL--10  Mental health - Non BC  1--VU--2 X5.  7--SL--8  Mental health - BC Government (Min. of  1--VU--2 X4.  5-NLNU-6  C h i l d welfare - Non BC  1--VU--2 X3.  3--SU--4  - Non BC  3--SU--4  Non p r o f i t s o c i e t i e s - BC  1--VU--2  3--SU--4  5-NLNU-6  153 CONTROL # X8.  Non p r o f i t s o c i e t i e s - Non BC  1--VU--2 X9.  5-NLNU-6  7--SL--8  9--VL--10  5-NLNU-6  7--SL--8  9--VL--10  Unemployed  1--VU--2 X10.  3--SU--4  3--SU--4  Any other area of employment or other  activity.  (Please Specify) 1--VU--2  3--SU--4  5-NLNU-6  7--SL--8  9--VL--10  154 CONTROL # PART 6 PLEASE FEEL FREE TO COMPLETE PART 6 NOW AND RETURN IT NOW, OR COMPLETE IT LATER AND MAIL IT TO ME WITHIN A FEW DAYS USING THE ENVELOPE PROVIDED I am also interested i n any observations you may have about why s o c i a l workers may value d i f f e r e n t employment s i t u a t i o n factors d i f f e r e n t l y with d i f f e r e n t lengths of experience. Some of the changes may have to do with experiences with c l i e n t s , some may have to do with experiences working within the agency, some may have to do with changes within the person, or there may be other f a c t o r s . Do you think that a l l of the employment s i t u a t i o n factors that are important to you now were equally important to you just before you entered c h i l d welfare? or do you think some of them are d i f f e r e n t ? A l l are the same Some.are d i f f e r e n t If some of them are d i f f e r e n t , which ones have changed, and what happened to cause you to put more value on some things and l e s s on others? (Use a d d i t i o n a l pages i f necessary.)  155 CONTROL #  156 APPENDIX B RESEARCH TERMS AND VARIABLES  W i t h i n the t e x t ,  each of the f o l l o w i n g v a r i a b l e s  except  SOVER, may be found prepended with an " F " , an "S", or a "D".  Terms which are not prepended r e f e r to the  concepts.  Terms which are prepended r e f e r to  which r e l a t e variables),  to the importance of that concept satisfaction  w i t h the concept  the d i s c r e p e n c y between the with a concept  following  variables (F  (S v a r i a b l e s ) ,  importance of and  satisfaction  (D v a r i a b l e s ) .  See Appendix A f o r exact wording of each of  the  questions.  SOVER - f a c e t - f r e e measure of  job  PAY - being w e l l p a i d f o r the  job  SUPER - having high q u a l i t y  satisfaction  supervision  PAPER - having l i t t l e paperwork MORALE - having a w o r k s i t e w i t h h i g h morale PHYSIC  - having a p h y s i c a l l y comfortable work environment  HOURS - being s a t i s f i e d  w i t h the hours of work per week  ACCOMP - g e t t i n g a sense of accomplishment from work PROMO - having o p p o r t u n i t i e s  f o r career  advancement  IMAGE - working f o r an agency with a good p u b l i c JOBSEC - having good job CLEAR - c l e a r l y d e f i n e d  security job  responsibilities  or  image  157 SPECIAL  - ability  CONTACT  - contact with other  FRINGE  - having  WRKLOAD - h a v i n g PRODEV - h a v i n g ACCESS  to develop  a  specialization  professionals  good employment a manageable support  - convenience  benefits  workload  for training  of workplace  access  FUNDS - a c c e s s t o r e s o u r c e s f o r c l i e n t s VAC  - having  STATUS  liberal  vacation  - h a v i n g h i g h s t a t u s w i t h i n the agency  CONSULT - a c c e s s PROFESS  to quality  - recognition  consultation  of s o c i a l  AUTO - h a v i n g d e c i s i o n m a k i n g COWORK - h a v i n g INFLU  time  competent  work  autonomy  co-workers  - b e i n g a b l e t o i n f l u e n c e d e c i s i o n s i n t h e agency  SUPRESP  - s u p e r v i s o r t r e a t s workers with  respect  AGERESP  - agency t r e a t s workers w i t h r e s p e c t  158  INTRODUCTION TO BE READ TO PARTICIPANTS A MESSAGE FROM ERNIE GQRRIE, MSW  STUDENT AND MSSH SOCIAL WORKER  Ernie Gorrie i s a s o c i a l worker working on h i s M.S.W. t h e s i s .  i n t h i s region, c u r r e n t l y  His research i s i n the area of job  s a t i s f a c t i o n among c h i l d welfare  l i n e s o c i a l workers.  His  research i s independent of the m i n i s t r y , but has the support of the  deputy-minister. He i s asking you to complete a questionnaire regarding your  job s a t i s f a c t i o n and f a c t o r s which r e l a t e to your o v e r a l l job satisfaction.  You w i l l have the opportunity to complete the  questionnaire during working hours, that i s , during t h i s meeting. During t h i s meeting, you w i l l be asked to return to your o f f i c e s , or to i n d i v i d u a l rooms, to complete the questionnaires. have 30 minutes to complete the questionnaires.  You w i l l  More time may  be  taken i f needed. Ernie thanks you f o r c o n s i d e r i n g p a r t i c i p a t i n g i n t h i s research.  You are under no o b l i g a t i o n to p a r t i c i p a t e i n t h i s  research.  P a r t i c i p a t i o n by as many s o c i a l workers as p o s s i b l e  w i l l increase the value of the research however, and increase the p r o b a b i l i t y that i t w i l l have i n f l u e n c e on the m i n i s t r y . Extensive measures have been taken to ensure c o n f i d e n t i a l i t y for p a r t i c i p a n t s i n the research.  Numbered questionnaires have  been randomly assigned to envelopes.  These envelopes have then  been randomly a l l o c a t e d to d i s t r i c t o f f i c e s .  I t would not have  been p o s s i b l e to record which o f f i c e s received which questionnaires.  It would not be p o s s i b l e to trace any numbered  160 INTRODUCTION AND INSTRUCTION SHEET RESEARCH TOPIC: FACTORS INFLUENCING OVERALL JOB SATISFACTION AMONG CHILD PROTECTION LINE SOCIAL WORKERS  Thank you f o r c o n s i d e r i n g p a r t i c i p a t i n g i n t h i s project.  Please  complete t h i s  be aware t h a t you are under no o b l i g a t i o n t o  questionnaire.  This research student  research  i s being  conducted by E r n i e G o r r i e , an M.S.W.  a t the U n i v e r s i t y o f B r i t i s h Columbia, and i s b e i n g  overseen by a f a c u l t y committee.  E r n i e G o r r i e c a n be c o n t a c t e d  by telephone a t 271-3083 t o answer q u e s t i o n s  regarding the  research. This research  i s independent o f the M i n i s t r y o f S o c i a l  S e r v i c e s and Housing, b u t i s supported by M.S.S.H. of t h i s r e s e a r c h may have s i g n i f i c a n t  The r e s u l t s  impact on some d e c i s i o n s  made w i t h i n the m i n i s t r y . The  purpose o f t h e r e s e a r c h  i s t o determine which  elements o f employment c o r r e l a t e with o v e r a l l  specific  job s a t i s f a c t i o n  among c h i l d p r o t e c t i o n d i r e c t p r a c t i c e s o c i a l workers. Statistical analysis w i l l  i n c l u d e determining  i n v a r i o u s demographic v a r i a b l e s (e.g. years  whether d i f f e r e n c e s o f e x p e r i e n c e ) have  an impact on these c o r r e l a t i o n s . If you chose t o p a r t i c i p a t e , you w i l l Data Sheet, and t h e q u e s t i o n n a i r e the q u e s t i o n n a i r e Sheets t o a lawyer  provided.  complete the C o n t r o l You w i l l  then  return  t o me i n one envelope and send t h e C o n t r o l (Mr.  D. Higham) i n a separate  envelope.  Data This  161 method was d e v e l o p e d t o ensure maximum c o n f i d e n t i a l i t y f o r respondents. All  questionnaires w i l l  No q u e s t i o n n a i r e w i l l No p e r s o n a l  be c o n f i d e n t i a l  information.  be made a v a i l a b l e t o t h e m i n i s t r y .  identification  i s i n c l u d e d on any q u e s t i o n n a i r e .  No d e c i s i o n r e g a r d i n g your employment s t a t u s w i t h the M i n i s t r y o f S o c i a l S e r v i c e s and Housing w i l l  be i n f l u e n c e d i n any  way by your d e c i s i o n t o p a r t i c i p a t e or n o t p a r t i c i p a t e i n t h i s research.  Your q u e s t i o n n a i r e w i l l  be used f o r r e s e a r c h  purposes  only. PLEASE DO NOT PUT YOUR NAME ON ANY PAGE OF THE QUESTIONNAIRE EXCEPT THE CONTROL DATA SHEET. You have t h e r i g h t t o d e c l i n e t o p a r t i c i p a t e i n t h i s research.  You have the r i g h t t o withdraw  any time.  You have the r i g h t t o d e c l i n e t o answer any or a l l  questions  from t h e r e s e a r c h a t  i n the q u e s t i o n n a i r e .  T y p i c a l time t o complete p a r t s 1 t o 5 of t h e q u e s t i o n n a i r e (multiple choice parts)  i s 30 minutes.  complete P a r t s 6 and 7 w i l l now or l a t e r .  vary.  The time r e q u i r e d to  These p a r t s may be completed  165 APPENDIX D GRAPH OF SELECTED DEMOGRAPHIC VARIABLES AGE Value L a b e l  Value  Under 25 Years 25 to 29 Years 30 to 34 Years 35 to 39 Years 40 to 44 Years 45 to 49 Years 50 to 54 Years 55 to 59 Years  Frequency  1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00  TOTAL  3 11 7 11 12 2 1 1 -9.00 49  Percent 6.1 22.4 14.3 22.4 24.5 4.1 2.0 2.0 1 100.0  Valid Cum Percent Percent 6.3 22.9 14.6 22.9 25.0 4.2 2.1 2.1 2.0 100.0  < 25 25-29 30-34 35-39 40-44 45-49 50-54 55-59  V a l i d Cases  48  M i s s i n g Cases  1  6.3 29.2 43.8 66.7 91.7 95.8 97.9 100.0 MISSING  166 GENDER Value Label Male Female  Value 1.00 2.00 9.00 TOTAL  Valid Cases  ASSN  48  Frequency  Percent  Valid Percent  Cum Percent  15 33 1  30.6 67.3 2.0  31.3 68.8 MISSING  31.3 100.0  49  100.0  Missing Cases  100.0  1  Membership in Professional Association  Value Label No Yes  Valid Cases  Value .00 1.00  49  Frequency 38 11  Percent  Valid Cum Percent Percent  77.6 22.4  Missing Cases  77.6 22.4  0  77.6 100.0  167  ATTEMPT  E f f o r t Made In Past Year To F i n d Other  Value L a b e l No Yes  I I. V a l i d Cases  Frequency  Percent  Valid Percent  .00 1.00  26 23  53.1 46.9  53.1 46.9  TOTAL  49  100.0  100.0  6  0  BELIEVE Years  Value  49  12  18  Value  Cum Percent 53.1 100.0  .I  .1  30  24  M i s s i n g Cases  B e l i e v e W i l l Be Working In T h i s Agency  Value Label No Yes  Employment  Frequency  Percent  In Two  Valid Percent  .00 1.00  17 32  34.7 65.3  34.7 65.3  TOTAL  49  100.0  100.0  Cum Percent 34.7 100.0  168  DEPEND  Number of Economic Dependants  Value Label  Value  Frequency Percent  Valid Percent  Cum Percent 34.7 53.1 65.3 91.8 100.0  .00 1.00 2.00 3.00 4.00  17 9 6 13 4  34.7 18.4 12.2 26.5 8.2  34.7 18.4 12.2 26.5 8.2  TOTAL  49  100.0  100.0  169 APPENDIX E DESCRIPTIVE LISTING OF F (IMPORTANCE) VARIABLES FOR SOCIAL WORKERS Number of V a l i d Observations Variable  FPAY  7.929 .774 -.810 7.000 10.00  Observations  Variable  8.143 1.972 -1.248 8.500 10.00  V a l i d Observations  Variable  Valid  - 49  5.776 -.656 -.130 9.000 10.00  V a l i d Observations  Mean Kurtosis Skewness Range Maximum  49.00  Missing  Observations  Supervision  Std Dev S . E . Kurt S . E . Skew Minimum Missing  1.575 .668 .340 3.00  1.857 .668 .340 1.50  Observations  Imp. o f Small Amt. of Paperwork  FPAPER  Mean Kurtosis Skewness Range Maximum  Variable  49  Std Dev S . E . Kurt S . E . Skew Minimum  Imp. of Q u a l i t y  FSUPER  Mean Kurtosis Skewness Range Maximum  =  Imp. of Pay  Mean Kurtosis Skewness Range Maximum Valid  (Listwise)  - 49  Std Dev S . E . Kurt S . E . Skew Minimum M i s s i n g Observations  Imp. of High S t a f f  FMORALE  8.469 .363 -.890 5.000 10.00  Observations  49  - 0  Morale  Std Dev S . E . Kurt S . E . Skew Minimum Missing  2.354 .668 .340 1.00  Observations  1.397  .668 .340 5.00  0  170 D e s c r i p t i v e L i s t i n g Of F V a r i a b l e s For S o c i a l Workers Number of V a l i d Observations Variable Enviro  FPHYSIC  Mean Kurtosis Skewness Range Maximum  7.061 .937 -.715 8.500 10.00  FHOURS  FACCOMP  Observations  Variable Mean Kurtosis Skewness Range Maximum  49  Std Dev S . E . Kurt S . E . Skew Minimum M i s s i n g Observations  1.839 .668 .340 1.50 - 0  Std Dev S . E . Kurt S . E . Skew Minimum Missing  2.089 .668 .340 1.50  Observations  Imp. of Sense of Accomplishment  8.510 10.344 -2.564 9.000 10.00  Mean Kurtosis Skewness Range Maximum Valid  - 49  7.429 - .029 - .695 8.500 10.00  Observations  Variable  49.00  Imp. of Work Being <=35 Hours  Mean Kurtosis Skewness Range Maximum Valid  =  Imp. of Comfortable P h y s i c a l Work  V a l i d Observations  Variable  (Listwise)  49  Std Dev S . E . Kurt S . E . Skew Minimum M i s s i n g Observations  1.566 .668 .340 1.00 - 0  Imp. o f O p p o r t u n i t y f o r Promotion  FPROMO  7.469 1.827 -1.497 9.000 10.00  V a l i d Observations  - 49  Std Dev S . E . Kurt S . E . Skew Minimum Missing  Observations  2.130 .668 .340 1.00 0  171  Descriptive Listing Of F Variables For Social Workers Number of Valid Observations (Listwise) = Variable im  FIMAGE  Imp. of Agency Having Positive Public 6.041 -.562 -.338 9.000 10.00  Mean Kurtosis Skewness Range Maximum  Valid Observations - 49  Variable  FJOBSEC  Mean Kurtosis Skewness Range Maximum  49.00  Std Dev S.E. Kurt S.E. Skew Minimum Missing Observations  Imp. of Having Good Job Security 7.388 - .316 - .685 7.000 10.00  Std Dev S.E. Kurt S.E. Skew Minimum  49  Variable FRESP Responsib  Imp. of Having Clearly Defined 6.959 - .444 - .374 7.000 10.00  Valid Observations - 49 Variable Special Mean Kurtosis Skewness Range Maximum  FSPECIAL  1.985 .668 .340 3.00  Missing Observations  Valid Observations  Mean Kurtosis Skewness Range Maximum  2.270 .668 .340 1.00  Std Dev S.E. Kurt S.E. Skew Minimum  1.577 .668 .340 3.00  Missing Observations  Imp. of Opportunity for Practice  6.694 .873 - .721 9.000 10.00  Valid Observations - 49  Std Dev S.E. Kurt S.E. Skew Minimum  1.957 .668 .340 1.00  Missing Observations - 0  172  Descriptive Listing Of F Variables For Social Workers Number of Valid Observations (Listwise) = Variable Oth Mean Kurtosis Skewness Range Maximum  FCONTACT  Imp. of Opportunity for Contact With  7.082 3.203 -1.292 9.000 10.00  Valid Observations - 49  Variable  FFRINGE  Mean Kurtosis Skewness Range Maximum  49.00  1.602 .668 .340 1.00  Std Dev S.E. Kurt S.E. Skew Minimum  Missing Observations - 0  Imp. of "Fringe Benefits" 7.939 - .203 - .607 6.000 10.00  Std Dev S.E. Kurt S.E. Skew Minimum  1.543 .668 .340 4.00  Missing Observations  Valid Observations  49  Variable  Imp. of Having Managable Workload  FWKSIZE  8.653 9.923 2.625 9.000 10.00  Mean Kurtosis Skewness Range Maximum  Std Dev S.E. Kurt S.E. Skew Minimum  1.611 .668 .340 1.00  Missing Observations  Valid Observations  49  Variable Develop  Imp. of Agency Support for Prof.  Mean Kurtosis Skewness Range Maximum  FPRODEV  8.694 -.471 -.563 4.000 10.00  Valid Observations  49  Std Dev S.E. Kurt S.E. Skew Minimum Missing Observations  1.060 .668 .340 6.00 0  173 D e s c r i p t i v e L i s t i n g Of F V a r i a b l e s For S o c i a l Workers Number of V a l i d Observations Variable  FACCESS  Valid  Observations  Variable  FFUNDS  FVAC  Valid  Mean Kurtosis Skewness Range Maximum Valid  1.973 .668 .340 2.00  Std Dev S . E . Kurt S . E . Skew Minimum Missing  Observations  1.611 .668 .340 3.00  Std Dev S . E . Kurt S . E . Skew Minimum Missing  Observations  Imp. of L i b e r a l V a c a t i o n Time  Observations  Variable  - 49  8.143 2.372 1.476 8.500 10.00  Mean Kurtosis Skewness Range Maximum  49.00  Imp. o f Funds/Resources  V a l i d Observations  Variable  49  8.347 2.326 1.519 7.000 10.00  Mean Kurtosis Skewness Range Maximum  =  Imp. of Easy Access to Workplace  6.367 - .769 - .197 8.000 10.00  Mean Kurtosis Skewness Range Maximum  (Listwise)  49  Std Dev S . E . Kurt S . E . Skew Minimum Missing  1.890 .668 .340 1.50  Observations  Imp. of Intra-Agency Job Status  FSTATUS  5.490 -.500 -.249 7.500 9.00  Observations  49  Std Dev S . E . Kurt S . E . Skew Minimum M i s s i n g Observations  1.924  .668 .340 1.50  - 0  174 D e s c r i p t i v e L i s t i n g Of F V a r i a b l e s Number o f V a l i d Variable  Observations  FCONSULT  Valid  4.000  FPROFESS  Observations  Variable  Kurtosis Skewness  Range  Maximum Valid  - 49  FCOWORK  49  Imp.  8.082  Mean  S t d Dev S.E. K u r t S.E. Skew Minimum  1.186 .668 .340  6.00  Missing Observations  -0  Imp. o f A g e n c y R e c o g n i t i o n o f SW a s  7.776 .028 .112 4.500 10.00  Observations  Variable  49.00  1.814 .668 .340 2.00  S t d Dev S.E. K u r t S.E. Skew Minimum  Missing Observations  -0  Imp. o f D e c i s i o n M a k i n g Autonomy  FAUTO  Mean Kurtosis Skewness Range Maximum Valid  - 49  8.306 3.083 1.688 8.000 10.00  Mean Kurtosis Skewness Range Maximum Valid  »  10.00  Observations  Variable Prof  (Listwise)  Imp. o f A c c e s s t o C o n s u l t a t i o n  8.571 -.553 -.529  Mean Kurtosis Skewness Range Maximum  F o r S o c i a l Workers  .445 -.390  5.000  S t d Dev S.E. K u r t S.E. Skew Minimum Missing  1.041 .668 .340 5.50  Observations  o fCompetent  Co-Workers  S t d Dev S.E. K u r t S.E. Skew Minimum  1.183 .668 .340 5.00  10.00  Observations  49  Missing  Observations  0  175 D e s c r i p t i v e L i s t i n g Of F V a r i a b l e s For S o c i a l Workers Number of V a l i d Observations Variable Decis  7.449 -.351 -.574 6.500 10.00  V a l i d Observations - 49  Mean Kurtosis Skewness Range Maximum Valid  9.122 1.337 -1.320 4.000 10.00  Observations  Variable  Std Dev S.E. Kurt S.E. Skew Minimum  1.656 .668 .340 3.50  Missing Observations  Imp. o f S u p e r v i s o r Respect f o r  Variable FSUPRES Employees  Valid  49.00  Imp.of A b i l i t y t o I n f l u e n c e Agency  FINFLU  Mean Kurtosis Skewness Range Maximum  Mean Kurtosis Skewness Range Maximum  (Listwise) •  FAGERESP  49  Missing  1.008 .668 .340 6.00  Observations  Imp. of Agency Respect f o r Employees  8.816 1.710 1.449 5.000 10.00  Observations  Std Dev S.E. Kurt S.E. Skew Minimum  49  Std Dev S.E. Kurt S.E. Skew Minimum Missing  Observations  1.314 .668 .340 5.00  176 Appendix F F (IMPORTANCE) VARIABLES LISTED IN ORDER OF MEANS Variable  Mean  S..D.  Lowest  FSUPRES FAGERESP FPRODEV FWKSIZE FCONSULT FACCOMP FMORALE FFUNDS FPROFESS FVAC FSUPER FCOWORK FFRINGE FPAY FAUTO FPROMO FINFLU FHOURS FJOBSEC FCONTACT FPHYSIC FRESP FSPECIAL FACCESS FIMAGE FPAPER FSTATUS  9. 12 8. 82 8. 69 8. 65 8. 57 8. 51 8. 47 8. 35 8. 31 8. 14 8. 14 8. 08 7. 94 7. 93 7. 78 7. 47 7. 45 7. 43 7. 39 7. 08 7. 06 6. 96 6. 69 6. 37 6. 04 5. 78 5. 49  1..01 1..31 1,.06 1..61 1..19 1..57 1..40 1..61 1..81 1..89 1..86 1..18 1..54 1..57 1..04 2..13 1..66 2..09 1..99 1..60 1..84 1.,58 1.,96 1.,97 2.,27 2.,35 1. 92  6. 00 5. 00 6. 00 1. 00 6. 00 1. 00 5. 00 3. 00 2. 00 1. 50 1. 50 5. 00 4. 00 3. 00 5. 50 1. 00 3. 50 1. 50 3. 00 1. 00 1. 50 3. 00 1. 00 2. 00 1. 00 1. 00 1. 50  Highest  N  10 .00 10 .00 10 .00 10 .00 10 .00 10 .00 10 .00 10 .00 10 .00 10 .00 10 .00 10 .00 10 .00 10 .00 10 .00 10 .00 10 .00 10 .00 10 .00 10 .00 10 .00 10 .00 10 .00 10 .00 10 .00 10 .00 9 .00  49 49 49 49 49 49 49 49 49 49 49 49 49 49 49 49 49 49 49 49 49 49 49 49 49 49 49  177 APPENDIX G GRAPHS OF F (IMPORTANCE)  VARIABLES RECODED TO FIVE LEVELS  Nl = Not Important; SI = Somewhat Important; MI = Moderately Important; QI = Quite Important; VI = Very Important FPAY  Importance of Pay  Value Label  Value  2.00 3.00 4.00 5.00 TOTAL  V a l i d Cases  49  Frequency  Percent  Valid Percent  1  2.0 14.3 44.9 38.8  2.0 14.3 44.9 38.8  7 22 19  100.0  M i s s i n g Cases  100.0  0  Cum Percent  2. 16.3 61.2 100.0  178 Nl = Not Important; SI - Somewhat Important; MI = Moderately Important; QI = Q u i t e Important; VI = Very Important FSUPER  Importance of Q u a l i t y of S u p e r v i s i o n  Value L a b e l  NI SI MI QI VI  Value  Frequency  1.00 3.00 4.00 5.00  8 15 25  TOTAL  49  1  Percent  2.0 16.3 30.6 51.0 100.0  1.00 i i i 1 2.00 I 0 3.00 4.00 5.00  Valid  Cases  49  M i s s i n g Cases  Valid Percent  2.0 16.3 30.6 51.0 100.0  Cum Percent  2. 18.4 49.0 100.0  179 NI = Not Important; SI = Somewhat Important; MI = Moderately Important; QI = Quite Important; VI » Very Important FPAPER  Importance of Small Amount of Paperwork  Value L a b e l  Value  1.00 2.00 3.00 4.00 5.00 TOTAL  Valid  Cases  FMORALE  49  Percent  Valid Percent  5  10.2 20.4 30.6 24.5 14.3  10.2 20.4 30.6 24.5 14.3  10 15 12 7  100.0  49  Value  3.00 4.00 5.00 TOTAL  Cases  49  10. 30.6 61.2 85.7 100.0  100.0  Morale  Frequency  Percent  Valid Percent  4  8.2 36.7 55.1  8.2 36.7 55.1  18 27 49  100.0  1.00 2.00 3.00 4.00 5.00  Valid  Cum Percent  M i s s i n g Cases  Importance of High S t a f f  Value L a b e l  NI SI MI QI VI  Frequency  M i s s i n g Cases  100.0  Cum Percent  8. 44.9 100.0  180 Nl = Not Important; SI = Somewhat Important; MI = Moderately Important; QI = Q u i t e Important; VI • Very Important FPHYSIC  Importance of Comfortable P h y s i c a l Environment  Value Label  Nl SI MI QI VI  Value  Frequency  Percent  Valid Percent  1  2.0 4.1 28.6 46.9 18.4  2.0 4.1 28.6 46.9 18.4  1.00 2.00 3.00 4.00 5.00  2 14 23 9  TOTAL  49  100.0  1.00 2.00 3.00 4, 00 5, 00  V a l i d Cases  49  M i s s i n g Cases  100.0  Cum Percent 2. 6.1 34.7 81.6 100.0  181 NI = Not Important; SI = Somewhat Important; MI = Moderately Important; QI = Quite Important; VI = Very Important FHOURS  Importance of Working No More Than 35 Hours  Value Label  Value  1.00 2.00 3.00 4.00 5.00 TOTAL NI SI MI QI VI  Frequency  Percent  Valid Percent  Cum Percent  1  2.0 6.1 22.4 30.6 38.8  2.0 6.1 22.4 30.6 38.8  2.0 8.2 30.6 61.2 100.0  3 11 15 19  100.0  49  100.0  1.00 2.00 3.00 4.00 5.00  V a l i d Cases FACCOMP  49  M i s s i n g Cases  Importance of G e t t i n g a Sense of Accomplishment  Value Label  Value  1.00 3.00 4.00 5.00 TOTAL  V a l i d Cases  49  Frequency  Percent  Valid Percent  1  2.0 4.1 36.7 57.1  2.0 4.1 36.7 57.1  2 18 28 49  100.0  100.0  M i s s i n g Cases  0  Cum Percent  2. 6.1 42.9 100.0  182 Nl » Not Important; SI = Somewhat Important; MI * Moderately Important; QI = Q u i t e Important; VI => Very Important FPROMO  Imp. of Opportunity for Promotion  Value Label  Value  1.00 2.00 3.00 4.00 5.00 TOTAL Nl SI MI QI VI  Frequency  Percent  Valid Percent  2  4.1 8.2 10.2 38.8 38.8  4.1 8.2 10.2 38.8 38.8  4 5 19 19 49  100.0  1.00 2.00 3.00 4.00 5.00  Valid  Cases  49  M i s s i n g Cases  100.0  Cum Percent  4. 12.2 22.4 61.2 100.0  183 Nl =» Not Important; SI = Somewhat Important; MI = Moderately Important; QI = Quite Important; VI = Very Important FIMAGE  Importance of Agency Having P o s i t i v e  Value Label  Value 1.00 2.00 3.00 4.00 5.00 TOTAL  Nl SI MI QI VI  Frequency  Percent  Valid Percent  Cum Percent  5  10.2 14.3 30.6 30.6 14.3  10.2 14.3 30.6 30.6 14.3  10.2 24.5 55.1 85.7 100.0  7 15 15 7 49  100.0  100.0  1.00 2.00 3.00 4.00 5.00  V a l i d Cases FJOBSEC  49  M i s s i n g Cases  Importance of Having Good Job S e c u r i t y  Value Label  Value 2.00 3.00 4.00 5.00 TOTAL  Nl SI MI QI VI  Image  Frequency  Percent  Valid Percent  5  10.2 18.4 38.8 32.7  10.2 18.4 38.8 32.7  9 19 16 49  100.0  1.00 2.00 3.00 4.00 5.00  V a l i d Cases  49  M i s s i n g Cases  100.0  Cum Percent 10. 28.6 67.3 100.0  184 NI = Not Important; SI = Somewhat Important; MI = Moderately Important; QI = Q u i t e Important; VI = Very Important FCLEAR  importance of Having C l e a r  Value L a b e l  Value  2.00 3.00 4.00 5.00 TOTAL NI SI MI QI VI  Responsibilities  Frequency  Percent  Valid Percent  3  6.1 32.7 44.9 16.3  6.1 32.7 44.9 16.3  16 22 8 49  100.0  Cunt Percent  6. 38.8 83.7 100.0  100.0  1.00 2.00 3.00 4.00 5.00  Valid  Cases  FSPECIAL  49  Imp. of O p p o r t u n i t y f o r P r a c t i c e S p e c i a l  Value L a b e l  Value  1.00 2.00 3.00 4.00 5.00 TOTAL NI 1.00 SI 2.00 MI 3.00 QI 4.00 VI 5.00  Valid  M i s s i n g Cases  Cases  Frequency  Percent  Valid Percent  2  4.1 6.1 30.6 42.9 16.3  4.1 6.1 30.6 42.9 16.3  3 15 21 8 49  100.0  100.0  Cum Percent  4. 10.2 40.8 83.7 100.0  185 Nl = Not Important; SI = Somewhat Important; MI = Moderately Important; QI = Q u i t e Important; VI = Very Important FCONTACT  Importance Contact With Others Doing S i m i l a r Work  Value Label  Value  1.00 2.00 3.00 4.00 5.00 TOTAL Nl SI MI QI VI  Percent  Valid Percent  Cum Percent  1  2.0 4.1 22.4 57.1 14.3  2.0 4.1 22.4 57.1 14.3  2.0 6.1 28.6 85.7 100.0  2 11 28 7  100.0  49  100.0  1.00 2.00 3.00 4.00 5.00  Valid  Cases  FFRINGE  49  M i s s i n g Cases  Imp. of "Fringe B e n e f i t s "  Value Label  Value  2.00 3.00 4.00 5.00 TOTAL Nl SI MI QI VI  Frequency  Frequency  Percent  Valid Percent  1  2.0 16.3 42.9 38.8  2.0 16.3 42.9 38.8  8 21 19 49  100.0  1.00 2.00 3.00 4.00 5.00  Valid  Cases  49  M i s s i n g Cases  100.0  Cum Percent  2. 18.4 61.2 100.0  186 NI = Not Important; SI = Somewhat important; MI = Moderately Important; QI = Quite Important; VI = Very Important FWKSIZE  Importance of Having a Managable Workload  Value Label  Value  Frequency  Percent  Valid Percent  1  2.0 4.1 22.4 71.4  2.0 4.1 22.4 71.4  1.00 3.00 4.00 5.00  2 11 35  TOTAL  49  100.0  Cum Percent 2. 6.1 28.6 100.0  100.0  NI 1.00 SI 2.00 I MI 3.00 QI 4.00 VI 5.00  V a l i d Cases FPRODEV  49  M i s s i n g Cases  Imp. of Agency Support f o r P r o f .  Value Label  Value 3.00 4.00 5.00 TOTAL  49  Develop  Frequency  Percent  Valid Percent  1  2.0 34.7 63.3  2.0 34.7 63.3  17 31 49  100.0  NI 1.00 I 0 SI 2.00 I 0 MI 3.00 QI 4.00 VI 5.00  V a l i d Cases  0  M i s s i n g Cases  100.0  Cum Percent 2. 36.7 100.0  187 N l = Not Important; SI = Somewhat Important; MI = Moderately Important; QI = Quite Important; VI = Very Important FACCESS  Importance of Ease of Access to Workplace  Value Label  Value  1.00 2.00 3.00 4.00 5.00 TOTAL Nl SI MI QI VI  Percent  Valid Percent  1 9 15 15 9  2.0 18.4 30.6 30.6 18.4  2.0 18.4 30.6 30.6 18.4  49  100.0  Cum Percent  2. 20.4 51.0 81.6 100.0  100.0  1.00 2.00 3.00 4.00 5.00  V a l i d Cases FFUNDS  49  M i s s i n g Cases  Importance of Access to  Value Label  Value  2.00 3.00 4.00 5.00 TOTAL Nl SI MI QI VI  Frequency  Funds/Resources  Frequency  Percent  Valid Percent  2 4 14 29  4.1 8.2 28.6 59.2  4.1 8.2 28.6 59.2  49  100.0  1.00 2.00 3.00 4.00 5.00  Valid  Cases  49  M i s s i n g Cases  100.0  Cum Percent  4. 12.2 40.8 100.0  188  NI - Not Important; SI = Somewhat Important; MI = Moderately Important; QI = Quite Important; VI « Very Important FVAC  Importance of Liberal Vacation Time  Value Label  NI SI MI QI VI  Value 1.00 2.00 3.00 4.00 5.00  2 4 17 25  TOTAL  49  1  Percent 2.0 4.1 8.2 34.7 51.0 100.0  Valid Cum Percent Percent 2.0 4.1 8.2 34.7 51.0  2. 6.1 14.3 49.0 100.0  100.0  1.00 2.00 3.00 4.00 5.00  Valid Cases FSTATUS  49  Missing Cases  Imp. of Intra-Agency Job Status Value  NI SI MI QI VI  Frequency  Frequency  1.00 2.00 3.00 4.00 5.00  11 20 12 3  TOTAL  49  3  Percent 6.1 22.4 40.8 24.5 6.1 100.0  1.00 2.00 3.00 4.00 5.00  Valid Cases  49  Missing Cases  Valid Percent 6.1 22.4 40.8 24.5 6.1 100.0  Cum Percent 6.1 28.6 69.4 93.9 100.0  189  Nl =» Not Important; SI = Somewhat Important; MI =» Moderately Important; QI = Quite Important; VI = Very Important FCONSULT  Importance of Access to Consultation  Value Label  Value  Frequency  Percent 6.1 36.7 57.1  3.00 4.00 5.00  18 28  TOTAL  49  100.0  I  I.  I  3  Valid Cum Percent Percent 6.1 36.7 57.1  6. 42.9 100.0  100.0  Nl 1.00 I 0 SI 2.00 I 0^^  I I 0  6  Valid Cases FPROFESS  12  18  Missing Cases  0  24  I 30  Imp. of Agency Recognition of SW as Prof  Value Label  Nl SI MI QI VI  49  I  Value  Frequency  1.00 2.00 3.00 4.00 5.00  1 4 15 28  TOTAL  49  1  Percent 2.0 2.0 8.2 30.6 57.1 100.0  1.00 2.00 3.00 4.00 5.00  Valid Cases  49  Missing Cases  Valid Cum Percent Percent 2.0 2.0 8.2 30.6 57.1 100.0  2. 4.1 12.2 42.9 100.0  190  \  Nl <• Not Important; SI = Somewhat Important; MI =» Moderately Important; QI » Q u i t e Important; VI = Very Important FAUTO  Importance of Having D e c i s i o n Making Autonomy  Value L a b e l  Value  3.00 4.00 5.00 TOTAL Nl SI MI QI VI  Percent  Valid Percent  5  10.2 69.4 20.4  10.2 69.4 20.4  34 10  100.0  49  Cum Percent  10. 79.6 100.0  100.0  1.00 I 0 2.00 I 0 3.00 4.00 5.00  Valid  Cases  FCOWORK  49  M i s s i n g Cases  Importance of Having Competent Co-Workers  Value L a b e l  value  3.00 4.00 5.00 TOTAL Nl SI MI QI VI  Frequency  Frequency  Percent  Valid Percent  3  6.1 59.2 34.7  6.1 59.2 34.7  29 17 49  100.0  1.00 I 0 2.00 I 0 3.00 4.00 5.00  Valid  Cases  49  M i s s i n g Cases  100.0  Cum Percent  6. 65.3 100.0  191 NI = Not Important; SI = Somewhat Important; MI = Moderately Important; QI = Quite Important; VI = Very Important FINFLU  Importance of I n f l u e n c i n g Agency D e c i s i o n s  Value Label  Value 2.00 3.00 4.00 5.00 TOTAL  NI SI MI QI VI  Frequency  Percent  Valid Percent  3  6.1 20.4 42.9 30.6  6.1 20.4 42.9 30.6  10 21 15  100.0  49  Cum Percent 6. 26.5 69.4 100.0  100.0  1.00 2.00 3.00 4.00 5.00  Valid  Cases  FSUPRESP  49  M i s s i n g Cases  Importance of S u p e r v i s o r Respect f o r Employees  Value Label  Value 3.00 4.00 5.00 TOTAL  Frequency  Percent  Valid Percent  1  2.0 14.3 83.7  2.0 14.3 83.7  7 41 49  100.0  NI 1.00 SI 2.00 I MI 3.00 QI 4.00 VI 5.00  Valid  Cases  49  M i s s i n g Cases  100.0  Cum Percent 2. 16.3 100.0  192 NI • Not Important; SI = Somewhat Important; MI - Moderately Important; QI - Quite Important; VI - Very Important FAGERESP . Importance of Agency Respect Value L a b e l  Value  3.00 4.00 5.00 TOTAL  Cases  49  Employees  Frequency  Percent  Valid Percent  4  8.2 16.3 75.5  8.2 16.3 75.5  8 37 49  100.0  NI 1.00 I 0 SI 2.00 I 0 MI 3.00 QI 4.00 VI 5.00  Valid  for  M i s s i n g Cases  100.0  Cum Percent  8. 24.5 100.0  APPENDIX H DESCRIPTIVE LISTING OF S (SATISFACTION) VARIABLES FOR SOCIAL WORKERS Variable  SOVER  5.745 - .820 - .233 9.000 10.00  Mean Kurtosis Skewness Range Maximum Valid  Satisfaction  Observations  Variable  SPAY  Mean Kurtosis Skewness Range Maximum  Mean Kurtosis Skewness Range Maximum Valid  Std Dev S . E . Kurt S . E . Skew Minimum Missing  Satisfaction 4.276 -1.196 .249 8.000 9.00  V a l i d Observations  Variable  49  SSUPER  - 49  Overall  2.287 .668 .340 1.00  Observations  With Pay Std Dev S . E . Kurt S . E . Skew Minimum  M i s s i n g Observations  2.486 .668 .340 1.00 - 0  S a t . With Q u a l i t y of S u p e r v i s i o n  6.867 .071 1.014 9.000 10.00  Observations  49  Std Dev S . E . Kurt S . E . Skew Minimum Missing  Observations  2.671 .668 .340 1.00  194 D e s c r i p t i v e L i s t i n g Of S V a r i a b l e s For S o c i a l Workers Variable  SPAPER  3.990 .608 .644 9.000 10.00  Mean Kurtosis Skewness Range Maximum Valid  Observations  Variable  SMORALE  SPHYSIC  Valid  Mean Kurtosis Skewness Range Maximum  Missing  2.053 .668 .340 1.00  Observations  Std Dev S . E . Kurt S . E . Skew Minimum M i s s i n g Observations  2.180 .668 .340 1.00 - 0  S a t . With P h y s i c a l Work Environment  Observations  Variable Week  49  4.724 - .770 - .017 8.000 9.00  Mean Kurtosis Skewness Range Maximum  Std Dev S . E . Kurt S . E . Skew Minimum  S a t . With S t a f f Morale  Observations  Variable  49  5.153 - .871 .113 8.000 9.00  Mean Kurtosis Skewness Range Maximum Valid  S a t . With Amount of Paperwork  SHOURS  49  Std Dev S . E . Kurt S . E . Skew Minimum Missing  2.294 .668 .340 1.00  Observations  Sat With # of Hours Being Worked per  4.500 - .597 .395 8.000 9.00  V a l i d Observations  - 49  Std Dev S . E . Kurt S . E . Skew Minimum Missing  Observations  2.155 .668 .340 1.00 0  195 D e s c r i p t i v e L i s t i n g Of S V a r i a b l e s For S o c i a l Workers Variable  SACCOMP  Mean Kurtosis Skewness Range Maximum  S a t . With Sense of Accomplishment  Std Dev  5.255 -.586 -.262 8.000 9.00  V a l i d Observations  S . E . Kurt S . E . Skew Minimum  - 49  M i s s i n g Observations  Number of V a l i d Observations Variable  SPROMO  Mean Kurtosis Skewness Range Maximum  4.602 -.794 -.150 7.000 8.00  SIMAGE  Valid  Mean Kurtosis Skewness Range Maximum Valid  Sat.  Observations  Variable  - 49  3.541 - .914 .021 6.000 7.00  Mean Kurtosis Skewness Range Maximum  =  .668 .340 1.00  - 0  47.00  S a t . With O p p o r t u n i t y f o r Promotion  V a l i d Observations  Variable  (Listwise)  2.067  49  Sat.  SJOBSEC  6.071 - .181 - .597 9.000 10.00  Observations  49  S t d Dev S . E . Kurt S . E . Skew Minimum M i s s i n g Observations  1.759 .668 .340 1.00 - 0  With P u b l i c Image of Agency Std Dev S . E . Kurt S . E . Skew Minimum  1.744 .668 .340 1.00  M i s s i n g Observations  With Job S e c u r i t y Std Dev S . E . Kurt S . E . Skew Minimum Missing  Observations  2.198 .668 .340 1.00  196 D e s c r i p t i v e L i s t i n g Of S V a r i a b l e s For S o c i a l Workers Variable SRESP Responsibilities Mean Kurtosis Skewness Range Maximum Valid  S a t . With C l a r i t y of Job 5.745 -.052 -.172 8.000 10.00  Observations  Variable SSPECIAL Specialization Mean Kurtosis Skewness Range Maximum Valid  Variable SCONTACT Other P r o f e s s i o n a l s  49  Observations  Std Dev S . E . Kurt S . E . Skew Minimum Missing  1.534 .668 .340 1.00  Observations  Sat. With O p p o r t u n i t y f o r Contact With  5.296 - .364 - .355 7.000 8.00  V a l i d Observations  Missing  1.627 .668 .340 2.00  Sat. With O p p o r t u n i t y f o r P r a c t i c e  5.071 .275 -.704 6.000 7.00  Observations  Mean Kurtosis Skewness Range Maximum  49  Std Dev S . E . Kurt S . E . Skew Minimum  - 49  Std Dev S . E . Kurt S . E . Skew Minimum Missing  Observations  1.695 .668 .340 1.00 0  197 D e s c r i p t i v e L i s t i n g Of S V a r i a b l e s For S o c i a l Workers Variable  S a t . With F r i n g e B e n e f i t s  SFRINGE  Mean Kurtosis Skewness Range Maximum  6.092 -.331 -.733 9.000 10.00  V a l i d Observations  Variable  SWKSIZE  - 49  Missing Observations  2.393 .668 .340 1.00 - 0  S a t . With Workload S i z e  3.684 .056 .841 9.000 10.00  Mean Kurtosis Skewness Range Maximum  Std Dev S . E . Kurt S . E . Skew Minimum  V a l i d Observations  - 49  Std Dev S . E . Kurt S . E . Skew Minimum Missing Observations  2.302 .668 .340 1.00 - 0  Variable SPRODEV Sat, With Agency Support f o r P r o f e s s i o n a l Development  3.439 .263 .815 9.000 10.00  Mean Kurtosis Skewness Range Maximum Valid  Observations  Variable Mean Kurtosis Skewness Range Maximum  49  Std Dev S . E . Kurt S . E . Skew Minimum Missing  2.216  .668 .340 1.00  Observations  S a t . With Ease of Access to Workplace  SACCESS  6.255 -.414 -.525 9.000 10.00  V a l i d Observations  - 49  Std Dev S . E . Kurt S . E . Skew Minimum Missing  Observations  2.310  .668 .340 1.00  0  198 D e s c r i p t i v e L i s t i n g Of S V a r i a b l e s For S o c i a l Workers Variable  S a t . With Funds/Resources  SFUNDS  Mean Kurtosis Skewness Range Maximum  3.802 -.508 .312 7.000 8.00  V a l i d Observations  Variable  4.418 1.163 .240 8.000 9.00  Observations  Variable Mean Kurtosis Skewness Range Maximum Valid  Mean Kurtosis Skewness Range Maximum  49  1.830 .674 .343 1.00  Observations  Time  Std Dev S . E . Kurt S . E . Skew Minimum M i s s i n g Observations  2.392 .668 .340 1.00 - 0  S a t . With Intra-Agency Job Status  SSTATUS  5.031 -.641 -.341 7.000 8.00  Observations  Variable  Missing  S a t . With V a c a t i o n  SVAC  Mean Kurtosis Skewness Range Maximum Valid  - 48  Std Dev S . E . Kurt S . E . Skew Minimum  SCONSULT  49  Std Dev S . E . Kurt S . E . Skew Minimum M i s s i n g Observations  1.975 .668 .340 1.00 - 0  S a t . With Access to C o n s u l t a t i o n  6.235 - .657 - .462 9.000 10.00  Std Dev S . E . Kurt S . E . Skew Minimum  2.531 .668 .340 1.00  199 D e s c r i p t i v e L i s t i n g Of S V a r i a b l e s For S o c i a l Variable SPROFESS Profession Mean Kurtosis Skewness Range Maximum  Variable  SAUTO  SCOWORK  Missing  Observations  2.286 .668 .340 1.00 - 0  1.890 .668 .340 1.00  Std Dev S . E . Kurt S . E . Skew Minimum Missing  Observations  S a t . With Co-Workers Competency  V a l i d Observations  Mean Kurtosis Skewness Range Maximum  - 49  7.073 - .000 - .841 6.500 9.50  Mean Kurtosis Skewness Range Maximum  S t d Dev S . E . Kurt S . E . Skew Minimum  S a t . With D e c i s i o n Making Autonomy  V a l i d Observations  Variable Decisions  - 49  5.867 .046 - .283 9.000 10.00  Mean Kurtosis Skewness Range Maximum  Variable  S a t . With Agency R e c o g n i t i o n of SW as a  4.541 -.781 .219 8.000 9.00  V a l i d Observations  Workers  SINFLU  - 48  Missing  Observations  S a t . With A b i l i t y  4.908 .611 .431 9.000 10.00  V a l i d Observations  - 49  1.618 .674 .343 3.00  Std Dev S . E . Kurt S . E . Skew Minimum  to Influence  Std Dev S . E . Kurt S . E . Skew Minimum Missing  Observations  Agency  1.842 .668 .340 1.00 - 0  200 D e s c r i p t i v e L i s t i n g Of S V a r i a b l e s For S o c i a l Workers Variable SSUPRES Employees Mean Kurtosis Skewness Range Maximum Valid  7.745 1.055 -1.410 9.000 10.00  Observations  Variable Mean Kurtosis Skewness Range Maximum Valid  Sat.  SAGERESP  49  49  2.462 .668 .340 1.00  Std Dev S . E . Kurt S . E . Skew Minimum Missing  observations  S a t . With Agency Respect  3.520 .776 .942 8.000 9.00  Observations  With Supervisor Respect f o r  f o r Employees  Std Dev S . E . Kurt S . E . Skew Minimum Missing Observations  1.893 .668 .340 1.00  201 APPENDIX I GRAPHS OF S (SATISFACTION) VARIABLES RECODED TO FIVE LEVELS VD = Very D i s s a t i s f i e d ; SD = Somewhat D i s s a t i s f i e d ; NN = Neither Satisfied Nor D i s s a t i s f i e d ; SS = Somewhat S a t i s f i e d ; VS = Very S a t i s f i e d SOVER  Satisfaction Overall  Value Label  Value  Frequency  Percent  Valid Percent  4  8.2 28.6 10.2 42.9 10.2  8.2 28.6 10.2 42.9 10.2  1.00 2.00 3.00 4.00 5.00  14 5 21 5  TOTAL  49  100.0  100.0  VD 1.00  V a l i d Cases  49  Missing Cases  0  Cum Percent 8. 36.7 46.9 89.8 100.0  202 Graphs Of S Variables Recoded to Five Levels SPAY  Satisfaction With Pay  Value Label  Value  Frequency  Percent  Valid Percent  1.00 2.00 3.00 4.00 5.00  14 15 6 13 1  28.6 30.6 12.2 26.5 2.0  28.6 30.6 12.2 26.5 2.0  TOTAL  49  100.0  Cum Percent 28. 59.2 71.4 98.0 100.0  100.0  VD 1.00 SD 2.00 NN 3.00 SS 4.00 VS 5.00  V a l i d Cases SSUPER  49  Missing Cases  Sat. With Quality of Supervision  Value Label  Value  Frequency  Percent  Valid Percent  6  12.2 8.2 12.2 34.7 32.7  12.2 8.2 12.2 34.7 32.7  1.00 2.00 3.00 4.00 5.00  4 6 17 16  TOTAL  49  100.0  VD 1.00 SD 2.00 NN 3.00 SS 4.00 VS 5.00  V a l i d Cases  49  Missing Cases  100.0  Cum Percent 12. 20.4 32.7 67.3 100.0  203 Graphs Of S Variables Recoded to Five Levels SPAPER  Sat. With Amount of Paperwork  Value Label  Value 1.00 2.00 3.00 4.00 5.00 TOTAL  Frequency 11 20 13 3 2 49  Percent 22.4 40.8 26.5 6.1 4.1 100.0  Valid Percent 22.4 40.8 26.5 6.1 4.1  Cum Percent 22. 63.3 89.8 95.9 100.0  100.0  VD 1.00 SD 2.00 NN 3.00 SS 4.00 VS 5.00  Valid Cases SMORALE  49  Missing Cases  Sat. With Staff Morale  Value Label  Value  Frequency  1.00 2.00 3.00 4.00 5.00  14 14 10 4  TOTAL  49  7  Percent 14.3 28.6 28.6 20.4 8.2 100.0  VD 1.00 SD 2.00 NN 3.00 SS 4.00 VS 5.00  Valid Cases  49  Missing Cases  Valid Percent 14.3 28.6 28.6 20.4 8.2 100.0  Cum Percent 14. 42.9 71.4 91.8 100.0  204 Graphs Of S Variables Recoded to Five Levels SPHYSIC  Sat. With Physical Work Environment  Value Label  Value  Frequency  Percent  Valid Percent  1.00 2.00 3.00 4.00 5.00  10 13 14 9 3  20.4 26.5 28.6 18.4 6.1  20.4 26.5 28.6 18.4 6.1  TOTAL VD SD NN SS VS  49  100.0  Cum Percent 20. 46.9 75.5 93.9 100.0  100.0  1.00 2.00 3.00 4.00 5.00  V a l i d Cases SHOURS  49  Missing Cases  Sat With # of Hours Being Worked per Week  Value Label  VD 1.00 SD 2.00 NN 3.00 SS 4.00 VS 5.00  Value  Frequency  Percent  Valid Percent  7  14.3 46.9 18.4 16.3 4.1  14.3 46.9 18.4 16.3 4.1  1.00 2.00 3.00 4.00 5.00  23 9 8 2  TOTAL  49  It?  I I. 0 V a l i d Cases  ill  1I ,1.  5  49  p.,  rnrnn  100.0  Cum Percent 14. 61.2 79.6 95.9 100.0  100.0  . nnnnn.nnnnni.nnnnnnnnnnnrnnnnnrrnnnnnnnn 1  rnnpHmnnnnnnnnnnnnnnnnnnnnrnnnnn  9  23  8 .1  10  . I, 15  Missing Cases  .1,  20 0  .I 25  205 Graphs Of S Variables Recocted to Five Levels SACCOMP  Sat. With Sense of Accomplishment '  Value Label  Value  Frequency  Percent  Valid Percent  1.00 2.00 3.00 4.00 5.00  5 12 15 14 3  10.2 24.5 30.6 28.6 6.1  10.2 24.5 30.6 28.6 6.1  TOTAL  49  100.0  Cum Percent 10. 34.7 65.3 93.9 100.0  100.0  VD 1.00 SD 2.00 NN 3.00 SS 4.00 VS 5.00  V a l i d Cases SPROMO  49  Missing Cases  Sat. With Opportunity for Promotion  Value Label  Value  Frequency  Percent  Valid Percent  1.00 2.00 3.00 4.00  6 18 17 8  12.2 36.7 34.7 16.3  12.2 36.7 34.7 16.3  TOTAL  49  100.0  VD 1.00 SD 2.00 NN 3.00 SS 4.00 j VS 5.00  V a l i d Cases  49  Missing Cases  100.0  Cum Percent 12. 49.0 83.7 100.0  206 Graphs Of S Variables Recocted to Five Levels SIMAGE  Sat. With Public Image of Agency  Value Label  Value  Frequency  Percent  Valid Percent  1.00 2.00 3.00 4.00  15 19 13 2  30.6 38.8 26.5 4.1  30.6 38.8 26.5 4.1  TOTAL  49  100.0  Cum Percent 30. 69.4 95.9 100.0  100.0  VD 1.00 SD 2.00 NN 3.00 SS 4.00 VS 5.00  V a l i d Cases SJOBSEC  49  Missing Cases  Sat. With Job Security  Value Label  Value  Frequency  Percent  Valid Percent  4  8.2 10.2 28.6 46.9 6.1  8.2 10.2 28.6 46.9 6.1  1.00 2.00 3.00 4.00 5.00  5 14 23 3  TOTAL  49  100.0  VD 1.00 SD 2.00 NN 3.00 SS 4.00 VS 5.00  V a l i d Cases  49  Missing Cases  100.0  Cum Percent 8. 18.4 46.9 93.9 100.0  207 Graphs Of S Variables Recoded to Five Levels SRESP  Sat. With C l a r i t y of Job R e s p o n s i b i l i t i e s  Value Label  Value  Frequency  Percent  Valid Percent  1  2.0 22.4 36.7 36.7 2.0  2.0 22.4 36.7 36.7 2.0  1.00 2.00 3.00 4.00 5.00  11 18 18 1  TOTAL  49  100.0  Cum Percent 2. 24.5 61.2 98.0 100.0  100.0  VD 1.00 SD 2.00 NN 3.00 SS 4.00 VS 5.00  V a l i d Cases SSPECIAL  49  Nissing Cases  Sat. With Opportunity for Practice Specialization  Value Label  Value  Frequency  Percent  Valid Percent  2  4.1 26.5 49.0 20.4  4.1 26.5 49.0 20.4  1.00 2.00 3.00 4.00  13 24 10  TOTAL  49  100.0  100.0  VD 1.00 SD 2.00 NN 3.00 SS 4.00 VS 5.00  V a l i d Cases  49  Missing Cases  0  Cum Percent 4. 30.6 79.6 100.0  208 Graphs Of S Variables Recoded to Five Levels SCONTACT Sat. With Opportunity for Contact With Other Professionals Value Label  Value  Frequency  Percent  Valid Percent  2  4.1 26.5 46.9 22.4  4.1 26.5 46.9 22.4  1.00 2.00 3.00 4.00  13 23 11  TOTAL  49  100.0  Cum Percent 4. 30.6 77.6 100.0  100.0  VD 1.00 SD 2.00 NN 3.00 SS 4.00 VS 5.00  V a l i d Cases SFRINGE  49  Missing Cases  Sat. With "Fringe Benefits"  Value Label  Value  Frequency  1.00 2.00 3.00 4.00 5.00  7 10 22 5  TOTAL  49  Percent  Valid Percent  Cum Percent  10.2 14.3 20.4 44.9 10.2  10.2 14.3 20.4 44.9 10.2  10.2 24.5 44.9 89.8 100.0  100.0  VD 1.00 SD 2.00 NN 3.00 SS 4.00 VS 5.00  V a l i d Cases  49  Missing Cases  100.0  209 Graphs Of S Variables Recoded to Five Levels SWKSIZE  Sat. With Workload Size  Value Label  Value 1.00 2.00 3.00 4.00 5.00 TOTAL  Frequency  Percent  Valid Percent  Cum Percent  19 15 8 5 2  38.8 30.6 16.3 10.2 4.1  38.8 30.6 16.3 10.2 4.1  38.8 69.4 85.7 95.9 100.0  49  100.0  100.0  VD 1.00 SD 2.00 NN 3.00 SS 4.00 VS 5.00  V a l i d Cases SPRODEV  49  Missing Cases  Sat. With Agency Support for Prof. Devel  Value Label  Value  Frequency  Percent  Valid Percent  1.00 2.00 3.00 4.00 5.00  19 16 8 5 1  38.8 32.7 16.3 10.2 2.0  38.8 32.7 16.3 10.2 2.0  TOTAL  49  100.0  VD 1.00 SD 2.00 NN 3.00 SS 4.00 VS 5.00  V a l i d Cases  49  Missing Cases  100.0  Cum Percent 38. 71.4 87.8 98.0 100.0  210 Graphs Of S Variables Recoded to Five Levels SACCESS  Sat. with Ease of Access to Workplace  Value Label  Value  Frequency  Percent  Valid Percent  4  8.2 8.2 34.7 28.6 20.4  8.2 8.2 34.7 28.6 20.4  1.00 2.00 3.00 4.00 5.00  4 17 14 10  TOTAL  49  100.0  Cum Percent 8. 16.3 51.0 79.6 100.0  100.0  VD 1.00 SD 2.00 NN 3.00 SS 4.00 VS 5.00 8 V a l i d Cases SFUNDS  49  12  I, 16  I 20  Missing Cases  Sat. With Funds/Resources  Value Label  Value  Frequency  Percent  Valid Percent  Cum Percent  1.00 2.00 3.00 4.00 -9.00  11 21 11 5 1  22.4 42.9 22.4 10.2 2.0  22.9 43.8 22.9 10.4 MISSING  22. 66.7 89.6 100.0  TOTAL  49  100.0  VD 1.00 SD 2.00 NN 3.00 SS 4.00 VS 5.00  V a l i d Cases  48  Missing Cases  100.0  211 Graphs Of S Variables Recoded to Five Levels SVAC  Sat. With Vacation Time  Value Label  Value  Frequency  Percent  Valid Percent  1.00 2.00 3.00 4.00 5.00  13 14 8 12 2  26.5 28.6 16.3 24.5 4.1  26.5 28.6 16.3 24.5 4.1  TOTAL  49  100.0  Cum Percent 26. 55.1 71.4 95.9 100.0  100.0  VD 1.00 SD 2.00 NN 3.00 SS 4.00 VS 5.00  V a l i d Cases SSTATUS  49  Missing Cases  Sat. With Intra-Agency Job Status  Value Label  Value  Frequency  Percent  Valid Percent  7  14.3 20.4 42.9 22.4  14.3 20.4 42.9 22.4  1.00 2.00 3.00 4.00  10 21 11  TOTAL  49  100.0  VD 1.00 SD 2.00 NN 3.00 SS 4.00 VS 5.00  V a l i d Cases  49  Missing Cases  100.0  Cum Percent 14. 34.7 77.6 100.0  212 Graphs Of S Variables Recoded to Five Levels SCONSULT  Sat. w i t h Access to Consultation  Value Label  Value  Frequency  1.00 2.00 3.00 4.00 5.00  9 8 15 12  TOTAL  49  5  Percent  Valid Percent  10.2 18.4 16.3 30.6 24.5  10.2 18.4 16.3 30.6 24.5  I  100.0  Cum Percent 10. 28.6 44.9 75.5 100.0  100.0  VD 1.00 SD 2.00 NN 3.00 SS 4.00 VS 5.00  V a l i d Cases SPROFESS  49  Missing Cases  Sat. With Agency Recognition of SW as a Profession  Value Label  Value  Frequency  Percent  Valid Percent  1.00 2.00 3.00 4.00 5.00  12 12 16 6 3  24.5 24.5 32.7 12.2 6.1  24.5 24.5 32.7 12.2 6.1  TOTAL  49  100.0  VD 1.00 SD 2.00 | NN 3.00 i SS 4.00 VS 5.00  V a l i d Cases  49  Missing Cases  100.0  Cum Percent 24. 49.0 81.6 93.9 100.0  213 Graphs Of S Variables Recoded to Five Levels SAUTO  Sat. With Decision Making Autonomy  Value Label  Value  Frequency  Percent  Valid Percent  1.00 2.00 3.00 4.00 5.00  3 9 16 18 3  6.1 18.4 32.7 36.7 6.1  6.1 18.4 32.7 36.7 6.1  TOTAL  49  100.0  Cum Percent 6. 24.5 57.1 93.9 100.0  100.0  VD 1.00 SD 2.00 NN 3.00 SS 4.00 VS 5.00  V a l i d Cases SCOWORK  49  Missing Cases  Sat. With Co-workers' Competency  Value Label  VD 1.00 SD 2.00 NN 3.00 SS 4.00 VS 5.00  V a l i d Cases  Value  Frequency  Percent  Valid Percent  6  12.2 14.3 55.1 16.3 2.0  12.5 14.6 56.3 16.7 MISSING  2.00 3.00 4.00 5.00 -9.00  7 27 8 1  TOTAL  49  100.0  100.0  Cum Percent 12. 27.1 83.3 100.0  214 Graphs Of S Variables Recoded to Five Levels SINFLU  Sat. With A b i l i t y to Influence Agency Decisions  Value Label  Value  Frequency  Percent  Valid Percent  3  6.1 42.9 32.7 14.3 4.1  6.1 42.9 32.7 14.3 4.1  1.00 2.00 3.00 4.00 5.00  21 16 7 2  TOTAL  49  100.0  Cum Percent 6. 49.0 81.6 95.9 100.0  100.0  VD 1.00 SD 2.00 NN 3.00 SS 4.00 VS 5.00  V a l i d Cases SSUPRES  49  Missing Cases  Sat. With Supervisor Respect for Employees  Value Label  Value  Frequency  Percent  Valid Percent  4  8.2 6.1 6.1 24.5 55.1  8.2 6.1 6.1 24.5 55.1  1.00 2.00 3.00 4.00 5.00  3 3 12 27  TOTAL  49  100.0  VD 1.00 SD 2.00 NN 3.00 SS 4.00 VS 5.00  V a l i d Cases  49  Missing Cases  100.0  Cum Percent 8. 14.3 20.4 44.9 100.0  215 SAGERESP  Sat. With Agency Respect for Employees  Value Label  Value 1.00 2.00 3.00 4.00 5.00 TOTAL  Frequency  Percent  Valid Percent  17 20 9 2 1  34.7 40.8 18.4 4.1 2.0  34.7 40.8 18.4 4.1 2.0  49  100.0  VD 1.00 SD 2.00 NN 3.00 SS 4.00 VS 5.00  V a l i d Cases  Missing Cases  100.0  Cum Percent 34. 75.5 93.9 98.0 100.0  216 APPENDIX J D (DISCREPANY) VARIABLES LISTED IN ORDER OF MEANS Variable  Mean  Std Dev  DAGERESP DPRODEV DWKSIZE DFUNDS DPROFESS DVAC DPAY DMORALE DACCOMP DHOURS DPROMO DINFLU DIMAGE DPHYSIC DCONSULT DAUTO DCONTACT DPAPER DFRINGE DSPECIAL DSUPRES DJOBSEC DSUPER DRESP DCOWORK DSTATUS DACCESS  5..30 5..26 4. .97 4. .53 3..77 3..72 3..65 3..32 3..26 2..93 2. .87 2..54 2..50 2..34 2..34 1, .91 1, .79 1. .79 1. .85 1. .62 1. .38 1. .32 1. .28 1. .21 1. .02 .46 .11  2,.37 2..57 3..52 2..82 3..23 3..72 3..25 2..46 2..58 3..21 3..11 2..65 3..43 3..27 2..62 2..04 2..48 4, .05 2..86 2..92 2..73 2..87 3..42 2..56 2..05 2..65 3..14  *  Minimum .00 .00 -9, .00 -2, .00 -6. .00 -5, .50 -4. .00 -3, .00 -4. .00 -4. .00 -4. .00 -3, .00 -4, .00 -5, .50 -2, .00 -1, .00 -4, .00 -9. .00 -3. .50 -4. .00 -3. .00 -5. .00 -7. .50 -5. .00 -3. .00 -5. .50 -7. .00  Maximum 9. .00 9. .00 9. ,00 9. .00 9. .00 9. .00 9. .00 8. .00 9. .00 9. .00 8. .00 8. .00 8. .00 9. .00 9. .00 7. .00 7. .00 9. .00 9. .00 8. .00 9. ,00 8. ,00 9. ,00 8. ,00 6. ,00 6. ,00 9. ,00  N 49 49 49 48 49 49 49 49 49 49 49 49 49 49 49 49 49 49 49 49 49 49 49 49 48 49 49  217 APPENDIX K CORRELATION MATRIX OF SIGNIFICANT PREDICTORS OF SOVER >rrelations: SACCOMP SWKSIZE SCONSULT SPAPER SVAC SSTATUS SINFLU SAUTO SPAY SPROFESS SSUPRESP SCLEAR SPROMO SFUNDS SHOURS )rrelations: SACCOMP SWKSIZE SCONSULT SPAPER SVAC SSTATUS SINFLU SAUTO SPAY SPROFESS SSUPRESP SCLEAR SPROMO SFUNDS SHOURS  SACCOMP  SWKSIZE  SCONSULT  SPAPER  1.0000 .4053* .1464 .2980 .3497* .1006 .0937 .1910 .0328 .0721 -.0310 .3437* .0653 .3693* .2147  .4053* 1.0000 .3452* .6586** .3439* .2253 .3243 .2378 .2053 .2778 .0242 .3640* .0718 .3098 .4118*  .1464 .3452* 1.0000 .3202 .5273** .5644** .3317 .3759* .2074 .4889** .6156** .2038 .2242 .2199 .2351  .2980 .6586** .3202 1.0000 .2174 .2330 .3000 .3244 .4575** .1002 .0821 .3372* .2443 .1290 .4125*  SVAC  SSTATUS  SINFLU  SAUTO  .0937 .3243 .3317 .3000 .2490 .3058 1.0000 .6161** .2322 .2254 .3908* .0430 .0735 .1912 .1465  .1910 .2378 .3759* .3244 .3418* .6326** .6161** 1.0000 .3441* .3661* .3385* .1735 .3456* .2370 .2455  .3497* .3439* .5273** .2174 1.0000 .3668* .2490 .3418* .2628 .2712 .3332 .1193 .4033* .3211 .0963  .1006 .2253 .5644** .2330 .3668* 1.0000 .3058 .6326** .3645* .6502** .4271* .2583 .4677** .1864 .1831  218 Correlation Matrix of S i g n i f i c a n t Predictors of SOVER Correlations :  SPAY  SACCOMP SWKSIZE SCONSULT SPAPER SVAC SSTATUS SINFLU SAUTO SPAY SPROFESS SSUPRESP SCLEAR SPROMO SFUNDS SHOURS  .0328 .2053 .2074 .4575** .2628 .3645* .2322 .3441* 1.0000 .3137 .1619 .1634 .5137** .1099 .3041  .0721 .2778 .4889** .1002 .2712 .6502** .2254 .3661* .3137 1.0000 .4216* .3990* .1618 .0088 .1992  -.0310 .0242 .6156** .0821 .3332 .4271* .3908* .3385* .1619 .4216* 1.0000 .1512 .3176 .1719 - .0057  Correlations :  SPROMO  SFUNDS  SHOURS  SACCOMP SWKSIZE SCONSULT SPAPER SVAC SSTATUS SINFLU SAUTO SPAY SPROFESS SSUPRESP SCLEAR SPROMO SFUNDS SHOURS  .0653 .0718 .2242 .2443 .4033* .4677** .0735 .3456* .5137** .1618 .3176 .0533 1.0000 .2167 .0943  .3693* .3098 .2199 .1290 .3211 .1864 .1912 .2370 .1099 .0088 .1719 .3132 .2167 1.0000 .1281  .2147 .4118* .2351 .4125* .0963 .1831 .1465 .2455 .3041 .1992 -.0057 .3212 .0943 .1281 1.0000  SPROFESS  SSUPRESP  SCLEAR .3437* .3640* .2038 .3372* .1193 .2583 .0430 .1735 .1634 .3990* .1512 1.0000 .0533 .3132 .3212  APPENDIX L T-TESTS OF F (IMPORTANCE) VARIABLES WITH TOTGOV AS GROUPING VARIABLE t-test for:  FSUPER  Group 1 Group 2  Imp. of Quality  Number of Cases  Mean  26 23  8.6731 7.5435  Pooled Variance F 2-Tail Value Prob. 1.68  2.21  FMORALE  Group 1 Group 2  Standard Deviation 1.555 2.016  .305 .420  Estimate  47  .032  Imp. of High Staff Morale  Number of Cases  Mean  26 23  8.9808 7.8913  Standard Deviation 1.081 1.507  Pooled Variance Estimate F 2-Tail Value Prob.  Standard Error  t Degrees of 2 - T a i l Value Freedom Prob.  .211  t-test for:  Supervision  ii t Degrees of 2 - T a i l § 1 Value Freedom Prob.  Standard Error .212 .314  220 T-Tests of F Variables with TOTGOV as Grouping Variable t-test for: Environment  FPHYSIC  Imp. of Comfortable Physical Work  Number of Cases  Mean  26 23  7.7500 6.2826  Group 1 Group 2  Pooled Variance F 2-Tail Value Prob. 1.92  1.423 1.970  3.01  FPROMO  .000  Mean  26 23  8.2115 6.6304  Ii t i value I  .004  Imp. of Opportunity for Promotion  Separate Variance  6.18  .279 .411  Estimate  47  Number of Cases Group 1 Group 2  F 2-Tail Value Prob.  Standard Error  t Degrees of 2 - T a i l Value Freedom Prob.  .118  t-test for:  Standard Deviation  2.65  Degrees of Freedom 28.24  Standard Deviation 1.079 2.681 Estimate 2-Tail Prob. .013  Standard Error .212 .559  221 T-Tests of F Variables with TOTGOV as Grouping Variable t-test for: Image  Imp. of Agency Having Positive Public  FIMAGE  Group 1 Group 2  Number of Cases  Mean  26 23  6.8654 5.1087  Pooled Variance F 2-Tail Value Prob. 1.04  .926  t-test for:  2.133 2.089  4.25  .001  .418 .436  Estimate  2.90  47  .006  Imp. of Having Managable Workload  Number of Cases  Mean  26 23  9.0962 8.1522  Standard Deviation .980 2.019  Separate Variance Estimate F 2-Tail Value Prob.  Standard Error  t Degrees of 2 - T a i l Value Freedom Prob.  FWKSIZE  Group 1 Group 2  Standard Deviation  t Value 2.04  Degrees of Freedom 30.94  2-Tail Prob. .050  Standard Error .192 .421  222 T-Tests of F Variables with TOTGOV as Grouping Variable t-test for: Professional  FPRODEV Imp. of Agency Support for Development  Group 1 Grouo 2  Number of Cases  Mean  26 23  8.9808 8.3696  Standard Deviation .922 1.130  Standard Error .181 .236  Pooled Variance Estimate F 2-Tail Value Prob. 1.50  324  t-test for: Profession  t Degrees of 2 - T a i l Value Freedom Prob. 2.08  FPROFESS  Group 1 Group 2  47  Imp. of Agency Recognition of SW as a  Number of Cases  Mean  26 23  9.0192 7.5000  Separate Variance F 2-Tail Value Prob. 4.48  .000  .043  t Value 3.07  Degrees of Freedom 30.50  Standard Deviation 1.025 2.169 Estimate 2-Tail Prob. .004  Standard Error .201 .452  223 T-Tests of F Variables with TOTGOV as Grouping Variable t-test for:  FCOWORK  Group 1 Group 2  Imp. of Competent Co-Workers  Number of Cases  Mean  26 23  8.4808 7.6304  Standard Deviation 1.081 1.150  Pooled variance Estimate F 2-Tail Value Prob. 1.13  .761  t Degrees of 2 - T a i l Value Freedom Prob. 2.67  47  .010  Standard Error .212 .240  

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.831.1-0098003/manifest

Comment

Related Items