UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The role enactment questionnaire : reliability and validity testing Sleigh, Kenna Marie 1995

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

Item Metadata

Download

Media
831-ubc_1995-0407.pdf [ 5.67MB ]
Metadata
JSON: 831-1.0086996.json
JSON-LD: 831-1.0086996-ld.json
RDF/XML (Pretty): 831-1.0086996-rdf.xml
RDF/JSON: 831-1.0086996-rdf.json
Turtle: 831-1.0086996-turtle.txt
N-Triples: 831-1.0086996-rdf-ntriples.txt
Original Record: 831-1.0086996-source.json
Full Text
831-1.0086996-fulltext.txt
Citation
831-1.0086996.ris

Full Text

/fi-^C/f2-THE ROLE ENACTMENT QUESTIONNAIRE: RELIABILITY AND VALIDITY TESTING by KENNA MARIE SLEIGH B.S.N., The University of B r i t i s h Columbia, 1990 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE IN NURSING in THE FACULTY OF GRADUATE STUDIES School of Nursing We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August 199 5 ® Kenna Marie Sleigh, 1995 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Scuao l C F /J^^si^JC. The University of British Columbia Vancouver, Canada Date flue^us r z-< , /'its DE-6 (2/88) 11 Abstract In dual-earner research, there i s a dearth of psychometrically sound instruments that measure r o l e q u a l i t y . The author examined the r e l i a b i l i t y and v a l i d i t y of a new instrument, the Role Enactment Questionnaire (REQ). The REQ measures two r o l e stressors: r o l e i n t e n s i t y and r o l e d i s p a r i t y , i n the paid-work, spouse, parent, and i n d i v i d u a l r o l e s . R e l i a b i l i t y was assessed by c a l c u l a t i n g the REQ 1s in t e r n a l consistency and t e s t - r e t e s t r e l i a b i l i t y . Construct v a l i d i t y was evaluated through confirmatory factor analysis (CFA) and hypothesis t e s t i n g . The REQ was administered to 165 dual-earner parents with children s i x years or younger. The sample was very homogeneous with respect to age, education, and income l e v e l s . With the exception of two subscales that captured the i n t e n s i t y dimension i n the paid-work and i n d i v i d u a l r o l e s , Cronbach's alphas were high. Test-retest calculations indicated that the REQ was stable. Ten hypotheses were developed from demographic predictors of r o l e i n t e n s i t y and r o l e d i s p a r i t y . Seven of the hypotheses were confirmed by s t a t i s t i c a l l y s i g n i f i c a n t r e s u l t s . CFA was somewhat supportive of construct v a l i d i t y but also demonstrated some lack of consistency i n capturing the construct being measured by the instrument. Hypothesis t e s t i n g i d e n t i f i e d the continuing influence I l l of gender r o l e norms on the dual-earner l i f e s t y l e , i n t h i s sample. A t r a d i t i o n a l d i v i s i o n of labour dominated the a l l o c a t i o n of domestic duties and work-role involvement. Nurses who work with dual-earner parents can use demographic predictors to i d e n t i f y i n d i v i d u a l s vulnerable to ro l e stress as measured by ro l e i n t e n s i t y and r o l e d i s p a r i t y . S i m i l a r l y , administrators and nurse educators can use these predictors to address issues a f f e c t i n g personnel and students, respectively. Although the study recommends some refinement of the REQ based on r e l i a b i l i t y and v a l i d i t y r e s u l t s , the instrument shows promise for use i n the f i e l d of dual-earner research. Table of Contents Abstract Table of Contents L i s t of Tables L i s t of Figures Acknowledgements CHAPTER ONE: INTRODUCTION Background Problem Statement Purpose Research Questions D e f i n i t i o n of Terms Conceptual Framework Hypotheses Role Intensity Role Disparity Conclusion Chapter Summary CHAPTER TWO: REVIEW OF THE LITERATURE The Nature and Impact of Roles Aspects of Role Quality Role Issues Associated with Dual-earner Families with Young Children Demands of the Parent Role Demands of the Spouse Role Demands of the Paid-work Role Demands of the Individual Role The Interaction of Role Demands and Attitudes Control A c c e p t a b i l i t y The Importance of Contextualization i n Dual-earner Research Instrumentation Description of the REQ Psychometric Properties of the REQ R e l i a b i l i t y V a l i d i t y C r i t i q u e of the REQ i v Page 11 i v v i i v i i i i x 1 1 7 7 7 8 9 10 10 12 13 13 15 16 20 24 24 26 28 30 32 33 3 5 38 41 44 44 45 46 47 V Methods of Supporting Construct V a l i d i t y 49 Conclusion 51 Chapter Summary 52 CHAPTER THREE: METHOD 53 Sample 53 Sample Recruitment 53 Sample C r i t e r i a 54 Inclusion C r i t e r i a 54 Exclusion C r i t e r i a 55 Study Procedures 55 R e l i a b i l i t y Testing 55 V a l i d i t y Testing 56 Hypothesis Testing 56 Confirmatory Factor Analysis 57 Data Analysis 57 Assumptions 58 Limitations 59 E t h i c a l Considerations 60 Benefits of P a r t i c i p a t i o n 61 Risks from P a r t i c i p a t i o n 61 Conclusion 61 Chapter Summary 62 CHAPTER FOUR: PRESENTATION AND DISCUSSION OF FINDINGS 63 Sample Charact e r i s t i c s 63 R e l i a b i l i t y 68 Internal Consistency 68 Test-retest R e l i a b i l i t y 73 Construct V a l i d i t y 77 Hypothesis Testing 77 Role Intensity 78 Role Disparity 93 Confirmatory Factor Analysis 102 Analysis one (intensity and disparity) 102 Analysis two (intensity i n four roles) 106 Analysis three (disparity i n four roles) 110 Chapter Summary 113 v i CHAPTER FIVE: CONCLUSIONS AND IMPLICATIONS 115 Conclusions from Tool Testing 115 R e l i a b i l i t y and V a l i d i t y of the REQ 115 Implications for Tool Development 118 Tool Revision 118 The Paid-work Role 118 The Spouse Role 12 0 The Parent Role 121 The Individual Role 121 Future Psychometric Testing of the REQ 122 U t i l i t y of the REQ for Nursing 12 3 Conclusions from Hypothesis Testing 124 Implications for Nursing 125 C l i n i c a l Practice 127 Nursing Education 127 Nursing Administration 128 Nursing Research 13 0 Chapter Summary 131 Concluding Comments 131 Thesis Summary 132 REFERENCES 134 APPENDIX A: The Role Enactment Questionnaire 156 APPENDIX B: Cover Letter 165 v i i L i s t of Tables Table Page Table 1. Number and ages of children 64 Table 2. Dual-earners' hours of work 66 Table 3 . Internal consistency: Comparison of for r o l e dimensions findings 69 Table 4. Internal consistency: Comparison of for r o l e components findings 70 Table 5. Test-retest r e l i a b i l i t y : Comparison findings for r o l e dimensions of 74 Table 6. Test-retest r e l i a b i l i t y : Comparison findings for r o l e components of 75 Table 7. Test scores for dual-earner men and women 78 Table 8. Intensity scores for dual-earner women 84 Table 9. Intensity scores for dual-earners 85 Table 10. Test scores i n the paid-work r o l e for dual-earners 89 Table 11. Number of working hours for men and women 92 Table 12 . Disparity scores for dual-earners 93 v m L i s t of Figures Figure Page Figure l . The antecedents of r o l e s t r a i n 10 ix Acknowledgements I wish to praise my thesis chair, Wendy H a l l , for her uncanny a b i l i t y to know when to push, when to p u l l , and when to l e t go. I am also indebted to the other members of my thesis committee, Marion Clauson and Ray Thompson, for t h e i r patience, support, and encouragement. I must also thank my employers, Dr. Anthony Chow and Dr. Grant Stiver, for giving me the time and space I needed to f i n i s h t h i s project. F i n a l l y , a huge debt of gratitude i s owed to my friends and family. Many thanks to Linda and Gary; Anna, Cyndi, and Linda; my s i s t e r , Chick; and my nephew, Michael. 1 CHAPTER ONE Background One of the most dramatic s o c i a l changes that has occurred i s the i n f l u x of women into the labour force (Ghalam, 1993). S t a t i s t i c s reveal that i n 1991, 45% of the labour force consisted of women, compared with 36% i n 1975 (Ghalam, 1993). A s i g n i f i c a n t feature of t h i s trend i s the increased number of married women p a r t i c i p a t i n g i n paid employment. In 1991, 56% of married Canadian women were employed outside the home (Ghalam, 1993). By v i r t u e of t h i s labour force p a r t i c i p a t i o n the number of dual-earner families has increased to 62%, making t h i s model of Canadian l i f e the norm (Ghalam, 1993). Of note i s the number of married women who are remaining i n paid-work positions during t h e i r c h i l d rearing years. Within t h i s group, women with preschool children have shown the greatest increase (Cowan & Cowan, 1992; Googins & Burden, 1987; Levanthal-Belfer, Matthews, & Rodin, 1989; Presser, 1988). Since 1981 the proportion of employed Canadian mothers with preschoolers rose from 42% to 57% (Ghalam, 1993). The dual-earner occupies the s o c i a l roles of worker, spouse, parent, and ind i v i d u a l (Hall, 1987). This multiple r o l e pattern leads to a complex l i f e s t y l e involving numerous obligations (Anderson-Kulman & Paludi, 1986). A concern 2 that the obligations associated with multiple r o l e s may prove challenging to individuals has increasingly permeated the research l i t e r a t u r e (Bolger, Delongis, Kessler, & Wethington, 1989; Hirsch & Rapkin, 1986; Verbrugge, 1986). Researchers examining multiple roles have been guided by one of two t h e o r e t i c a l perspectives: the s c a r c i t y hypothesis or the expansion hypothesis (Froberg, Gjerdingen, & Preston, 1986; Marks, 1977). Both perspectives are based on the concept of human energy. However, the outcomes of multiple r o l e involvement predicted by the two hypotheses are dia m e t r i c a l l y opposed. The s c a r c i t y hypothesis predicts r o l e s t r a i n for multiple r o l e incumbents, while the expansion hypothesis predicts r o l e g r a t i f i c a t i o n (Goode, 1960; Marks, 1977). The presence of findings supportive of both perspectives suggests multiple roles can r e s u l t i n both harmful and b e n e f i c i a l outcomes (Hall, 1993; Marshall & Barnett, 1993). Of the two hypotheses, the expansion hypothesis has received the most empirical support (Barnett, Biener, & Baruch, 1987; Froberg et a l . , 1986; Repetti, Matthews, & Waldron, 1989). However, both the expansion hypothesis and the s c a r c i t y hypothesis are limited because they focus on the number of roles an i n d i v i d u a l performs (Piechowski, 1992; Pietromonaco, Manis, & Frohardt-Lane, 1986). Froberg et a l . (1986) argue that "the consequence of r o l e 3 accumulation should depend not only on the number of rol e s occupied but on types of roles and t h e i r c h a r a c t e r i s t i c s " (p. 84). Variations i n the qu a l i t y of roles allow for considerable i n d i v i d u a l differences i n the outcomes associated with ro l e s . Study findings that affirm the benign e f f e c t of multiple roles but f a i l to consider r o l e q u a l i t y may mask the r i s k s of ro l e s t r a i n (Froberg et a l . , 1986; Hirsch & Rapkin, 1986; Marshall & Barnett, 1993). The dual-earner couple with young children represents a population subgroup that experiences intense and frequently non-negotiable r o l e demands. Such ro l e demands may attenuate the b e n e f i c i a l e f f e c t s of multiple roles and r e s u l t i n r o l e s t r a i n (Kessler & McRae, 1982; Waldron & Jacobs, 1989). To address these issues, Hirsch and Rapkin (198 6) suggest researchers i d e n t i f y the conditions which account for such divergent outcomes as ro l e s t r a i n and r o l e g r a t i f i c a t i o n . Unfortunately, t h i s l i n e of research has not been systematically investigated (Burke & McKeen, 1988; Guelzow, Bird, & Koball, 1991). Currently, no empirical model i s consistently used to explain the re l a t i o n s h i p of rol e antecedents to ro l e outcomes. This r e s u l t s i n considerable d i f f i c u l t y interpreting and synthesizing extant study findings. 4 In order to determine the e f f e c t s of r o l e antecedents on r o l e outcomes i t i s important to develop and t e s t comprehensive models. Pear l i n (1989) suggests researchers use a s o c i o l o g i c a l model of the stress process to investigate r o l e phenomena. Ward (1986) has developed a t h e o r e t i c a l framework which describes the antecedents and consequences of r o l e stress. Ward's (1986) framework proposes r o l e stress antecedents which "do not always lead to r o l e s t r a i n , but i n fact may produce the opposite, desirable" outcome of r o l e g r a t i f i c a t i o n (Ward, 1986, p. 44). Mediating conditions determine the r o l e outcome. Integral to operationalizing and t e s t i n g the antecedent components of Ward's (1986) model i s r e l i a b l e and v a l i d instrumentation. In order to be useful an instrument must be consistent with Ward's (1986) framework, r e l a t e to t h e o r e t i c a l l y clear constructs, and be capable of simultaneously measuring r o l e stress antecedents which a r i s e from the work and family domains (Frone, Russell & Cooper, 1991; Parry & Warr, 1980). Furthermore, the t o o l should be suitable for both genders. Unfortunately, current published instruments f a i l to meet these c r i t e r i a . Some tools are not structured to capture antecedents of r o l e s t r a i n without the assumption of inherent negative outcomes. The Perceived Stress Scale (Cohen, Kamarck, & Mermelstein, 1983) precludes the 5 p o s s i b i l i t y of a po s i t i v e r o l e outcome for respondents with high scores. The Perceived Work-Family C o n f l i c t Scale (Loerch, Russell, & Rush, 1989) also implies r o l e s t r a i n i s an inev i t a b l e outcome of ro l e stress. The Stress Index (Reifman, Biernat, & Lang, 1991) muddies the conceptual waters by combining items that measure r o l e stressors with items that measure r o l e s t r a i n . Other instruments have been developed a t h e o r e t i c a l l y and are tailor-made for use i n a single study (Amatea & Fong, 1991; Hibbard & Pope, 1985; Kandel, Davies, & Raveis, 1985; Sears & Galambos, 1992; Voydanoff, 1988). Such tools do not accurately r e f l e c t r o l e s t r a i n antecedents. Moreover, c r i t i c s lament the use of "one-shot" investigator-designed instruments because psychometric evaluation i s seldom available (Johnson, 1989; Pietromonaco et a l . , 1986). Many instruments f a i l to measure the combined e f f e c t s of family and work ro l e s . The Stress Diagnostic Survey (Ivancevich & Matteson, 1980) and Burke's (1988) Work Stressors t o o l measure work stress, only. S i m i l a r l y , the Parenting Stress Index (Abidin, 1980) and the Parenting Daily Hassles Scale used by Crnic and Booth (1991) both tap soley into the family r o l e domain. Other instruments are unsuitable for men. Parry and Warr (1980) developed the HERS attitude scales for working 6 class women with young children. The Lengacher Role Strain Inventory (Lengacher, 1993) measures ro l e s t r a i n i n female nursing students. Hall's Everyday Stressors Index (1990) i s used exclusively i n research that studies mothers. Baruch and Barnett's (1986) rewards and concerns scales are theory driven and assess the parental, occupational, and marital roles for both genders. However, the instruments do not i s o l a t e and quantify discrete r o l e antecedents that may impact on r o l e outcomes. Moreover, rewarding c h a r a c t e r i s t i c s of each r o l e are subtracted from d i s t r e s s i n g r o l e c h a r a c t e r i s t i c s which produces a balance score (Baruch & Barnett, 1986). This method of scoring masks the r o l e aspects that contribute to outcomes. Hall's (1993) Role Enactment Questionnaire (REQ) i s an instrument that may not be subject to the l i m i t a t i o n s associated with other t o o l s . The REQ (Appendix A) measures two r o l e stress antecedents proposed i n Ward's (1986) framework, r o l e i n t e n s i t y and r o l e d i s p a r i t y . The t o o l i s capable of measuring work and non-work stress i n both genders. Preliminary t e s t i n g suggests the REQ i s r e l i a b l e and v a l i d (Hall, 1993). But, before t h i s t o o l can be used by researchers or c l i n i c i a n s to measure ro l e stress antecedents inherent i n the dual-earner l i f e s t y l e , in-depth psychometric t e s t i n g must be conducted. 7 Problem Statement Progress towards understanding r o l e s t r a i n i n the dual-earner population i s hampered because current instruments do not focus on r o l e stress or antecedents to s t r a i n . The recently developed REQ (Hall, 1993) i s a t o o l that measures r o l e i n t e n s i t y and r o l e d i s p a r i t y i n the paid-work, spouse, parent, and ind i v i d u a l r o l e s . Furthermore, r o l e i n t e n s i t y and ro l e d i s p a r i t y are concepts that are not ubiquitously associated with r o l e s t r a i n . These concepts also have the poten t i a l to r e s u l t i n ro l e g r a t i f i c a t i o n . Thus the REQ may be u t i l i z e d i n research examining the po s i t i v e and negative outcomes of ro l e q u a l i t y . Although preliminary t e s t i n g supports the r e l i a b i l i t y and v a l i d i t y of the t o o l , more extensive psychometric t e s t i n g i s warranted. Therefore, the r e l i a b i l i t y and v a l i d i t y of the REQ require assessment. Purpose The purpose of t h i s study i s to assess the r e l i a b i l i t y and v a l i d i t y of the REQ which measures two antecedents of rol e s t r a i n : r o l e i n t e n s i t y and ro l e d i s p a r i t y . Research Questions 1. What i s the inter n a l consistency of the Role Enactment Questionnaire? 2. What i s the t e s t - r e t e s t r e l i a b i l i t y of the Role Enactment Questionnaire? 8 3. What i s the construct v a l i d i t y of the Role Enactment Questionnaire? D e f i n i t i o n of Terms 1. Internal consistency: Internal consistency evaluates the homogeneity of an instrument and i s defined as the "consistency of performance of a group of i n d i v i d u a l s across the items on a single t e s t " (Waltz & Bausell, 1981, p. 62) . 2. Test-retest r e l i a b i l i t y : Test-retest r e l i a b i l i t y measures the s t a b i l i t y of an instrument and i s defined as "the consistency of performance one measure e l i c i t s from one group of subjects on two separate measurement occasions" (Waltz, Strickland, & Lenz, 1984, p. 135). 3. Construct v a l i d i t y : A "tool's construct v a l i d i t y i s based on the extent that a t e s t measures a t h e o r e t i c a l construct or t r a i t " (LoBiondo-Wood & Haber, 1986, p. 187). Conceptual Framework Hall's (1993) ro l e antecedent framework i s used to support hypotheses that t e s t the construct v a l i d i t y of the REQ. The framework proposes that s t r a i n and g r a t i f i c a t i o n r e s u l t from r o l e stress antecedents for in d i v i d u a l s engaged i n the paid-work, spouse, parent, and i n d i v i d u a l r o l e s . Two antecedents are i d e n t i f i e d , r o l e i n t e n s i t y and r o l e d i s p a r i t y . Role i n t e n s i t y describes the amount of time and e f f o r t an i n d i v i d u a l devotes to the four r o l e components. 9 Role d i s p a r i t y represents the discrepancy an i n d i v i d u a l experiences between ro l e "expectations, e x i s t i n g attitudes, and actual behaviours" (Hall, 1993, p. 59). Because the l i t e r a t u r e demonstrates that r o l e stress antecedents and demographic resources are correlated, H a l l (1993) suggests researchers may predict r o l e i n t e n s i t y and ro l e d i s p a r i t y by assessing demographic resources. Thus, i n the conceptual framework, demographic variables such as number of children, number of paid-work hours, education, income, gender, and age are depicted as a f f e c t i n g the degree of r o l e i n t e n s i t y and d i s p a r i t y that i n d i v i d u a l s experience (see Figure 1). 10 Figure 1. The antecedents of ro l e s t r a i n Demographic R e s o u r c e s rtiilttiiiiil « # c h i l d r e n * ; *s# work hrs*; ^ E d u c a t i o n s sslncome ' , , ; sGender Age R o l e I n t e n s i t y Hi in | Work r o l e s | Spouse r o l e j P a r e n t r o l e i n d i v i d u a l r o l e s Hi: I n t e r v e n i n g C o n d i t i o n s mm R o l e D i s p a r i t y f=>^  C o p i n g (=jf>^ R o l e S t r a i n R o l e G r a t i f i c a t i o n Note. From "Development and early t e s t i n g of the Role Enactment Questionnaire" by W. A. H a l l , 1993. Canadian Journal of Nursing Research. 25(4), p. 65. Adapted by permission. Hypotheses The following hypotheses predict the relationships between demographic variables and the indicators of ro l e stress: r o l e i n t e n s i t y and role d i s p a r i t y . Role Intensity 1. Women are primarily responsible for c h i l d care (Crompton, 1991; Marshall & Barnett, 1993). Thus, dual-earner mothers w i l l score higher on the role i n t e n s i t y dimension of the parenting role than dual-earner fathers. 11 2. Both genders tend to overperform i n one primary r o l e : Women focus on the family r o l e and men focus on the work r o l e (Barnett et a l . , 1987; Froberg et a l . , 1986). Therefore, dual-earner fathers w i l l have higher r o l e i n t e n s i t y scores i n the paid-work r o l e than dual-earner mothers. 3. Greater numbers of children w i l l r e s u l t i n increased domestic r e s p o n s i b i l i t i e s for women (Cleary & Mechanic, 1983; Schnittger & Bird, 1990; Verbrugge, 1986). I t i s hypothesized that dual-earner mothers with more than one c h i l d w i l l have higher scores on the r o l e i n t e n s i t y dimension of the parent and spouse roles than dual-earner mothers with one c h i l d . 4. Individuals who earn high incomes probably experience greater occupational r o l e demands and r e s p o n s i b i l i t i e s . Therefore, dual-earners with high incomes w i l l have higher r o l e i n t e n s i t y scores i n the paid-work r o l e than dual-earners with low incomes. 5. An increase i n the number of paid-work hours increases an individual's work ro l e demands. Thus, dual-earners who work f u l l - t i m e w i l l have higher r o l e i n t e n s i t y scores i n the paid-work r o l e than dual-earners who work part-time. 12 Role Disparity 6. Dual-earners with more than one c h i l d w i l l experience greater d i s p a r i t y i n the spouse and i n d i v i d u a l r o l e s than dual-earners with one c h i l d . 7. Dual-earners who work f u l l - t i m e (over 30 hours a week) w i l l experience greater d i s p a r i t y i n the paid-work r o l e than dual-earners who work part-time (30 hours a week or l e s s ) . 8. Some men experience d i f f i c u l t y creating a s a t i s f y i n g fatherhood r o l e (Barnett et a l . , 1987; Hanson & Bozett, 1987; Rotundo, 1985; Scott & Alwin, 1989). Barriers to parenting for fathers include: heavy work demands, ambiguous ro l e expectations, and the reluctance of mothers to y i e l d c h i l d care t u r f (Baruch & Barnett, 1986; Cowan & Cowan, 1988; Hanson, 1985). I t i s expected that dual-earner fathers w i l l have higher role d i s p a r i t y scores i n the parent r o l e than dual-earner women. 9. Women are less l i k e l y than men to engage i n a c t i v i t i e s that are s e l f - r e l a t e d (Hall, 1992; Schnittger & Bird, 1990). Thus, dual-earner mothers w i l l have higher r o l e d i s p a r i t y scores i n the ind i v i d u a l r o l e than dual-earner fathers. 10. The l i t e r a t u r e reports that women with more education experience greater congruence between t h e i r employment status and t h e i r conception of appropriate female 13 behaviour (Kessler & McRae, 1982). Therefore, dual-earner mothers working f u l l - t i m e who have college or professional preparation w i l l have lower r o l e d i s p a r i t y scores i n the paid-work r o l e than dual-earner mothers working f u l l - t i m e who have a high school diploma or l e s s . Conclusion Dual-earner parents of preschoolers experience unique r o l e conditions that may lead to r o l e s t r a i n or g r a t i f i c a t i o n . The challenge for researchers i s to determine how r o l e conditions influence outcomes. Hall' s (1993) REQ may be an instrument i d e a l l y suited to meet t h i s challenge. The REQ i s suitable for both genders and quantifies two indicators of r o l e stress: i n t e n s i t y and d i s p a r i t y , i n the paid-work, spouse, parent, and i n d i v i d u a l r o l e s . Preliminary psychometric t e s t i n g of the REQ has yielded promising r e s u l t s . However, additional t e s t i n g must be performed before the REQ can be used by researchers or c l i n i c a l p r a c t i t i o n e r s . This study proposes to address t h i s problem by t e s t i n g the r e l i a b i l i t y and v a l i d i t y of the REQ. Chapter Summary In the f i r s t chapter, the background issues associated with dual-earner research were introduced. The problem statement, purpose, and research questions were presented. In addition, the reader was provided with a d e f i n i t i o n of the terms used i n the study. F i n a l l y , the conceptual 14 framework and the hypotheses derived from i t were deta i l e d . In order to contribute to the knowledge base, the study "must be placed i n the context of what s c i e n t i f i c work has gone before" (Wilson, 1987, p. 95). Therefore, i n the second chapter, the reader i s offered a comprehensive and c r i t i c a l review of the l i t e r a t u r e . 15 CHAPTER TWO Review of the Literature Due to increases i n job opportunities, changes i n sex-role attitudes, and a steady decline i n fa m i l i e s ' spending power, women have increased t h e i r p a r t i c i p a t i o n i n the labour force (Smith, 1979). The most s t a r t l i n g aspect of t h i s trend i s the number of mothers who work during t h e i r children's preschool years. In addition to the f i n a n c i a l necessity to continue working while children are very young, mothers are unwilling to interrupt t h e i r career plans (Power & Parke, 1984). Women's labour force p a r t i c i p a t i o n has resulted i n a s i g n i f i c a n t increase i n the number of dual-earner families with young children. Unfortunately, the l i t e r a t u r e r e f l e c t s , a disturbing lack of consensus regarding the e f f e c t s of combining the paid-work, spouse, parent, and in d i v i d u a l roles (Hall, 1993; Johnson & Johnson, 1980). The following review of the l i t e r a t u r e addresses a number of areas associated with t h i s issue including: (a) the nature and impact of roles, (b) aspects of ro l e quality, (c) r o l e issues associated with dual-earner families with young children, (d) the importance of contextualization i n dual-earner research and, (e) instrumentation. 16 The Nature and Impact of Roles Two dominant t h e o r e t i c a l perspectives i n the l i t e r a t u r e propose c o n f l i c t i n g outcomes associated with multiple r o l e involvement: the s c a r c i t y hypothesis and the expansion hypothesis (Froberg et a l . , 1986). The s c a r c i t y hypothesis posits that multiple roles are harmful because they act as a drain on the individual's energy resources (Goode, 1960). Energy i s u t i l i z e d when individuals meet ro l e demands; as r o l e s accumulate, demands increase, and obligations mount. Attempts to f u l f i l t o t a l r o l e obligations r a p i d l y deplete the i n d i v i d u a l ' s f i n i t e energy reserves. Energy supplies become exhausted and the i n d i v i d u a l i s overwhelmed. The r e s u l t i s r o l e s t r a i n , "the f e l t d i f f i c u l t y i n f u l f i l l i n g r o l e obligations" (Goode, 1960, p. 483) . Although some ro l e s t r a i n i s inevitable, high l e v e l s may prove deleterious (Goode, 1960). Therefore, the i n d i v i d u a l attempts to reduce the l e v e l of s t r a i n experienced. The i n d i v i d u a l makes ro l e decisions and bargains which are designed to diminish r o l e demands (Barnett et a l . , 1987; Goode, 1960; Pietromonaco et a l . , 1986). If r o l e s t r a i n i s not reduced to manageable le v e l s the i n d i v i d u a l may experience mental and/or physical d i s t r e s s (Goode, 1960; Scott & Alwin, 1989). Researchers have used a variety of outcome variables to measure r o l e s t r a i n . Strain i s r e f l e c t e d i n negative 17 a f f e c t i v e states and may manifest as: fatigue, exhaustion, embarrassment, worry, d i s s a t i s f a c t i o n , somatization, discomfort, depression, anxiety, obsessive compulsiveness, and a n g e r / h o s t i l i t y (McBride, 1990; Ward, 1986). Indices of job d i s s a t i s f a c t i o n , such as absenteeism and turnover, are also used to operationalize r o l e s t r a i n (Ward, 1986). H a l l (1991) i d e n t i f i e s marital c o n f l i c t , family d i s s a t i s f a c t i o n , and i l l n e s s as empirical referents of r o l e s t r a i n . In contrast to the s c a r c i t y hypothesis, proponents of the expansion hypothesis argue that s o c i a l involvements v i t a l i z e individuals (Marks, 1977). Marks (1977) proposes a c t i v i t y stimulates the body to produce energy and that human beings have perpetually renewing energy resources. Conditions determine whether or not energy i s available for ro l e a c t i v i t i e s (Marks, 1977). "Persons construct t h e i r response to the demands of others" (Marks, 1977, p. 927). Thus, energy can be "found for anything to which we are highly committed" (Marks, 1977, p. 927). Roles i n which an in d i v i d u a l i s committed "may even create energy for use i n that r o l e or i n other r o l e performances" (Marks, 1977, p. 926) . Multiple r o l e l i f e s t y l e s can also be extremely g r a t i f y i n g . Society provides "both monetary and non-monetary rewards for ro l e performance" (Hirsch & Rapkin, 1986, p. 1237). P r i v i l e g e s , status, security, self-esteem, 18 personality enrichment, and s o c i a l relationships are associated with multiple r o l e enactment (Froberg et a l . , 1986; P u g l i e s i , 1989; Thoits, 1983). In addition, multiple r o l e incumbents who experience g r a t i f i c a t i o n i n one r o l e may be protected from negative experiences i n other ro l e s (Barnett, Marshall, & Sayer, 1992). The expansion hypothesis also acknowledges the existence of r o l e s t r a i n , but s t r a i n i s attributed to an imbalance i n commitment le v e l s among various roles i n the set (Marks, 1977). When one rol e i s preferred over others, the i n d i v i d u a l spends more time performing a c t i v i t i e s associated with i t (Marks, 1977). Over time, the preferred r o l e usurps time and energy previously a l l o t t e d to other r o l e s . When the in d i v i d u a l neglects r o l e a c t i v i t i e s associated with the undercommitted roles, r o l e s t r a i n occurs. Findings i n the l i t e r a t u r e support both the s c a r c i t y and the expansion hypotheses. However, studies have consistently shown that individuals occupying the greatest number of role s report the lowest l e v e l s of s t r a i n (Amatea & Fong, 1991; Barnett, Marshall, & Singer, 1992; Hibbard & Pope, 1985; Pietromonaco et a l . , 1986; Thoits, 1983; Waldron & Jacobs, 1989). The contention that there i s a simple r e l a t i o n s h i p between ro l e quantity and ro l e s t r a i n has not been supported 19 i n the l i t e r a t u r e . Theorists surmise that r o l e s t r a i n and r o l e g r a t i f i c a t i o n may r e s u l t from the mediation of stressors associated with multiple r o l e involvement (McBride, 1990; Piechowski, 1992; Rice, Frone, & McFarlin, 1992). However, confirmation of t h i s hypothesis awaits the t e s t i n g of comprehensive models which include the antecedents and consequences of r o l e stress (Burke, 1988). Ward (1986) provides a clear conceptual framework that l i n k s antecedent stressors to consequences (role s t r a i n or r o l e g r a t i f i c a t i o n ) through mediating conditions. In t h i s framework, the following stressors precede r o l e outcomes: r o l e c o n f l i c t , r o l e accumulation, r o l e r i g i d i t i e s , r o l e ambiguity, r o l e incompetence, r o l e o v e r q u a l i f i c a t i o n , r o l e incongruity, r o l e uncertainty, r o l e incompatibility, r o l e tedium, r o l e d i s p a r i t y , and r o l e i n t e n s i t y . Intervening conditions determine whether these stressors r e s u l t i n r o l e s t r a i n or r o l e g r a t i f i c a t i o n . The d i f f i c u l t y with Ward's (1986) conceptual framework i s that none of the antecedent stressors are well defined. Theorists emphasize the importance of c l e a r l y defining and contextualizing antecedent stressors. Campbell and Moen (1992) believe that understanding these stressors may lead to solutions to r o l e s t r a i n . Moreover, Pear l i n (1989) i n s i s t s the antecedents to s t r a i n : 20 need to be understood i n terms of process, whereby broad structured and i n s t i t u t i o n a l forces [and] constellations ... converge ... to a f f e c t peoples* well-being. We must guide our e f f o r t s not simply by i d e n t i f y i n g and adding together a l l factors that might contribute to the variance of an outcome but also by asking how and why these contributions came about (p. 249) . Fa i l u r e to understand t h i s process l i m i t s researchers' a b i l i t i e s to implicate s p e c i f i c r o l e a t t r i b u t e s as aspects of r o l e stress which p o t e n t i a l l y lead to r o l e s t r a i n . Furthermore, without a clear understanding of the inte r a c t i o n among ro l e stress and mediators, c l i n i c i a n s are hampered from developing research-based strategies that may prevent or reduce r o l e s t r a i n (Wethington & Kessler, 1986). Aspects of Role Quality Most researchers have s h i f t e d t h e i r attention from r o l e quantity to r o l e q u a l i t y . Froberg et a l . , (1986) suggest r o l e q u a l i t y be considered because "the same ro l e can be experienced d i f f e r e n t l y depending on the person and the circumstances" (p. 87). "Both gender-socialized i n d i v i d u a l c h a r a c t e r i s t i c s and the broader s o c i o c u l t u r a l context may play a r o l e " i n determining ind i v i d u a l s ' responses (Piechowski, 1992, p. 134). Thus, "the perceived q u a l i t y of the r o l e appears to be a better predictor of well-being than 21 the mere occupancy of that p o s i t i o n " (Hirsch & Rapkin, 1986, p. 1238) . Part of the examination of ro l e q u a l i t y includes a closer look at ro l e conditions. The following conditions have been i d e n t i f i e d as sources of s t r a i n : r o l e incongruity, r o l e incompatibility, r o l e c o n f l i c t , r o l e ambiguity, and ro l e overload (Burr, 1976; Goode, 1960; Ward, 1986). However, only a few of these constructs have been tested empirically. Researchers have examined the ef f e c t s that r o l e ambiguity, r o l e c o n f l i c t , and ro l e overload exert on outcomes (Avant, 1988; Elman & Gi l b e r t , 1984; Pleck, 1985; Rudd & McKenry, 1986; Shehan, 1984). Some investigators f i n d that these constructs have negative e f f e c t s . For instance, Bacharach and Bamberger (1992) demonstrate r o l e ambiguity, c o n f l i c t , and overload are related to burn-out and turn-over intentions. Barling and Macewen (1992) report r o l e c o n f l i c t and ambiguity are associated with impaired concentration and depression. Role c o n f l i c t i s linked to low l e v e l s of marital adjustment (Greenhaus, Bedeian, & Mossholder, 1987) and depression, anxiety, and somatization in men and women (Greenglass, Pantony, & Burke, 1988). In contrast, other researchers conclude r o l e ambiguity, ro l e overload, and r o l e c o n f l i c t have no e f f e c t on outcomes (Coverman, 1989; Voydanoff, 1988). Voydanoff (1988) reports 22 r o l e ambiguity and c o n f l i c t are not related to work/family c o n f l i c t — a construct the author describes as being " s i m i l a r to, but narrower than, r o l e s t r a i n " (p. 749). S i m i l a r l y , Amatea and Fong (1991) f i n d "neither reported r o l e demand l e v e l nor i n t e r n a l r o l e c o n f l i c t are p o s i t i v e l y correlated with s t r a i n at a s i g n i f i c a n t l e v e l " (p. 427). Verbrugge (1986) notes r o l e overload (denoted by long work hours) i s not correlated with poor health. Role c o n f l i c t and r o l e overload have also been associated with p o s i t i v e outcomes. Baruch and Barnett (1986) conclude minimal l e v e l s of r o l e c o n f l i c t are a stimulus to performance. In an ethnography, Jones (1993) reports administrators experienced an "energizing e f f e c t r e s u l t i n g from the ongoing challenge of dealing with c o n f l i c t " (p. 136). Kandel et a l . (1985) f i n d subjects with high workload l e v e l s report less r o l e - r e l a t e d stress. Methodological shortcomings may account for contradictory findings. In t h e i r meta-analysis of r o l e c o n f l i c t and ambiguity, Fisher and Gitelson (1983) conclude some v a r i a b i l i t y i n c o r r e l a t i o n a l r e s u l t s across samples may be due to s t a t i s t i c a l a r t i f a c t s and the confounding e f f e c t of moderators. Coverman (1989) i n s i s t s discrepancies may also be related to researchers' tendency to use r o l e ambiguity, c o n f l i c t , and overload interchangeably, even though each construct represents a d i s t i n c t conceptual 23 domain. Moreover inconsistent findings may r e s u l t from var i a t i o n s i n the d e f i n i t i o n s and operationalizations of the constructs (Johnson, 1989). Furthermore many instruments u t i l i z e d to measure the key constructs possess only moderate degrees of v a l i d i t y and r e l i a b i l i t y (Johnson, 1989). F i n a l l y , researchers disagree on how to model the constructs' t h e o r e t i c a l relationships with r o l e s t r a i n . Anderson-Kulman and Paludi (1986) hypothesize r o l e c o n f l i c t functions as a precursor to r o l e s t r a i n . P e a r l i n (1989) and Repetti et a l . (1989) treat r o l e c o n f l i c t and overload as two types of r o l e s t r a i n , rather than precursors to r o l e s t r a i n . Several t h e o r i s t s conceptualize r o l e c o n f l i c t and overload as the primary ingredients of r o l e s t r a i n (Guelzow, et a l . , 1991; Ross & Mirowsky, 1988). Researchers must c l e a r l y define and operationalize r o l e stress constructs. "The manner i n which [man] formulates concepts influences his understanding ... and becomes a source of d i f f i c u l t y when the meanings of concepts are diverse and the referents poorly s p e c i f i e d " (Berthold, 1964, p. 406). The lack of c l a r i t y i n the conceptualization and operationalization of r o l e constructs l i m i t s the usefulness of extant study findings. In order to operationalize constructs s a l i e n t to r o l e outcomes, t h e o r i s t s must f i r s t develop more rigorous ways of conceptualizing r o l e q u a l i t i e s (Amatea & Fong, 1991). 24 Voydanoff (1988) suggests looking at r o l e demands, the s t r u c t u r a l c h a r a c t e r i s t i c s of r o l e s . Role Demands Associated with Dual-earner Families with Young Children The l i t e r a t u r e indicates dual-earner parents experience intense r o l e demands during t h e i r children's preschool years. These demands are numerous and frequently non-negotiable. Dual-earners are challenged by the accumulation of r o l e demands associated with the parent, paid-work, spouse, and i n d i v i d u a l roles (Greenhaus, 1988; K e l l y & Voydanoff, 1986). Demands of the Parent Role Researchers have long suspected that younger children are associated with higher parental r o l e demands (Gove & Geerkin, 1977; Umberson & Gove, 1989). P e a r l i n and Schooler (1978) i n s i s t that "as children grow older time demands decline i n a jagged curve from the near continuous demands of newborn children to the r e l a t i v e l y complete autonomy of young adults" (p. 110). Moen and Dempster-McClain (1987) report "parents spend more time with t h e i r children i f the youngest c h i l d i s under six than i f the youngest i s of school age" (p. 585). Scarr, P h i l l i p s , and McCartney (1989) indicate women who are mothers of babies and young children spend more hours on t h e i r family roles than do mothers with older children. 25 Due to t h e i r high dependency l e v e l , young children create greater workloads (Walker & Best, 1991). Moreover young children are less able than older children to a s s i s t with household chores (Gove & Geerkin, 1977; Schnittger & Bird, 1990). Parents with large families experience even greater r o l e demands (Greenhaus & Beutell, 1985; Moen & Dempster-McClain, 1987). The number of children i s p o s i t i v e l y related to the number of hours parents spend performing c h i l d care and housekeeping tasks (Voydanoff, 1988). Responsibility for parenting also functions as a heavy r o l e demand (Pearlin & Schooler, 1978). Numerous studies suggest primary r e s p o n s i b i l i t y for c h i l d care continues to be the province of women (Hibbard & Pope, 1985; Kach & McGhee, 1982; Marshall & Barnett, 1993; Michelson, 1985; Pear l i n & Schooler, 1978; Scarr et a l . , 1989; Scott & Alwin, 1989; Suitor, 1991; Verbrugge, 1983). Although men are becoming increasingly involved with t h e i r children, they are "not a c t u a l l y responsible for the c h i l d care i n the way that t h e i r wives" are (Olds, Schwartz, Eisen, Betcher, & Van Ni e l , 1993, p. 11). Even when parents share c h i l d care " i t i s the women who most often implement the sharing—remembering when tasks need to be done, assigning tasks, and checking on task progress" (Schnittger & Bird, 1990, p. 201). 26 Mothers are also primarily responsible for ensuring t h e i r children's health and happiness (Marshall & Barnett, 1993; McBride, 1990; Michelson, 1985; Pea r l i n & Schooler, 1978). Changes i n ideology concerning human nature add to parental r e s p o n s i b i l i t y . In contrast with an e a r l i e r view i n which the c h i l d was considered a bundle of impulses requiring parental control, the current view i s that there i s r i c h p o t e n t i a l i n every c h i l d . Mothers are responsible for developing the p o t e n t i a l into a well-adjusted adult (Johnson & Johnson, 198 0). Demands of the Spouse Role A s i g n i f i c a n t demand for the dual-career couple i s creating and sustaining a harmonious spousal r e l a t i o n s h i p (Crouter, Perry-Jenkins, Huston, & McHale, 1987; Vannoy & P h i l l i b e r , 1992). Towards t h i s end couples must devise a mutually s a t i s f y i n g d i v i s i o n of household labour. H a l l ' s (1987) sample of working mothers f e l t that negotiating domestic r e s p o n s i b i l i t i e s with t h e i r husbands was "a way of maintaining t h e i r marital r e l a t i o n s h i p s " (p. 192). The l i t e r a t u r e suggests household chores are not evenly divided between men and women (Benin & A g o s t i n e l l i , 1988; Crouter et a l . , 1987; Fish, New, & Van Cleave, 1992). Women accept greater r e s p o n s i b i l i t y for domestic chores than men (Darling-Fisher & Tiedje, 1990; Geerken & Gove, 1983) . Women i n dual-earner families are more l i k e l y to r e t a i n sole 27 r e s p o n s i b i l i t y for grocery shopping and cooking (Marat & Finlay, 1984). Gunter and Gunter (1991) remark "women's motivation for the performance of household tasks ... l i k e l y represents the b e l i e f that i t [is] t h e i r r e s p o n s i b i l i t y , while men [are] more l i k e l y to f e e l that these tasks [are] not t h e i r job, and when they do them, i t [is] merely to help out" (p. 562). An additional r o l e demand for dual-earners i s f i n d i n g time to spend together. The pressing nature of couples' work schedules can mean a loss i n togetherness (Hall, 1992). The presence of young children also l i m i t s the time spouses can spend together, except, perhaps i n child-care roles (Kingston & Nock, 1987). Furthermore, by the time couples have fi n i s h e d work, t i d i e d the house, and put the children to bed, they are exhausted. There may be " l i t t l e emotional or physical energy l e f t to r e l a t e i n a meaningful way to a spouse" (Googins, 1991, p. 166). Kingston and Nock (1987) report dual-earner couples t r y hard to share time. Compared to single-earners, dual-earners spend only 30 minutes a day less with t h e i r spouses. However, although the amount of time together i s related to marital quality, Kingston and Nock (1987) conclude "the kind of time together also matters. The more time together i n a c t i v i t i e s such as eating, playing, and conversing the more s a t i s f y i n g the marriage" (p. 399). 28 Meleis and Stevens (1992) state that the most frequently c i t e d sources of marital s a t i s f a c t i o n for c l e r i c a l women are companionship and r e c i p r o c i t y of a f f e c t i o n . Some research suggests c e r t a i n spousal r o l e demands may be neglected by dual-earner parents with young children (Chassin, Zeiss, Cooper, & Reaven, 1985; Hughes, Galinsky, & Morris, 1992). Some couples with children under f i v e report a decrease i n intimacy and an increase i n c o n f l i c t (Olds et a l . , 1993). After analyzing 27 studies, Smith (1985) concludes that, compared to single-earner couples, dual-earners have a less rewarding sex l i f e and obtain lower scores on communication measures. Demands of the Paid-work Role Combining the parent r o l e with the paid-work r o l e increases o v e r a l l r o l e demands. Often young parents are establishing a career which may require a s i g n i f i c a n t degree of commitment to work-related tasks (Schnittger & Bird, 1990). Performing well i n the work ro l e can demand more time and e f f o r t than individuals expect (Greenhaus et a l . , 1987). Moen and Dempster-McClain (1987) report that, for t h e i r sample of dual-earners, the demands of the job were greater than either spouse preferred. An increase i n the number of paid-work hours increases work-role demands (Rosenfield, 1989). Employment 29 c h a r a c t e r i s t i c s such as job-related t r a v e l and commuting contribute to the amount of time parents spend i n the paid-work r o l e (Burke & McKeen, 1988). Moreover, working overtime and moonlighting create time shortages among employed parents (Voydanoff, 1988)• For one sample of employed parents, "overtime worked and the degree of d i s s a t i s f a c t i o n with work hours and scheduling" acted as s i g n i f i c a n t predictors of job tension (Kelly & Voydanoff, 1985, p. 373). The l i t e r a t u r e indicates men experience heavier paid-work demands than women. Sekaran (1985) reports family interests superseded career concerns for some professional women who perceived t h e i r careers as being highly s a l i e n t . Men of a l l marital and parental status categories generally work more hours a week than women (Googins, 1991). In Canada, women devote "an average of 6.0 hours per day to paid work and education compared with 7.2 hours for men" (Ghalam, 1993). Evidence suggests that i n order to cope with the dual-earner l i f e s t y l e , families l i m i t the job involvement of mothers (Froberg et a l . , 1986). Theorists a t t r i b u t e differences i n paid-work commitments to both genders' tendency to overperform i n t h e i r core r o l e s : Men focus on the work r o l e and women focus on the family r o l e (Barnett et a l . , 1987; Cleary & Mechanic, 1983). Bielby (1992) believes women are less 30 committed to the occupational r o l e due to gender differences i n the workplace. The majority of employed women "occupy jobs t r a d i t i o n a l l y assigned to females and characterized by low power, prestige, and pay" (Pittman & Orthner, 1988, p. 227). Women's limited access to occupational rewards and opportunities may reduce t h e i r attachment to the work r o l e (Bielby, 1992). Demands of the Individual Role Both experts and the general public are becoming increasingly aware of the important benefits associated with self-nurturance. I t i s no longer considered s e l f i s h for adults to devote time to exercise, s o c i a l interactions, and hobbies (Googins, 1991). However, parent, paid-work, and spouse r o l e demands l i m i t the amount of time a dual-earner can spend i n the indi v i d u a l r o l e (Geerken & Gove, 1983; Michelson, 1985; Schnittger & Bird, 1990). Women are at r i s k of neglecting s e l f - r e l a t e d a c t i v i t i e s . Researchers believe that when women are confronted with multiple r o l e demands they often place t h e i r own needs l a s t on the l i s t of p r i o r i t i e s (McLaughlin, Cormier, & Cormier, 1988; Walker & Best, 1991). Time for s e l f i s viewed "as one more thing to schedule into an already h e c t i c routine" (McLaughlin et a l . , 1988, p. 192). Walker and Best (1991) suggest one way employed mothers manage parenting within the context of time constraints i s 31 by adapting t h e i r l i f e s t y l e to these constraints. Women diminish t h e i r attention to personal health, s e l f -a c t u a l i z a t i o n , exercise, n u t r i t i o n , interpersonal support, and stress management (Walker & Best). The authors conclude f u l l - t i m e employed mothers of infants manage parenting r e s p o n s i b i l i t i e s i n ways that "do not damage t h e i r image of themselves as mothers, but ... they may do so i n part at the expense of t h e i r own well-being" (p. 85). Other studies lend credence to t h i s analysis. Anderson-Kulman and Paludi (1986) state the majority of working mothers i n t h e i r sample experienced a shortage of time for nearly everything and that t h i s time shortage was most accentuated i n s e l f - r e l a t e d areas (e.g., community a c t i v i t i e s , hobbies, reading, and physical f i t n e s s ) . Reifman et a l . (1991) suggest f u l l - t i m e employed women spend less time engaged i n s e l f - r e l a t e d a c t i v i t i e s such as personal care, sleep, and le i s u r e than part-time and nonemployed women. Verhoef, Love, and Rose (1992) note young mothers are the least l i k e l y to engage i n regular exercise. Women are more l i k e l y than men to ignore demands associated with the in d i v i d u a l r o l e (Meleis & Stevens, 1992; Schnittger & Bird, 1990). H a l l (1992) confirms these findings. Unlike t h e i r wives who put a l l other needs f i r s t , dual-earner husbands indicated that they refuse to "put 32 t h e i r needs l a s t " (Hall, 1992, p. 36). Furthermore, H a l l (1992) reports that mothers who pursue "involvement i n family work to the detriment of t h e i r own needs" experience feelings of exhaustion, resentment, and anger (p. 37). In summary, i t i s clear the parental, paid-work, spousal, and i n d i v i d u a l roles generate intense demands fo r dual-earner parents with young children (McBride, 1990). Demographic findings suggest each r o l e domain has d i f f e r e n t salience for, and exerts d i f f e r e n t r o l e pressures on men and women. Men experience heavy work-role demands, whereas women experience heavy family r o l e demands. However, Voydanoff (1988) i n s i s t s the combination of work and family s t r u c t u r a l c h a r a c t e r i s t i c s "contribute independently and a d d i t i v e l y " to parents' r o l e demands. Thus the stressors i n the work and family domains can e a s i l y accumulate and may r e s u l t i n negative r o l e outcomes (Greenhaus & Parasuraman, 1987) . The Interaction of Role Demands and Attitudes Notwithstanding the considerable demands made on dual-earner parents with young children, inconsistencies s t i l l e x i s t i n the l i t e r a t u r e regarding the r e l a t i o n s h i p of r o l e demands to outcomes (Rosenfield, 1989; Voydanoff, 1988). Knowledge of r o l e demands, i n i s o l a t i o n , does not predict r o l e s t r a i n (Amatea & Fong, 1991; Greenhaus & Parasuraman, 1987; Hibbard & Pope, 1985; Voydanoff, 1988). To predict 33 s t r a i n , researchers must understand the r o l e incumbent's cognitive interpretations of ro l e demands (Bacharach & Bamberger, 1992; Crnic & Booth, 1991; Jacobson, 1989; Verbrugge, 1986). Verbrugge (1986) i n s i s t s r o l e "burdens are not inherent i n the objective a c t i v i t i e s and r e s p o n s i b i l i t i e s adults have ... Instead, burdens reside i n subjective reactions to one's a c t i v i t i e s " (Verbrugge, 1986, p. 74). At present, the inte r a c t i o n between an individual's feelings of control and ac c e p t a b i l i t y of ro l e demands i s thought to determine r o l e s t r a i n (Biener et a l . , 1987; McBride, 1990; P u g l i e s i , 1988; Rosenfield, 1989; Verbrugge, 1986). Control Control i s an in t e g r a l part of ro l e q u a l i t y . According to transactional models of stress, perceptions of overload and u n c o n t r o l l a b i l i t y are the key components of a s t r e s s f u l experience (Reifman et a l . , 1991). I f one's rol e s combine "psychologically demanding tasks with a low l e v e l of control over the tasks, t h i s combination exacts a major t o l l by simultaneously creating arousal and f r u s t r a t i o n " (Barnett et a l . , 1987, p. 131). Studies examining the eff e c t s of ro l e demands on women support t h i s perspective. In Piechowski's (1992) review of the l i t e r a t u r e on multiple r o l e women, the author reports "variables of both r o l e demands and control emerge as 34 s i g n i f i c a n t factors predicting mental health outcome" (p. 137). Rosenfield (1989) concludes that f u l l - t i m e employed women with children experience more anxiety and depression because the high demands of t h e i r combined roles lead to perceptions of low personal control. However, the educational l e v e l of women may a f f e c t t h e i r perceptions of control. K e l l y and Voydanoff (1986) speculate women with some college education experience greater control over work ro l e demands because they have "jobs with r e l a t i v e l y high lev e l s of f l e x i b i l i t y and autonomy" (p. 372). Simi l a r l y , H a l l (1990) suggests "the greater knowledge and s k i l l s associated with better education may increase ... [women's] feelings of competence, mastery, and control" (p. 76). Like women, men experience fewer negative outcomes i n the paid-work and parent roles when they control demands. In an all-male sample, Karasek (1979) reports "the most s t r e s s f u l set of job conditions combines having l i t t l e control over pacing of tasks or the a l l o c a t i o n of resources—and having highly psychologically demanding tasks" (Barnett et a l . , 1987, p. 131). Pleck (1985) i d e n t i f i e s control as an important factor influencing a man's adaptation to fatherhood. Remarking on the p o s i t i v e e f f e c t of c h i l d care on dual-earner fathers, Pleck (1985) writes, " i n the context of the time demands faced by the 35 dual-earner family, a c t i v e l y p a r t i c i p a t i n g i n family work may provide the husband with a sense of control, a f e e l i n g there i s something he can do to respond to these pressures" (p. 116) . Control i s associated with improved adjustment of the dual-earner couple to the paid-work and spousal r o l e s . Parents, as well as married couples who work non-standard hours, experience greater job and family s a t i s f a c t i o n when they control t h e i r work schedules (Burke & McKeen, 1988; Hughes et a l . , 1992; Staines & Pleck, 1986). As the study of multiple roles i s refined, Piechowski (1992) predicts "making d i s t i n c t i o n s between the demands and stresses of roles and the l e v e l of control over those demands may prove important .... Research should be designed to examine the possible i n t e r a c t i v e e f f e c t s of these fac t o r s " (p. 137). A valuable asset to future research with the dual-earner population would be a t o o l that measures respondents* perception of control over some r o l e demands. Acc e p t a b i l i t y The a c c e p t a b i l i t y of ro l e demands also contributes to ro l e q u a l i t y . Many researchers have observed that indivi d u a l s develop subjective predispositions towards r o l e demands (Brett & Yogev, 1988; Chassin et a l . , 1985). Ac c e p t a b i l i t y of a ro l e demand contributes to an indivi d u a l ' s p o s i t i v e evaluation of a ro l e (Campbell & Moen, 36 1992; Cowan & Cowan, 1988). In turn, an ind i v i d u a l ' s evaluation of his/her r o l e a f f e c t s the ro l e outcome (Baruch & Barnett, 1986; Crouter et a l . , 1987; Guelzow et a l . , 1991; Kessler & McRae, 1982; Suitor, 1991). However, the ac c e p t a b i l i t y of r o l e demands i s l i k e l y to vary i n the dual-earner population. Even though the d a i l y r e a l i t i e s of the dual-earner l i f e s t y l e obligate men and women to share family and work ro l e demands, p r e v a i l i n g s o c i e t a l norms and families of o r i g i n may influence how acceptable a demand i s to the ind i v i d u a l (Cowan & Cowan, 1988). This i s due to the personal expression of gender r o l e expectations (Burr, 1976; Woods, 1985). Gender r o l e expectations are "a set of preferences, rewards, tastes, and goals that a person learns because he or she happens to be male or female" (Scanzoni, 1978, p. 6). Gender r o l e expectations prescribe the t r a d i t i o n a l d i v i s i o n of labour; household tasks are women's work while employment-related tasks are men's work (Gunter & Gunter, 1991). Segregation of tasks by gender also operates within the context of housework. Gender appropriate chores for women include meal preparation and cleaning. Men's chores include outdoor tasks, maintaining the car, and paying b i l l s ( B l a i r & Johnson, 1992). T r a d i t i o n a l roles, as determined by gender norms, seem extremely tenacious (Schroeder, Blood, & Maluso, 1992). 37 Although "women have become r e l a t i v e l y l i b e r a t e d with respect to demanding equal r i g h t s at work, ... they f i n d i t harder to re l i n q u i s h deeply held b e l i e f s concerning the woman's proper r o l e i n the home" (Scott & Alwin, 1989, p. 498). An employed wife may request her husband's assistance i n performing "female" tasks but she may do so amid ambivalence over changing her t r a d i t i o n a l r o l e . S i m i l a r l y , men may express discomfort at being asked to do women's work and r e s i s t adopting feminine chores l i k e changing diapers or cleaning the bathroom (Googins & Burden, 1987). Researchers note i n d i v i d u a l differences i n r o l e a c c e p t a b i l i t y among various demographic groups of dual-earners (Smith, 1985). For example, women with advanced educational preparation are l i k e l y to f i n d extensive paid-work r o l e demands more acceptable than women with less extensive formal t r a i n i n g . Theorists o f f e r a cogent explanation for t h i s finding. Education i s p o s i t i v e l y correlated with an androgynous sex r o l e orientation (Kessler & McRae, 1982). Women with androgynous orientations may express them through congruence between t h e i r employment demands and t h e i r conception of appropriate female behaviour (Repetti et a l . , 1989). 38 The Importance of Contextualization in Dual-earner Research In addition to examining the int e r a c t i o n of r o l e demands and attitudes, Johnson (1989) suggests dual-earners be studied as they function i n the work and non-work domains. Several studies document the i n t e r a c t i o n a l e f f e c t s of marital, parental, and occupational c h a r a c t e r i s t i c s on rol e outcomes (Barnett & Marshall, 1992; Barnett, Marshall, & Sayer, 1992; Marshall & Barnett, 1991). Theorists encourage investigators to use a broad eco l o g i c a l perspective that includes the work and family domains i n addition to the interse c t i o n of the two domains (Frone et a l . , 1991; Parasuraman, Greenhaus, & Granrose, 1992). As well, research should study both genders i n both domains. "Most studies of job stress have focused on male samples, while most studies of family stress have focused on female samples" (Hughes et a l . , 1992, p. 32). However, the roles of dual-earner men and women are becoming more undifferentiated and both genders are increasingly being exposed to si m i l a r r o l e demands (Higgins, Duxbury, & Lee, 1994; Spitze, 1988). Researchers should monitor t h i s trend because i t has s i g n i f i c a n t implications for dual-earner men. As r o l e patterns of men and women converge, additional burdens may be v i s i t e d upon men (Googins & Burden, 1987). They may be 39 expected to p a r t i c i p a t e more a c t i v e l y i n c h i l d rearing and housework. Men may f i n d i t taxing to f u l f i l the demands of the family r o l e i n addition to the demands of the work r o l e (Hanson & Bozett, 1987). Pleck (1976) predicts that i f men expand the scope of t h e i r family roles but f a i l to reduce t h e i r commitment to the occupational r o l e they w i l l l i k e l y experience d i f f i c u l t i e s meeting the cumulative r o l e demands. Some studies support Pleck's claim (Crouter et a l . , 1987; Greenberger & O'Neil, 1993; Presser, 1988). Men are increasing t h e i r family work which may be contributing to t h e i r experience of stress. Presser (1988) i d e n t i f i e s a preponderance of father care i n young dual-earner households when mothers work evening and night s h i f t s . Other researchers note fathers i n dual-earner families perform twice as many c h i l d care a c t i v i t i e s alone as t h e i r counterparts i n single-earner families (Crouter et a l . , 1987). Baruch and Barnett (1986) report fathers i n t h e i r study perceive that p a r t i c i p a t i o n i n c h i l d care i n t e r f e r e s with t h e i r careers. Ventura (1987) confirms t h i s observation and states fathers who are engaged in c h i l d care and household tasks experience stress meeting the family's f i n a n c i a l needs. For men p a r t i c i p a t i n g i n Greenberger and O'Neil's (1993) study, absorption i n 40 work-related a c t i v i t i e s was associated with c o n f l i c t among work, parental, and marital r o l e s . In addition to providing a basis for monitoring men's responses to increased family r o l e p a r t i c i p a t i o n , studies which include both spouses permit greater contextualization of the dual-earner experience. Grossman (1988) contends "a convincing study of a parent or of a c h i l d cannot be made i n i s o l a t i o n , but must be examined i n the context of at lea s t the immediate family" (p. 100). Pearl i n (1989) explains context-sensitive studies are important because "one does not act alone as an incumbent of a r o l e . Instead, one r o l e i s part of a larger r o l e set or of a c o n s t e l l a t i o n of complementary roles around which important interpersonal r e l a t i o n s are structured" (p. 242). Meleis and Stevens (1992) caution investigators that unless the q u a l i t y of the experiences i n each r o l e are contextually understood "the development of resources to deal with these experiences [w i l l ] proceed slowly" (p. 24). Instrumentation that i s appropriate for both genders would make context-sensitive studies methodologically f e a s i b l e . As t h i s discussion demonstrates, the prevalence of dual-earner families i n contemporary society makes i t imperative that researchers gain a clearer understanding of the r o l e dynamics characterizing t h i s l i f e s t y l e (Greenhaus, 1988). Although r o l e s t r a i n has been i d e n t i f i e d as a 41 s i g n i f i c a n t problem, researchers have yet to i d e n t i f y the determinants of s t r a i n for dual-earners with young children (Kelly & Voydanoff, 1985). Mounting evidence suggests the int e r a c t i o n of ro l e demands and r e s p o n s i b i l i t y i n addition to the r o l e incumbent's perceptions of control and ac c e p t a b i l i t y may contribute to either r o l e g r a t i f i c a t i o n or ro l e s t r a i n . Theorists urge researchers to assess both genders as they function i n the work and non-work domains, es p e c i a l l y i n the context of the high demands associated with rearing young children. Instrumentation In order to extend dual-earner research more attention must be given to the development of operational d e f i n i t i o n s and empirical measures (Campbell & Moen, 1992). McBride (1990) suggests that researchers "operationalize concepts with f u l l regard for how complicated the phenomena are [and] ... develop instruments that address the problems of ex i s t i n g scales" (p. 382). Many researchers have r e l i e d upon q u a l i t a t i v e evidence of subjective state or unvalidated measures with unknown psychometric properties (Parry & Warr, 1985). Moreover researchers have u t i l i z e d open-ended questions and one- or two-item scales which are associated with r e l i a b i l i t y problems (Greenhaus & Beutell, 1985). Greenhaus (1988) urges researchers to pay close attention to the measures 42 used to assess key concepts. Empirical research i s lim i t e d by the r e l i a b i l i t y and v a l i d i t y associated with the measures. Available instruments are not suitable for determining the influence that r o l e demands and attitudes exert on s t r a i n and g r a t i f i c a t i o n for multiple r o l e incumbents. Some tools predict r o l e s t r a i n , only (Cohen et a l . , 1983; Loerch et a l . , 1989). Other instruments tap into single r o l e domains and are incapable of assessing the combined e f f e c t s of family and paid-work r o l e demands (Abidin, 1980; Burke, 1988; Crnic & Booth, 1991; Johnson, 1989). With the exception of Baruch and Barnett's (1986) rewards and concerns scales, no instrument quantifies men's experiences i n the family r o l e . Unfortunately, the scoring of the rewards and concerns scales does not permit the i d e n t i f i c a t i o n of discrete r o l e antecedents that may lead to s t r a i n or g r a t i f i c a t i o n . The REQ i s an instrument that may not be subject to the li m i t a t i o n s associated with current t o o l s . The REQ i s a 12 6 item questionnaire that objectively measures a multidimensional construct. However, before the REQ can be recommended for use, researchers must have a cle a r understanding of the tool's purpose, target population, variables of interest, format, and psychometric properties (Waltz et a l . , 1984). 43 The REQ was developed i n order to quantify the r o l e q u a l i t y of dual-earner parents with young children. The REQ i s a norm-referenced t o o l that taps into four major rol e s including: paid-work, spouse, parent, and i n d i v i d u a l . H a l l (1993) contends the q u a l i t y of the i n d i v i d u a l ' s enactment of these roles i s a c r i t i c a l factor influencing r o l e outcomes. Role enactment consists of two dimensions, r o l e i n t e n s i t y and r o l e d i s p a r i t y . Role i n t e n s i t y i s "defined as the amount of time (preemptiveness) and e f f o r t (organismic involvement) devoted to s p e c i f i c r o l e components ... [This dimension consists] of personal perceptions of the amount of time, e f f o r t , and r e s p o n s i b i l i t y ... associated with the four major r o l e s " (Hall, 1993, p. 59). Role d i s p a r i t y i s defined as: discrepancy between expectations, e x i s t i n g attitudes, and actual behaviours. I t captures the i n d i v i d u a l s ' perceptions of t h e i r a b i l i t i e s to control t h e i r r o l e i n t e n s i t i e s as paid worker, i n d i v i d u a l , and spouse and t h e i r attitudes about the l e v e l of i n t e n s i t y they experience i n the four roles (Hall, 1993, p. 59). 44 Description of the Role Enactment Questionnaire The REQ i s divided into four sections which examine the paid-work, spouse, parent, and i n d i v i d u a l r o l e s . The questionnaire includes a t o t a l of 42 statements about r o l e demands. Each statement i s comprised of three items. The f i r s t item describes a r o l e demand and measures the respondent's time or energy expenditure i n meeting t h i s demand. The second item assesses the respondent's perception of the r o l e demand's ac c e p t a b i l i t y . The t h i r d item captures either the degree of r e s p o n s i b i l i t y or control the respondent associates with the demand. Responses are quantified using a f i v e step L i k e r t scale. For example: None A great deal The amount you dress your children 1 2 3 4 5 A. How acceptable i s t h i s to you? 1 2 3 4 5 B. How much r e s p o n s i b i l i t y do you 1 2 3 4 5 take for this? (Hall, 1993) Psychometric Properties of the REQ Wilson (1987) i n s i s t s before an instrument can be used i n research, minimal lev e l s of r e l i a b i l i t y and v a l i d i t y must be established. The r e l i a b i l i t y of measurement "refers to the consistency, accuracy, and pr e c i s i o n of the measures taken" (Wilson, 1987, p. 192). Waltz et a l . (1984) recommend estimating the r e l i a b i l i t y of the norm-referenced instrument by determining i t s i n t e r n a l consistency and 45 t e s t - r e t e s t r e l i a b i l i t y . R e l i a b i l i t y i s a necessary but not a s u f f i c i e n t condition for v a l i d i t y (Waltz et a l . , 1984). To be v a l i d an instrument must r e l i a b l y measure "what i t i s supposed to measure" (Wilson, 1987, p. 184). An instrument should demonstrate content v a l i d i t y which " i s the extent to which the items included on a t e s t are a representative sample of the important and relevant elements of a w e l l - s p e c i f i e d content or behavioral universe" (Brown, 1970, p. 155). To date, H a l l has presented the REQ to two groups. The author p i l o t tested the instrument on a ten couple sample. The volunteers were asked to indicate the length of time i t took to complete the REQ. They also evaluated the instrument's format and c l a r i t y of instructions. "After p i l o t t e s t i n g , revisions were made to correct any items considered ambiguous, ir r e l e v a n t , or unacceptable" (Hall, 1993, p. 61). R e l i a b i l i t y and v a l i d i t y t e s t i n g were then conducted using a 55 couple sample (Hall, 1993). R e l i a b i l i t y ^ From the 55 couple convenience sample, H a l l (1993) gathered data to t e s t the REQ 1s r e l i a b i l i t y . Obtaining a 71% response rate, H a l l (1993) calculated i n t e r n a l consistency and t e s t - r e t e s t r e l i a b i l i t y for the two dimensions of the REQ and the four r o l e categories. Using Cronbach's alpha, H a l l (1993) calculated high i n t e r n a l consistencies on the two r o l e dimensions—0.89 for i n t e n s i t y 46 and 0.90 for d i s p a r i t y . However, for the r o l e categories, alpha c o e f f i c i e n t s ranged from 0.39 to 0.93. For t e s t - r e t e s t r e l i a b i l i t y H a l l (1993) determined the Pearson's r at 0.73 for the d i s p a r i t y dimension and 0.80 for the i n t e n s i t y dimension. Pearson's r ranged from 0.62 to 0.87 for each of the four r o l e categories. V a l i d i t y . The process of evaluating content v a l i d i t y i s " r a t i o n a l and judgmental, and attempts to rate the adequacy of sampling" (Brown, 1970, p. 155). Since "content v a l i d i t y i s largely a function of how an instrument i s developed," researchers can i n f e r the REQ's content v a l i d i t y from the systematic method H a l l used to develop the t o o l (Waltz et a l . , 1984, p. 142). To derive the i n t e n s i t y and d i s p a r i t y dimensions H a l l considered c l a s s i c a l r o l e t h e o r i s t s (Burr, 1976; Sarbin & Allen, 1968) i n addition to Ward's (1986) model of ro l e s t r a i n . Furthermore, the REQ's item pool was generated from parents' descriptions of relevant r o l e components, obtained by H a l l (1987, 1991) from q u a l i t a t i v e studies investigating the dual-earner experience. F i n a l l y , H a l l supplemented these findings with information acquired from an extensive review of the l i t e r a t u r e . H a l l also supports content v a l i d i t y by respecting the p r i n c i p l e of representative sampling which requires items to be chosen from the universe of possible content " i n due 47 proportion or frequency" (Brown, 1970, p. 136). H a l l (1993) reports that "parents with preschool children emphasized parenting and spousal roles to a greater extent than work and i n d i v i d u a l r o l e s " (p. 60). Therefore, H a l l chose a greater number of items to measure the parent and spouse r o l e s . A review of the instrument by experts i n family nursing and work and family research provides additional support for the content v a l i d i t y of the REQ ( P o l i t & Hungler, 1991). The experts were asked to assess whether the items on the instrument s a t i s f a c t o r i l y represent the behaviours i n the s p e c i f i e d domains. In accordance with the experts' recommendations, H a l l (1993) revised the item pool. Critique of the Role Enactment Questionnaire The REQ i s a newly developed instrument that shows great promise for measuring the antecedents of r o l e s t r a i n in dual-earners with young children. The l i t e r a t u r e supports the notion that the construct, r o l e enactment, plays a s i g n i f i c a n t part i n the r o l e s t r a i n process. P i l o t t e s t i n g of the REQ suggests that the t o o l i s objective, e f f i c i e n t , and acceptable to research subjects (Hall, 1993). High i n t e r n a l consistency and t e s t - r e t e s t r e l i a b i l i t y estimates were obtained i n a convenience sample of 55 couples (Hall, 1993). H a l l (1993) also demonstrates substantial content v a l i d i t y for the instrument. 48 Additional psychometric t e s t i n g i s required. The t e s t -r e t e s t estimates may be suspect. H a l l reports a delay of one month before respondents returned the second set of questionnaires. Thus, the second questionnaire "may have been completed from 3 to 6 weeks af t e r the f i r s t one" (Hall, 1993, p. 62). Wilson (1987) suggests that i t i s important to consider the time i n t e r v a l s between the f i r s t and second administration of questionnaires. Ideally the time i n t e r v a l should be "long enough to avoid carryover e f f e c t s and short enough to avoid changes i n the construct being measured" (Wilson, 1987, p. 193). Since the s t a b i l i t y of the construct i s unknown, variations i n the time i n t e r v a l s among subjects may have confounded the t e s t - r e t e s t c a l c u l a t i o n s . Hall's small sample size may account for i n t e r n a l consistency measurements that vary widely for the four r o l e s . H a l l (1993) contends that "because alpha i s dependent on the t o t a l t e s t variance and the length of the t e s t , i t i s not unusual to have lower alphas with a short t e s t and a small sample s i z e " (p. 62). However, i t would be i n s t r u c t i v e to i d e n t i f y roles which demonstrate unacceptably low alphas. The REQ may require r e v i s i o n i n order to ensure that a l l the items within the instrument are consistent with each other. More comprehensive t e s t i n g of the REQ's v a l i d i t y must be conducted before the t o o l can be recommended for use. 49 Wilson (1987) notes that "content v a l i d i t y , although necessary, i s not a s u f f i c i e n t i n d i c a t i o n that the instrument measures what i t i s intended to measure, because i t i s based on subjective judgment" (p. 194). In order to demonstrate the v a l i d i t y of the REQ, an examination of the tool ' s construct v a l i d i t y must be undertaken. Methods of Supporting Construct V a l i d i t y Crocker and Algina (1986) state that construct v a l i d i t y may be supported by establishing c o r r e l a t i o n a l evidence of a r e l a t i o n s h i p between a new instrument and a c r i t e r i o n . The c r i t e r i o n may be a mature instrument which i s known to accurately measure a s i m i l a r construct. However, c r i t e r i o n - r e l a t e d v a l i d a t i o n i s unsuitable for t h i s study because no instrument measures a construct s i m i l a r to r o l e enactment. An a l t e r n a t i v e method recommended by Crocker and Algina (1986) i s hypothesis t e s t i n g . The authors suggest researchers (a) derive from an e x p l i c i t l y stated theory hypotheses about how those who d i f f e r on the construct are expected to d i f f e r on demographic c h a r a c t e r i s t i c s , (b) gather empirical data to t e s t the hypothesized rel a t i o n s h i p s , and (c) determine i f the data are consistent with the hypotheses. Factor analysis (FA) i s a s t a t i s t i c a l technique commonly used to est a b l i s h construct v a l i d i t y (Brown, 1970). 50 FA "aims to summarize the i n t e r r e l a t i o n s h i p s among variables i n a concise but accurate manner" (Gorsuch, 1974, p. 2). The goal i s to " i d e n t i f y those variables which are related enough to be placed under the same l a b e l " (Gorsuch, 1974, p. 7) • Two types of FA are mentioned i n the l i t e r a t u r e : exploratory and confirmatory. Confirmatory factor analysis (CFA) i s widely acknowledged as the more sophisticated technique (Tabachnick & F i d e l l , 1989). Furthermore, CFA i s superior to exploratory factor analysis because i t allows the researcher to specify the structure of the r e l a t i o n s i n the model on substantive grounds. Another advantage of CFA i s that i t can help to resolve the problem of measurement error, because i t indicates the degree to which multiple indicators of a concept r e f l e c t the underlying concept; the degree to which they r e f l e c t anything else i s considered measurement error (Coverman, 1989, p. 972). As the preceding discussion on instrumentation attests, there i s a need i n dual-earner research for r e l i a b l e and v a l i d tools that measure s a l i e n t r o l e c h a r a c t e r i s t i c s . The REQ i s a newly developed t o o l that quantifies two dimensions of r o l e enactment: r o l e i n t e n s i t y and r o l e d i s p a r i t y . Developed from two q u a l i t a t i v e studies that investigated the experiences of dual-earners with young 51 children (Hall, 1987; H a l l , 1991), the REQ demonstrates a high degree of content v a l i d i t y . Furthermore, psychometric t e s t i n g suggests the instrument i s i n t e r n a l l y consistent and has good t e s t - r e t e s t r e l i a b i l i t y . However, methodological l i m i t a t i o n s i n Hall's (1993) study design indicate additional r e l i a b i l i t y and v a l i d i t y t e s t i n g i s advisable. Conclusion Concerns about the well-being of dual-earners with preschool children make the investigation of t h i s population a research imperative. Recently, th e o r i s t s revised an e a r l i e r assumption that r o l e accumulation leads to either negative or p o s i t i v e outcomes. Instead, i t appears that r o l e q u a l i t y has the pote n t i a l to create r o l e s t r a i n or r o l e g r a t i f i c a t i o n , depending on mediating influences. The in t e r a c t i o n of r o l e demands and the r o l e incumbent's attitudes may determine which r o l e outcome occurs. The development of r e l i a b l e and v a l i d instrumentation has lagged behind other t h e o r e t i c a l advances. With the exception of the REQ, available tools are not designed to measure r o l e c h a r a c t e r i s t i c s thought to impact on s t r a i n and g r a t i f i c a t i o n for both genders at the work/family nexus. However, further psychometric t e s t i n g must be conducted on the REQ before t h i s t o o l can be used i n c l i n i c a l or research settings. 52 Chapter Summary Chapter two presented a c r i t i c a l review of the l i t e r a t u r e . Shortcomings of extant studies were described. The c r i t i q u e highlighted the relevance of the REQ for researchers investigating r o l e q u a l i t y . However, i t was demonstrated that the instrument i s r e l a t i v e l y untried, psychometrically, and psychometric standards are not negotiable. In chapter three, the methods used to examine the r e l i a b i l i t y and v a l i d i t y of the REQ are detailed. The chapter describes the recruitment strategies and study procedures. Lastly, a b r i e f examination of the e t h i c a l considerations and li m i t a t i o n s related to the study are presented. 53 CHAPTER THREE Method In the following section the sample sel e c t i o n , recruitment strategies, and methods for t e s t i n g the r e l i a b i l i t y and v a l i d i t y of the REQ are described. Sample Nunnally (1978) suggests researchers use a large sample to psychometrically t e s t new instruments. "A bare minimum in item analysis i s f i v e persons for each item ... [although] a safer number i s ten persons per item" (p. 298). Since the REQ consists of 12 6 items, the sample should include between 630 and 1260 ind i v i d u a l s . However, the recruitment of such a large sample i s not f e a s i b l e , given that t h i s project i s a Magistral thesis and the intent i s to gain an understanding of the research process. Thus, although the l i m i t a t i o n s of the sample siz e are recognized, the convenience sample only consists of 165 i n d i v i d u a l s . Sample Recruitment In order to r e c r u i t subjects, advertisements were placed i n the newsletters of the following i n s t i t u t i o n s : BC Hydro, VanCity Credit Union, the University of B r i t i s h Columbia, and the Vancouver Hospital and Health Sciences Centre. E-mail messages were transmitted to a l l BC Tel 54 employees who used o f f i c e computers. B u l l e t i n s requesting research subjects were mass-mailed to 650 secretaries and 320 managers employed by the University of B r i t i s h Columbia. Posters describing the study were displayed i n community centres, day care centres, and churches i n the Vancouver area. Dual-earner parents of young children known to the investigator were inv i t e d to p a r t i c i p a t e . Dual-earners were requested to contact the investigator. When contacted, the investigator screened the in d i v i d u a l , and i f e l i g i b l e , the in d i v i d u a l and his/her spouse were enrolled i n the study. A cover l e t t e r (Appendix B) and the f i r s t copy of the REQ were then sent to each spouse. Sample C r i t e r i a To be e l i g i b l e for p a r t i c i p a t i o n i n the study, prospective respondents were screened for the following i n c l u s i o n and exclusion c r i t e r i a . Inclusion C r i t e r i a 1. Couples have l i v e d together for at lea s t one year. 2. Men and women are over 19 years of age. 3. Study participants understand written English. 4. Families may be blended or t r a d i t i o n a l but must include at least one c h i l d under the age of s i x . 55 5. Both partners are employed and each i s employed at least 20 hours a week. 6. Individuals have been employed i n the same job for over three months. Exclusion C r i t e r i a 1. Chronic i l l n e s s of either spouse or c h i l d ( r e n ) . Chronic i l l n e s s i s defined as any physical or psychological condition preventing any family member from attending day care, school, or work more than f i v e days between July 1993 and November 1994. 2. Couples and t h e i r child(ren) who l i v e with extended family. Study Procedures The following section describes the procedures used to evaluate the r e l i a b i l i t y and v a l i d i t y of the REQ. R e l i a b i l i t y Testing In order to c o l l e c t data which would permit assessments of the REQ's in t e r n a l consistency and t e s t - r e t e s t r e l i a b i l i t y , study respondents were mailed two sets of questionnaires. The following procedure was followed: 1. F i r s t mailing: Copy #1 of the REQ and the demographic data c o l l e c t i o n sheet were mailed to, and completed by, the dual-earner couple. 56 2. Second mailing: Two weeks a f t e r the co-investigator received copy #1, copy #2 of the REQ was mailed to, and completed by, the couple. 3. Respondents were a l l o t t e d two weeks to respond to each mailing. Couples who did not return a completed questionnaire were contacted by telephone and, when necessary, mailed a duplicate copy of the questionnaire. 4. Respondents who f a i l e d to return questionnaires were contacted by telephone every two weeks. V a l i d i t y Testing To support construct v a l i d i t y , hypothesis t e s t i n g and confirmatory factor analysis (CFA) were performed. Hypothesis Testing In order to generate and t e s t the hypotheses that examine the v a l i d i t y of r o l e i n t e n s i t y and r o l e d i s p a r i t y , Hall's (1993) modification of Ward's (1986) t h e o r e t i c a l framework was used. From the empirical l i t e r a t u r e , demographic groups were i d e n t i f i e d which were most l i k e l y to experience high or low scores on the i n t e n s i t y and d i s p a r i t y dimensions. The demographic data c o l l e c t e d were used to confirm or r e j e c t study hypotheses. If the REQ accurately measured the construct described by H a l l , then the scores of 57 dual-earners may be predicted from t h e i r demographic c h a r a c t e r i s t i c s . Confirmatory Factor Analysis The purpose of factor analysis i s to i d e n t i f y groupings of variables that correlate with the same factor. Ideally, the correlations between the variables and the factor are high while the correlations between these variables and the al t e r n a t i v e factor(s) are low (Nunnally, 1978). In CFA the dominant factors are hypothesized p r i o r to analysis. For the present study, three analyses of the t e s t items were performed. The goal of the f i r s t analysis was to extract two factors corresponding to r o l e i n t e n s i t y and r o l e d i s p a r i t y . A second analysis was performed to confirm the existence of i n t e n s i t y i n each of the four r o l e components. The t h i r d analysis sought to confirm the existence of d i s p a r i t y i n each of the four r o l e components. Evidence that these factors exist, as hypothesized, lends support for the construct v a l i d i t y of the REQ. Data Analysis 1. To determine i n t e r n a l consistency, only data from the f i r s t administration of the REQ were analyzed. Cronbach's alpha was calculated for each r o l e component and 58 for i n t e n s i t y and d i s p a r i t y i n each of the four r o l e components. 2. To determine t e s t - r e t e s t r e l i a b i l i t y , scores from the f i r s t administration of the REQ were correlated with scores from the second administration. Pearson's r was used to calculate t e s t - r e t e s t r e l i a b i l i t y for each r o l e component and for i n t e n s i t y and d i s p a r i t y i n each of the four r o l e components. 3. Hypothesis t e s t i n g was performed using either a t - t e s t or a one-way Analysis of Variance (ANOVA). The sig n i f i c a n c e l e v e l was set at .05. 4. To determine the construct v a l i d i t y of the REQ, the te s t items were subjected to CFA using the P r i n c i p a l Components method of analysis, with varimax r o t a t i o n and Kaiser normalization. Assumptions 1. Individuals are capable of evaluating t h e i r d a i l y r o l e a c t i v i t i e s . 2. Individuals are s e l f - r e f l e c t i v e and capable of describing t h e i r f e e l i n g s . 3. Individuals vary i n t h e i r performance of, and attitudes towards, r o l e a c t i v i t i e s . 59 4. Individuals can communicate t h e i r experiences to others. Limitations 1. The study r e l i e d on a convenience sample. This introduces possible s e l e c t i o n bias which may l i m i t the g e n e r a l i z a b i l i t y of findings. 2. The sample i s biased i n favour of the w e l l -educated. Caution must be exercised i n generalizing the findings of t h i s study to s u b s t a n t i a l l y d i f f e r e n t populations. 3. The small sample siz e and r e l a t i v e homogeneity of sample c h a r a c t e r i s t i c s l i m i t s the variance of scores thereby r e s u l t i n g i n lower r e l i a b i l i t y estimates. 4. Due to the small sample siz e , construct v a l i d i t y work i n t h i s study must be considered preliminary and tenuous. 5. The demographic subsamples are very small. The reader i s cautioned not to place undue confidence i n s i g n i f i c a n t findings based on small subsamples. 6. R e l i a b i l i t y and v a l i d i t y measures are s i t u a t i o n s p e c i f i c . G e n e r a l i z a b i l i t y of these c a l c u l a t i o n s i s l i m i t e d to the study sample. 60 7. Repeated exposure to the REQ may lead to d i f f e r e n t degrees of memory r e c a l l among respondents. Individuals who remember questionnaire items may duplicate t h e i r e a r l i e r responses. P o l i t and Hungler (1991) suggest that t h i s phenomenon may r e s u l t i n spuriously high t e s t - r e t e s t r e l i a b i l i t y c o e f f i c i e n t s . 8. Thirty-three percent of respondents delayed returning the second questionnaire for one to three months. These respondents may have completed the second questionnaire anywhere from 3 to 14 weeks a f t e r the f i r s t questionnaire. Changes may have occurred i n the nature of the construct over such a protracted period of time. Thus, measurement may r e f l e c t construct s t a b i l i t y rather than instrument s t a b i l i t y . E t h i c a l Considerations A cover l e t t e r was sent to po t e n t i a l subjects informing them of the purpose of the study (Appendix B). Completion of the REQ served as an ind i c a t i o n of informed consent. The study protocol, measures, and contact l e t t e r s were approved by the UBC Behavioral Screening Committee for Research and Other Studies Involving Human Subjects. To protect respondents' c o n f i d e n t i a l i t y , questionnaires were numbered. Only the investigator had access to the p a r t i c i p a n t s ' names. 61 Benefits of P a r t i c i p a t i o n 1. Increased understanding of oneself i n r e l a t i o n to r o l e a c t i v i t i e s . 2. The opportunity to express personal views regarding r o l e a c t i v i t i e s . 3. The opportunity to p a r t i c i p a t e i n a worthwhile project that may help others adapt to the dual-earner l i f e s t y l e . Risks from P a r t i c i p a t i o n 1. Respondents may gain insights into r o l e - r e l a t e d a c t i v i t i e s . This may a l t e r previous perceptions and cause psychological d i s t r e s s . 2. Completing the questionnaire may take respondents' time away from other a c t i v i t i e s and increase respondents' perception of time pressure. Conclusion Research into the dual-earner l i f e s t y l e i s hampered by a lack of instruments that measure the antecedents of r o l e s t r a i n . Preliminary t e s t i n g of the REQ suggests that t h i s questionnaire has promise and represents a s i g n i f i c a n t improvement over currently available t o o l s . The procedures followed to address psychometric r e l i a b i l i t y and v a l i d i t y 62 contribute to the understanding and value of t h i s instrument. Chapter Summary In chapter three, the methods used to evaluate the psychometric properties of the REQ were described. The chapter presented recruitment strategies and sampling c r i t e r i a used by the study investigator. Procedures used to te s t the REQ's r e l i a b i l i t y and v a l i d i t y were examined. Data analysis techniques were i d e n t i f i e d . F i n a l l y , e t h i c a l considerations, assumptions, and l i m i t a t i o n s of the study were discussed. Chapter four presents the r e s u l t s of data analysis. Findings from the psychometric t e s t i n g of the REQ are described and discussed i n the context of the l i t e r a t u r e . 63 CHAPTER FOUR Presentation and Discussion of Findings The following chapter summarizes the r e s u l t s from the r e l i a b i l i t y and v a l i d i t y t e s t i n g of the REQ. The chapter describes: (a) the sample c h a r a c t e r i s t i c s , (b) support f o r r e l i a b i l i t y , s p e c i f i c a l l y i n t e r n a l consistency and te s t - r e t e s t r e l i a b i l i t y and, (c) support for construct v a l i d i t y using hypothesis t e s t i n g and factor analysis. In each section, the findings are described and discussed i n the context of the l i t e r a t u r e . Sample Characteristics A convenience sample of 104 dual-earner couples was recruited. Of the 2 08 parents enrolled, 165 completed and returned the f i r s t questionnaire, y i e l d i n g a response rate of 79%. However, only 134 individuals completed and returned both the f i r s t and second questionnaires. Chi-square analysis, (2, N = 165) = 9.78, p = .007, revealed that the two groups were comparable demographically, d i f f e r i n g only on the number of children i n the home; partici p a n t s with fewer children had a s i g n i f i c a n t l y greater tendency to complete both questionnaires. Due to the success of a campus mass-mailing recruitment strategy, the majority of dual-earners enrolled were 64 employed by the University of B r i t i s h Columbia. A l l par t i c i p a n t s were rearing at least one c h i l d under the age of s i x . Most of the children were under the age of three. Almost one-half of the sample had at least one c h i l d 18 months or younger (see Table 1). Table 1 Number and ages of children (N = 165) Number of Ages of Children Children 6 yrs 3 yrs 18 mo. 1 2 3 & less & less & less Percent 49% 46% 5% 100% 85% 48% Number 80 76 9 165 140 80 The sample r e f l e c t s the trend for mothers to return to work before t h e i r children reach school-age (Googins & Burden, 1987; Schroeder et a l . , 1992; Walker & Best, 1991). Since 1970, the greatest increase i n labour force p a r t i c i p a t i o n has occurred for mothers with preschool children (Mathews & Rodin, 1989). Interestingly, a "husband's po t e n t i a l to become a good income earner does not 65 make i t more l i k e l y for mothers of young children to be out of the labour force" (Matthews & Rodin, 1989, p. 1391). The majority of the men and women were well-educated. Seventy-three percent had either obtained a diploma, f i n i s h e d college, or earned a uni v e r s i t y degree. Twenty-five percent of these part i c i p a n t s were prepared at the masters or doctoral l e v e l . Only 1% had not graduated from high school. Most of the dual-earners were quite mature (M = 34.9, SD = 5.2). Only 14% of the participants were under 30 which r e f l e c t s the trend towards delayed childbearing (Bielby, 1992). Higgins et a l . (1994) suggested women delay c h i l d b i r t h i n order to estab l i s h t h e i r careers. Ghalam (1993) reported that the mean age when managerial and professional women i n Canada have t h e i r f i r s t c h i l d i s 31. Consistent with t h e i r education and age, the majority of dual-earners were f i n a n c i a l l y successful. Chi-square analysis showed that men earned more money than women (2, N = 160) = 22.55, p_<.001. For personal income, 25% of the part i c i p a n t s reported earning between $40,000 and $49,999 a year, while 28% made $50,000 or more a year. Only 4% of the sample earned less than $2 0,000 a year. Study participants worked from 18 to 80 hours a week, with a mean of 3 6.7 (SD = 9.2) paid-work hours. Chi-square analysis demonstrated a s i g n i f i c a n t difference i n the number 66 of paid hours men and women worked (1, N = 165) = 22.52, E<.001. Over one-third of the women worked t h i r t y hours a week or less, while 94% of the men worked more than 30 hours a week (see Table 2). Table 2 Dual-earners' hours of work (N = 165) Work hours Gender Men Women N % N % Under 2 0 hrs 0 0 3 4% 2 0 to 30 hrs 5 6% 27 32% Over 30 hrs 77 94% 53 64% Ghalam (1993) reported that i n 1991, 26% of a l l employed Canadian women worked part-time, compared with only 9% of employed men. Forty percent of women aged 25 to 44 prefer part-time work and more than half of these women c i t e personal or family r e s p o n s i b i l i t i e s as t h e i r reason for working part-time. Flex-time was assessed for respondents. Burke and McKeen (1988) defined flex-time as employment which allows 67 employees a "certain freedom i n choosing t h e i r times of a r r i v a l and departure" (p. 34). Chi-square analysis revealed that the a v a i l a b i l i t y of flex-time was independent of gender (1, N = 163) = 2.15, p = .34. For 50% of dual-earners, flex-time was available; however, 48% reported not being able to access flex-time, and p a r t i a l flex-time was available to the remaining 2% of the sample. About one-third (34%) of sample parti c i p a n t s did not have help at home. As demonstrated by chi-square analysis dual-earners who worked over 3 0 hours a week were no more l i k e l y to have help than dual-earners who worked under 3 0 hours a week (1, N = 163) = 2.33, p_ = .12. For the 66% of dual-earners who reported having assistance, the most frequently c i t e d sources of help were nannies (39%) and family babysitters (35%). Thus, c h i l d care was the most prevalent form of home assistance. Ghalam (1993) i d e n t i f i e d babysitters, rather than nannies, as the most important source of c h i l d care assistance for Canadian dual-earners. Babysitters provide care for 37% of children under three and 31% of children aged three to f i v e . Parents also r e l y on r e l a t i v e s for c h i l d care; 24% of children under three and 16% aged three to f i v e are cared for by a r e l a t i v e . In the majority of cases, the grandparent i s the r e l a t i v e giving care (Crompton, 1991). 68 Googins (1991) reported a small percentage of dual-earner parents e n l i s t i n g help. In hi s sample, only 10% of hourly workers and 12% of managers employed others to a s s i s t with home chores. In a sample of professional women, McLaughlin et a l . (1988) observed that 4 5% of respondents used outside help. Moreover, 67% of the women requested help from a family member on a d a i l y basis. R e l i a b i l i t y The next section presents the REQ's in t e r n a l consistency and t e s t - r e t e s t r e l i a b i l i t y estimates. The findings are compared with r e l i a b i l i t y estimates obtained by H a l l (1993). Internal Consistency Internal consistency measures the homogeneity of the REQ, the extent to which items measure the same c h a r a c t e r i s t i c . "Unless t e s t scores are consistent, they cannot be related to other variables with any degree of confidence" (Brown, 1970, p. 76). Using Cronbach's alpha, i n t e r n a l consistency was calculated for the REQ. The d i s p a r i t y dimension demonstrated high r e l i a b i l i t y with an alpha of .91 for the ov e r a l l scale (see Table 3). Alphas were also obtained for d i s p a r i t y i n each of the four r o l e components and ranged from .78 to .91. 69 Table 3. Internal consistency: Comparison of findings for r o l e dimensions Dimension Number of items Sleigh (n=165) H a l l (n=110) 1. Intensity: Paid-work Spouse Parent Individual 7 26 28 6 33 75 92 ,53 39 78 93 52 Total 2. Disparity: Paid-work Spouse Parent Individual Total 67 14 19 14 12 59 88 77 84 86 85 ,91 90 76 83 89 ,82 ,90 Internal consistency also proved to be substantial for the i n t e n s i t y dimension. The alpha for the enti r e scale was .88 (see Table 3). However, alphas were low for the 70 paid-work r o l e subscale (.33) and modest for the i n d i v i d u a l r o l e subscale (.53). Both alphas were less than .70, the minimal l e v e l of r e l i a b i l i t y recommended by Nunnally (1978) for exploratory research. Table 4 . Internal consistency: Comparison of findings for r o l e components Cronbach's alpha Number of Sleigh H a l l Role items (n=165) (n=110) Paid-work 21 . 66 . 69 Spouse 45 .82 .85 Parent 42 .93 .94 Individual 18 .86 .83 Comparison of t h i s author's sample and H a l l ' s (1993) sample indicates s i m i l a r r e l i a b i l i t y measurements for r o l e dimensions and components (see Tables 3 & 4). In general, the REQ demonstrates excellent i n t e r n a l consistency. However, the persistent f a i l u r e of the i n d i v i d u a l and paid-work i n t e n s i t y subscales to achieve minimal l e v e l s of r e l i a b i l i t y warrants further scrutiny. 71 Low i n t e r n a l consistency may be due to a lack of v a r i a b i l i t y i n the scores. For the i n d i v i d u a l r o l e items, standard deviations ranged from .86 to 1.14. Standard deviations for the paid-work r o l e ranged from .86 to 1.15. Limited score dispersion may be due to the homogeneous sample tested. Nunnally (1978) emphasized the importance of score dispersion to r e l i a b i l i t y . "The size of the r e l i a b i l i t y c o e f f i c i e n t i s d i r e c t l y related to the standard deviation of obtained scores for any sample of subjects" (p. 241). Furthermore, "the r e l i a b i l i t y c o e f f i c i e n t w i l l be larger for samples of subjects that vary more with respect to the t r a i t being investigated" (Nunnally, 1978, p. 241). Clear item wording and e x p l i c i t instructions are elements that impact on r e l i a b i l i t y (Nunnally, 1978). Measurement error i s introduced when respondents do not understand the wording of items. Respondents i n t h i s author's sample were confused by the phrase, "the amount that you ..." They suggested i n t h e i r written comments that "amount" should be defined, for example, amount of time, amount of energy, or frequency of a c t i v i t y . In addition, respondents received no instructions on how to deal with non-applicable items. Dual-earners chose a varie t y of ways to score non-applicable items. Responses which do not r e f l e c t v a l i d perceptions introduce unnecessary 72 error into the scores. Inconsistency i n answering these items may have compromised r e l i a b i l i t y . In the author's sample, the format of the questionnaire confused respondents. On the printed forms, no spaces were inserted between items 13, 14, and 15. The items appeared as; A great None deal 13. The amount you do dishes. 1 2 3 4 5 (a) How acceptable i s t h i s to you? 1 2 3 4 5 (b) How much r e s p o n s i b i l i t y do you 1 2 3 4 5 take for this? 14. The amount you vacuum/sweep/mop. 1 2 3 4 5 (a) How acceptable i s t h i s to you? 1 2 3 4 5 (b) How much r e s p o n s i b i l i t y do you 1 2 3 4 5 take for this? 15. The amount you do laundry. 1 2 3 4 5 (a) How acceptable i s t h i s to you? 1 2 3 4 5 (b) How much r e s p o n s i b i l i t y do you take for this? 1 2 3 4 5 Respondents found i t d i f f i c u l t to track statements across the page to the correct l i n e of numerical responses. It i s quite l i k e l y that t h i s v i s u a l l y t r y i n g format led to a c e r t a i n degree of fatigue, impatience, and carelessness i n respondents. Substandard alphas also occur when scales are short. "The si z e of the r e l i a b i l i t y c o e f f i c i e n t i s based on both the average c o r r e l a t i o n among items (internal consistency) and the number of items" (Nunnally, 1978, p. 230). For in t e n s i t y , both the i n d i v i d u a l and paid-work r o l e subscales 73 are very short, containing only s i x and seven items, respectively. The brevity of the subscales probably had a s i g n i f i c a n t impact on r e l i a b i l i t y estimates. As a r o l e component the i n d i v i d u a l r o l e subscale consists of 18 items, while the paid-work r o l e subscale contains 21 items. Cronbach's alphas for the i n d i v i d u a l and paid-work subscales for the r o l e components (see Table 4) were much higher than those obtained for these subscales i n the i n t e n s i t y dimension (see Table 3). Alphas for the i n d i v i d u a l and paid-work subscales for the r o l e components ranged from .66 to .69. and .86 to .83, respectively. Test-retest R e l i a b i l i t y Just as i n t e r n a l consistency measures the consistency of performance over items, t e s t - r e t e s t r e l i a b i l i t y measures the consistency or s t a b i l i t y of performance over time. Researchers calculate the s t a b i l i t y of an instrument by t e s t i n g the i n d i v i d u a l twice and c o r r e l a t i n g the two sets of scores. For the REQ, i n t e n s i t y and d i s p a r i t y produced Pearson's r ranging from .69 to .80 for the four r o l e components (see Table 5). With a c o e f f i c i e n t of .80, the i n t e n s i t y dimension demonstrated minimally greater s t a b i l i t y than the d i s p a r i t y dimension (r = .76). 74 Table 5 Test-retest r e l i a b i l i t y : Comparison of findings for r o l e dimensions Pearson 1 s r Sleigh H a l l Role Intensity Disparity Intensity Disparity Paid-work . 69 .73 .79 . 62 Spouse .73 .72 .77 . 68 Parent .82 . 64 . 87 . 68 Individual .70 . 64 . 62 . 65 Overall .80 .76 .80 .73 Test-retest r e l i a b i l i t y was calculated on the four r o l e components. Pearson's r ranged from .69 to .74 for each of the r o l e s (see Table 6). For the t o t a l r o l e scale, Pearson's r measured .72. 75 Table 6 Test-retest r e l i a b i l i t y : Comoarison of findings for r o l e components Pearson's r Role Sleigh H a l l Paid-work .73 . 60 Spouse . 69 .71 Parent . 74 . 85 Individual .72 . 66 Compared to H a l l (1993), s t a b i l i t y was s l i g h t l y lower for t h i s author's sample (see Tables 5 & 6). H a l l (1993) obtained Pearson's r ranging from .62 to .87 for the r o l e components. However, H a l l (1993) reported a possible lag time of three to six weeks between the f i r s t and second administration of the questionnaire. In the author's sample, study respondents were even more tardy. For 33% of the individuals, the possible i n t e r v a l between questionnaire administrations ranged from three to 14 weeks. "In general ... the magnitude of the c o r r e l a t i o n decreases over time" (Brown, 1970, p. 62). 76 S i g n i f i c a n t changes may have occurred i n the respondents' r o l e experiences over such a protracted period of time. In order to determine the acceptable l e v e l of r e l i a b i l i t y for a new instrument, Brown (1970) suggested comparing the r e l i a b i l i t y c o e f f i c i e n t of the new t o o l with an established t o o l that measures the same construct. Although Hal l ' s instrument operationalizes a novel construct, r e l i a b i l i t y may be compared with Barnett and Baruch's (1986) rewards and concerns scales. Barnett et a l . (1992) measured the t e s t - r e t e s t r e l i a b i l i t y of t h e i r instruments using randomly sampled scores from 10% of women (n = 40) who par t i c i p a t e d i n a longitudinal study. The authors obtained Pearson's r .88 for the job rewards and concerns scales, .82 for the parent rewards scale, and .70 for the parent concerns scale. In a l a t e r study of dual-earners, Barnett, Brennan, and Marshall (1994) calculated t e s t - r e t e s t r e l i a b i l i t y . Ten percent of study participants (n = 64) were randomly sampled and interviewed a f t e r the o r i g i n a l contact. The authors reported that for the parent concerns and rewards scales Pearson's r was .83 for men and .81 for women. However, Barnett et a l . (1994) conducted some retests up to three months af t e r the o r i g i n a l t e s t . Thus, the authors probably measured construct s t a b i l i t y rather than t o o l s t a b i l i t y . 77 Comparisons between Baruch and Barnett's (1986) instruments and the REQ suggest that the rewards and concerns scales are s l i g h t l y more stable. However, " r e l i a b i l i t y c o e f f i c i e n t s are s p e c i f i c to the t e s t , the t e s t i n g s i t u a t i o n , and the sample being tested" (Brown, 1970, p. 74). The differences between the r e l i a b i l i t y estimates may be p a r t l y due to the homogeneous, non-random samples recruited by H a l l (1993) and Sleigh. In the preceding section, the discussion focused on the r e l i a b i l i t y of the REQ. To complete the psychometric evaluation of the REQ, the next section reviews findings supporting the instrument's construct v a l i d i t y . Construct V a l i d i t y The following section presents the hypotheses, the r e s u l t s of the hypothesis tes t i n g , and a discussion of the r e s u l t s . Hypothesis Testing In order to examine the construct v a l i d i t y of the REQ, ten hypotheses were generated for the r o l e dimensions: r o l e i n t e n s i t y and r o l e d i s p a r i t y . Table 7 presents a summary of the t e s t scores. 78 Table 7 Test scores for dual-earner men and women (N = 165) Men Women Range of Variable possible scores Mean SD Mean SD Intensity Work 7 - 3 5 18.0 3.2 15.9 2.9 Spouse 30 - 130 88.9 11.8 90.8 11.1 Parent 28 - 140 99.9 15.8 119.2 11.6 Individual 5 - 3 0 17.0 3.7 17.0 2.9 Disparity Work 14 - 70 34.6 8.0 34.0 10.6 Spouse 1 9 - 9 5 46.4 9.0 49.4 11.0 Parent 1 4 - 7 0 28.3 8.1 27.4 8.4 Individual 1 2 - 6 0 31.1 8.7 35.5 8.5 Role i n t e n s i t y . HI. Dual-earner mothers w i l l obtain higher i n t e n s i t y scores i n the parent r o l e than dual-earner fathers. The hypothesis was confirmed by a one-tailed t - t e s t t(163) = 8.96, p_<.001. Since 85% of the sample consisted of dual-earner couples rearing one or more children under the age of three (see Table 1), the finding suggests that mothers of very young children invest more time, e f f o r t , and r e s p o n s i b i l i t y i n the parenting r o l e than fathers. The finding i s well supported i n the l i t e r a t u r e (Hanson & Bozett, 1987). Researchers have documented that employed women spend more hours i n c h i l d care than employed men (Darling-Fisher & Tiedje, 1990; Hochschild, 1989; Scarr et a l . , 1989). Baruch and Barnett (1986) reported the t o t a l i n t e r a c t i o n time between parents and children was 44.5 hours a week for women and 29.5 hours a week for men. Googins (1991) noted that employed female parents spent 24.2 hours on c h i l d care a week while employed male parents spent 14.9 hours. Gender segregation of tasks may account for some of the discrepancy between men's and women's c h i l d care contributions. Women generally assume r e s p o n s i b i l i t y f or children's physical care and perform duties such as bathing, dressing, feeding, and s e t t l i n g children: tasks which are routine, non-discretionary, and time-consuming (Darling-Fisher & Tiedje, 1990). Men, on the other hand, tend to a s s i s t by babysitting and playing with t h e i r children: a c t i v i t i e s which occur less frequently and may be scheduled at one's convenience (Power & Parke, 1984). In addition to expending more time and e f f o r t than men i n the parent r o l e , mothers r e t a i n primary r e s p o n s i b i l i t y 80 for t h e i r children (Barnett et a l . , 1987; Johnson & Johnson, 1980). Many husbands relegate c h i l d care to t h e i r wives and provide help only when s p e c i f i c a l l y asked. Thus, when fathers engage i n c h i l d care, they frequently assume the r o l e of "helper" (Gunter & Gunter, 1991). In order to assess r e s p o n s i b i l i t y , some researchers measured the amount of time men and women spent i n sole charge of t h e i r children. Studies revealed that fathers spent less time alone with t h e i r children than mothers (Baruch & Barnett, 1986; Olds et a l . , 1993). Baruch and Barnett (1986) state "fathers spend an average of 5.5 hours a week in t e r a c t i n g alone with t h e i r children" while "mothers spend an average of 19.6 hours" (p. 987). Olds et a l . (1993) also found gender predicted solo involvement i n c h i l d care; "less than one quarter (of husbands) care for t h e i r children more than four hours a week without t h e i r wives present" (p. 8). The REQ scores accurately r e f l e c t one of the most commonly reported phenomena i n the l i t e r a t u r e , mothers* dominance of the parent r o l e . The REQ also captures parents' involvement i n a variety of c h i l d care a c t i v i t i e s . Items tap into physical care (bathing, diapering, feeding) as well as a u x i l i a r y care (babysitting, playing). With the REQ, a r i c h data base i s generated which may be used by 81 investigators to track evolving trends i n the gender d i s t r i b u t i o n of c h i l d care tasks. Moreover, the REQ permits the investigator to d i r e c t l y measure parents' r e s p o n s i b i l i t y for c h i l d care. Parents are asked to quantify t h e i r l e v e l of r e s p o n s i b i l i t y for each c h i l d care task. This represents a s i g n i f i c a n t improvement over i n d i r e c t methods of assessing r e s p o n s i b i l i t y . H2. Dual-earner fathers w i l l obtain higher r o l e i n t e n s i t y scores i n the paid-work r o l e than dual-earner mothers. The hypothesis was confirmed by a one-tailed t - t e s t t(163) = 4.56, p<.001. Fathers appeared to invest more time, energy, and r e s p o n s i b i l i t y i n the paid-work r o l e than mothers (see Table 7). Indeed, chi-square analysis (1, N = 165 = 22.52, p<.001) revealed that men were more l i k e l y to work i n excess of 30 hours a week then women. Dual-earner fathers averaged 40.7 (SD = 8.6) hours a week i n paid-work as compared to dual-earner mothers who averaged 32.7 (SD = 7.9) hours a week. A negative c o r r e l a t i o n , r(134) = -.33, p<.001, was found between hours spent i n paid work and the int e n s i t y of the parent r o l e for both men and women. The men i n the sample spent more time doing paid work but experienced less i n t e n s i t y i n the parent ro l e while the women experienced the opposite time commitments. 82 Differences i n paid-work hours for men and women have been a consistent finding i n the l i t e r a t u r e (Higgins et a l . , 1994). Googins (1991) compared the paid-work hours of parents employed by a large American corporation, and found that fathers worked an average of 44.7 hours a week while mothers worked 3 9.7 hours a week. In t h e i r study of Canadian public and private sector employees, Higgins et a l . (1994) found that fathers with young children spent 44.0 hours a week engaged i n paid work, while mothers spent 41.8 hours a week. Fathers' high paid-work i n t e n s i t y scores i n conjunction with a negative c o r r e l a t i o n between paid-work hours and parenting i n t e n s i t y may r e f l e c t couples' attempts to balance t h e i r r o l e commitments. Men and women may be l i m i t i n g the time, e f f o r t , and r e s p o n s i b i l i t y involved i n one r o l e so they can function more productively i n another r o l e . Thus, each spouse may deal with multiple-role demands by assigning p r i o r i t y to those roles that hold more meaning and relevance for them. Cowan and Cowan (1988) suggested that parenthood exerts a conservative influence on rol e s . The authors stated that, regardless of where a couple begins on the t r a d i t i o n a l - t o -e g a l i t a r i a n continuum, "men's and women's family r o l e arrangements become increasingly t r a d i t i o n a l " a f t e r the b i r t h of t h e i r f i r s t c h i l d (p. 106). Furthermore, "as 83 psychological involvement i n parenthood increases, i t i s accompanied by a sense of decreasing involvement i n the r o l e of spouse and lover for both mothers and fathers, and contrasting changes i n work involvement for husbands and wives" (p. 118). Thus, the adaptation to parenthood, l i k e other major l i f e t r a n s i t i o n s , "create[s] functional r e l a t i o n s h i p s that were not present before, and ... unhook[s] functional relationships that existed before the l i f e change occurred" (p. 126). Surprisingly, the i n t e n s i t y scores i n the paid-work component were quite low for such a well-educated and highly-paid sample. However, H a l l maintains the REQ i s not designed to measure pressures associated with the paid-work r o l e (personal communication, June 19, 1995). Instead, H a l l hopes to capture respondents 1 perceptions of work-related demands that might intrude on the family domain. H3. Dual-earner mothers with more than one c h i l d w i l l have higher scores on the r o l e i n t e n s i t y dimension of the parent and spouse roles than dual-earner mothers with one c h i l d . The hypothesis was not confirmed. One-tailed t t e s t s revealed no differences i n i n t e n s i t y scores between the two groups (see Table 8) for the parent, t(81) = 0.32, p = .40, and spouse role s , t(81) = 0.31, p = .37. However, the 84 subsamples were small, n = 40 for women with one c h i l d , and n = 43 for women with more than one c h i l d . Table 8 Intensity scores for dual-earner women (N = 83) 1 c h i l d > 1 c h i l d Range of Role possible scores Mean SD Mean SD Spouse 30 - 130 91.2 9.6 90.4 12.4 Parent 28 - 140 118.6 11.4 119.8 11.9 Secondary analysis examined the e f f e c t the number of children had on inte n s i t y scores for men and women (see Table 9). A one-tailed t t e s t indicated that, while there was no difference between groups i n the parent r o l e , t(163) = 0.25, p_ =.40, dual-earners with one c h i l d scored higher on the i n t e n s i t y dimension of the spouse r o l e than dual-earners with more than one c h i l d t(163) = 1.68, p_ = .04. 85 Table 9 Intensity scores for dual-earners (N = 165) 1 c h i l d > 1 c h i l d Range of Role possible scores Mean SD Mean SD Spouse 30 - 130 91.4 10.1 88.4 12.5 Parent 28 - 140 109.4 15.3 109.3 18.3 I t was curious no difference was found between groups for i n t e n s i t y i n the parent r o l e even though 38% of the sample were r a i s i n g two or more children under the age of three. Common sense dictates c h i l d care tasks must increase when there are more young children i n the home. Moreover, mothers would be expected to bear the brunt of the burden. Lennon, Wasserman, & Allen , (1991) reported fathers of infants and toddlers were not as involved i n c h i l d care as fathers of older children. However, i t must be kept i n mind that the subscale quantifies parents' perceptions of r o l e i n t e n s i t y . The subscale i s not an objective measurement of the number of hours respondents invest i n r o l e a c t i v i t i e s . Possibly, 86 parents with large families perform more c h i l d care tasks but they do not perceive the experience to be unduly time-consuming or strenuous. Perhaps additional children provide parents with greater rewards, as well as greater demands (Pleck, 1985). Marks (1977) argued that rewarding roles tend to generate rather than drain energy. Such an e f f e c t would explain why dual-earners with more than one c h i l d do not perceive more parent r o l e i n t e n s i t y than dual-earners with one c h i l d . A l t e r n a t i v e l y , parents with one c h i l d may have higher standards than parents with more than one c h i l d ; parents with larger families may be better organized. Parents with additional children may perform one c h i l d care task for a l l the children at the same time, for example, bathing or playing. For the spouse ro l e , the data suggested that parents with one c h i l d invested more time, e f f o r t , and r e s p o n s i b i l i t y i n housework and/or spousal interactions than parents with two or more children. This finding i s d i f f i c u l t to interpret because of Hall's (1993) construction of the subscale. Four items address aspects of the spousal r e l a t i o n s h i p and 11 items assess the indiv i d u a l ' s perception of the amount of housework (s)he performs. Unfortunately, the questionnaire does not allow the investigator to separate the rel a t i o n s h i p and housework components so as to 87 determine each component's contribution to the o v e r a l l score. Two explanations may account for the high i n t e n s i t y scores found i n parents with one c h i l d . High scores may r e f l e c t respondents' perception of a more intense spousal r e l a t i o n s h i p . Subscale items assess the amount of time and attention spouses devote to each other. Couples with one c h i l d probably have more opportunities to in t e r a c t with each other than couples with two or more children. An a l t e r n a t i v e explanation i s that respondents with one c h i l d perform more housework than respondents with two or more children. Even though researchers suggest that additional children create more housework (Coverman, 1989; Pearlin, 1975), i t i s possible that couples with larger families lower t h e i r housekeeping standards and ac t u a l l y perform less housework than couples with smaller f a m i l i e s . Unless the investigator i s able to separate spousal in t e r a c t i o n scores from housekeeping scores, i t i s not possible to i d e n t i f y which of the preceding explanations might account for the differences between the groups. Calculation of the scores for each component would require the investigator to i d e n t i f y respondents i n the comparison groups and rescore the subscale. The process would be laborious and time-consuming. Furthermore, d i v i d i n g the 88 spouse r o l e into two subscales compromises the accuracy of the r e l i a b i l i t y estimates. H4. Individuals earning high incomes w i l l experience greater r o l e i n t e n s i t y i n the paid-work r o l e than in d i v i d u a l s earning low incomes. The hypothesis was confirmed by one-way ANOVA, F(2, 157) = 33.24, p_ = .03. Individuals earning $40,000 or more scored higher on the in t e n s i t y dimension than indivi d u a l s earning $30,000 - $39,999 or ind i v i d u a l s earning $10,000 - $29,999. This finding makes i n t u i t i v e sense. Individuals are paid high s a l a r i e s to compensate for heavy work-role demands. Although highly paid individuals would be expected to experience intense r o l e demands, the scores were mid-range. The median score for individ u a l s earning $40,000 or more a year was 17.4. The highest score that could be achieved i n t h i s subscale i s 35 (see Table 10). 89 Table 10 Test scores i n the paid-work r o l e for dual-earners (N = 160) Test Scores Range of Income possible scores Mean SD $10,000 - $29,999 7 - 3 5 15.8 2.9 $30,000 - $39,999 7 - 3 5 17.0 3.8 $40,000 and over 7 - 3 5 17.4 3.0 It must be remembered i t was not Hall's intention to assess the pressures i n t r i n s i c to the paid-work r o l e . Rather, H a l l composed subscale items i n order to ascertain the degree to which paid-work r o l e demands intruded on the family domain. While H a l l must be commended for t r y i n g to ascertain respondents' a b i l i t y to compartmentalize r o l e s , her approach introduces l o g i c a l inconsistencies between roles, v i s - a - v i s the in t e n s i t y dimension. For the parent, spouse, and in d i v i d u a l r o l e s , the REQ assesses respondents * perception of the amount of time and e f f o r t they invest i n ro l e a c t i v i t i e s . Unlike the paid-work r o l e , items do not measure the degree to which these roles 90 intrude on other ro l e s . Thus, the operationalization of in t e n s i t y i s not consistent among the four r o l e subscales. H5. Dual-earners who work f u l l - t i m e w i l l have higher r o l e i n t e n s i t y scores i n the paid-work r o l e than dual-earners who work part-time. The hypothesis was confirmed by a one-tailed t - t e s t t(163) = 2.77, p_<.001. Comparison groups included part-time employees working under 3 0 hours a week and f u l l - t i m e employees working 3 0 or more hours a week. This finding may r e f l e c t the confounding of gender with paid-work hours. I t has already been shown that men work more hours than women and t h i s phenomenon contributes to men's perception of in t e n s i t y i n the paid-work r o l e . In the present sample, f u l l - t i m e employees consisted mostly of men, while the part-time employees consisted mostly of women (see Table 11). Chi-square analysis, (1, N = 165) = 15.40, p_<«001, confirmed s i g n i f i c a n t differences i n paid-work hours between genders. 91 Table 11 Number of working hours for men and women (N = 165) Total hours Men Women Under 30 hours 3 28 3 0 or more hours 79 55 Future studies should investigate the e f f e c t that the number of paid-work hours has on women. Findings i n the l i t e r a t u r e are equivocal. Some researchers maintain many married women prefer part-time employment (Moen & Dempster-McCain, 1987; Olds et a l . , 1993). Rosenfield (1989) suggested part-time employment reduces o v e r a l l r o l e demands for women who have heavy family r e s p o n s i b i l i t i e s . Arber, G i l b e r t , and Dale (1985) indicated part-time work has more b e n e f i c i a l e f f e c t s for women under 40 than f u l l - t i m e work. However, Anderson-Kulman and Paludi (198 6) reported mothers f e e l pressured and "torn among the demands of being a worker, mother, and/or wife," regardless of t h e i r employment status (p. 250). Greenhaus and Beutall (1985) have suggested mothers with part-time jobs may be spread t h i n l y because they tend to assume r e s p o n s i b i l i t y for a l l c h i l d care and housekeeping tasks. Darling-Fisher and Tiedje (1990) concurred, noting women employed part-time 92 receive even less help from t h e i r spouses than f u l l - t i m e homemakers. The authors concluded women l i m i t t h e i r work hours and attempt to "do i t a l l " with t h e i r children. In the present study, paid-work hours are negatively correlated with parent r o l e intensity, r(134) = -.33, p_<.001, suggesting that dual-earners channel energy from the paid-work r o l e into the parent r o l e . The c o r r e l a t i o n i s p a r t i c u l a r l y s a l i e n t for women because they are more l i k e l y than men to work fewer hours. Barnett et a l . (1987) believed parent r o l e stressors are more potent than paid-work r o l e stressors. Parental work occurs i n a context of high demands and low control (Rosenfield, 1989). Karasek (1979) associated these conditions with stress. Thus, i t would be i n t e r e s t i n g to determine whether or not women act u a l l y benefit from exchanging paid-work r o l e demands for parent-role demands. The preceding discussion i l l u s t r a t e s the importance of observing the int e r r e l a t i o n s h i p s among ro l e s . F a i l u r e to examine work and family roles simultaneously deprives investigators of an opportunity to explore the complementary nature of ro l e s . The REQ i s to be lauded for capturing a ro l e dynamic that may have a s i g n i f i c a n t impact on r o l e outcomes for mothers i n dual-earner fa m i l i e s . 93 Role d i s p a r i t y . H6. Dual-earners with more than one c h i l d w i l l experience greater d i s p a r i t y i n the spouse and i n d i v i d u a l roles than dual-earners with one c h i l d . The hypothesis was confirmed. One-tailed t - t e s t s revealed a s i g n i f i c a n t difference between the two groups for the spouse, t(163) = 2.31, p = .01, and i n d i v i d u a l r o l e s t(163) = 2.19, p = .02. The groups showed a clear-cut difference i n r o l e perceptions: Parents with one c h i l d reported low l e v e l s of r o l e d i s p a r i t y (meaning the i n t e n s i t y of t h e i r r o l e demands was acceptable). Parents with more than one c h i l d reported higher lev e l s of r o l e d i s p a r i t y (see Table 12). Table 12 Disparity scores for dual-earners (N = 165) 1 c h i l d > 1 c h i l d Range of Role possible scores Mean SD Mean SD Spouse 1 9 - 9 5 46.0 9.0 49.7 10.9 Parent 1 4 - 7 0 26.0 7.3 28.8 9.0 Some studies have suggested greater numbers of children adversely a f f e c t parents' marital q u a l i t y (Cleary & Mechanic, 1983; Coverman, 1989; Vannoy & P h i l l i b e r , 1992; 94 Verbrugge, 1986). Guelzow et a l . (1991) reported an i n d i r e c t association between the number of children i n the home and marital stress. Nock and Kingston (1982) concluded that the decreased a v a i l a b i l i t y of time, and competition for time, accounts for the e f f e c t . These authors f i n d the "presence of children, e s p e c i a l l y young children, seems to reduce the time spouses spend together except, perhaps, i n c h i l d care r o l e s " (p. 392). Chassin et a l . (1985) opined that when couples experience time pressures they p r i o r i t i z e the parent and worker roles at the expense of the spousal r o l e . Voyandoff (1988) maintained that children increase housework and housework i s generally performed by the wife. Pe a r l i n (1975) suspected that additional children lead to wives* disenchantment with homemaking. Coverman (1989) agreed implicating increased housework as the reason that wives' marital s a t i s f a c t i o n i s negatively affected by the number of children i n the home. However, husbands may also experience a decline i n marital q u a l i t y when there are more children i n the home (Vannoy & P h i l l i b e r , 1992). Suitor (1991) believed t h i s r e f l e c t s the delegation of housework from wives to husbands. Marital q u a l i t y suffers i f men's "contribution to household labour i s more extensive than they desire" (p. 227). On the other hand, H a l l (1993) suggested that men simply lack time to maintain the marital r e l a t i o n s h i p . 95 E a r l i e r work by Chassin et a l . (1985) lends credence to Hall ' s contention. The authors reported when fathers have d i f f i c u l t y f u l f i l l i n g the husband and father r o l e s , the husband r o l e i s l e f t wanting. Unfortunately, the REQ does not permit a cl e a r i n t e r p r e t a t i o n of the study findings. The t o o l f a i l s to separate housework from the spousal r e l a t i o n s h i p . Thus, i t i s unknown which aspect of the spouse r o l e may be responsible for d i f f e r i n g d i s p a r i t y scores between the groups. The l i t e r a t u r e does not address the e f f e c t s the number of children have on parents' i n d i v i d u a l r o l e s . However, i t i s l o g i c a l to assume d i s p a r i t y i n the i n d i v i d u a l r o l e i s linked to parents' lack of time. Studies indicate, although dual-earner parents experience pressure due to time shortages, parents with more than one c h i l d experience even greater time pressures (Verbrugge, 1986). Schnittger and Bird (1990) suggested parents p r i o r i t i z e i n order to cope. A popular coping method i s to l i m i t parents* avocational a c t i v i t i e s . This study's findings may also be related to the ages of the children. Younger children have higher dependency needs than older children and may require more physical and emotional care. Thus, young children create a parenting 96 r o l e that i s p a r t i c u l a r l y labour-intensive (Walker & Best, 1991) . Parents of infants and toddlers may ignore the demands of the i n d i v i d u a l r o l e . Studies that report dual-earners* lack of time for exercise, community a c t i v i t i e s , and hobbies support t h i s interpretation. However, researchers have often r e s t r i c t e d t h e i r samples to parents with very young children (Anderson-Kulman & Paludi, 1986; Walker & Best, 199 0). Thus, the population sampled i n these studies may account for the consistency of the findings. H7. Dual-earners who work f u l l - t i m e (30 or more hours a week) w i l l experience greater d i s p a r i t y i n the paid-work r o l e than dual-earners who work part-time (under 30 hours a week). A one-tailed t - t e s t confirmed the difference between the two groups t(163) = 1.98, p_ = .03. The mean score for part-time workers was mid-range, ind i c a t i n g they perceived paid-work r o l e i n t e n s i t y to be neither acceptable/ unacceptable nor controllable/uncontrollable. Full-time workers' scores were moderately high suggesting they found work i n t e n s i t y somewhat unacceptable and uncontrollable. Unfortunately, the small sample siz e precluded entering gender as a control. Analysis of the demographic data revealed that women accounted for 90% of the part-time workers (see Table 11). Lower d i s p a r i t y scores for part-97 time workers may r e f l e c t women's preference for less i n t e n s i t y i n the paid-work r o l e . Thus, a comparison of hours of work was somewhat misleading because the gender issue may have confounded comparisons. H8. Dual-earner fathers w i l l experience greater d i s p a r i t y i n the parent r o l e than dual-earner mothers. The hypothesis was rejected using a one-tailed t - t e s t , t(163) = 0.67, p = .25. Men's d i s p a r i t y scores were higher than women's (see Table 7), but the difference was not s t a t i s t i c a l l y s i g n i f i c a n t . Since power i s .72, i t i s un l i k e l y that the finding resulted from a beta error. Although there were s i g n i f i c a n t differences i n in t e n s i t y scores between the two genders t(163) = 8.96, p<.001, the men and women shared s i m i l a r perceptions regarding the ac c e p t a b i l i t y of t h e i r r o l e demands. The low di s p a r i t y scores for men and women indicate both genders perceived the amount of time, e f f o r t , and r e s p o n s i b i l i t y they invested i n the parent r o l e as acceptable. The f a i l u r e to f i n d high d i s p a r i t y scores for men does l i t t l e to support the notion that today's generation of fathers desires greater p a r t i c i p a t i o n i n c h i l d rearing and are frustrated by exclusion. Rotundo (1985) believed the adoption of a more nurturant model of fatherhood would be problematical because i t "involves a substantial recasting of ... manhood, 98 womanhood, and family l i f e . I t demands new emotional st y l e s [and] i t e n t a i l s d i f f e r e n t notions of male and female" (p. 16). Moreover, Rotundo suggested men may have d i f f i c u l t y r e l i n q u i s h i n g the t r a d i t i o n a l s t y l e of fatherhood. Men may be "too thoroughly ingrained with 'male' values of ambition and achievement to devote much time to substantial d a i l y care" (Rotundo, 1985, p. 20). The data also suggest that the women i n t h i s sample do not want or expect more involvement i n c h i l d care from t h e i r spouses. These women seem content with the i n t e n s i t y of t h e i r parent r o l e . The l i t e r a t u r e provides some support for t h i s f i nding. Many women are not d i s s a t i s f i e d with being primary providers of c h i l d care (Darling-Fisher & Tiedje, 1990; Lennon et a l . , 1991). Indeed, some women covertly exclude men from d i r e c t c h i l d care (Hanson et a l . , 1985). Glendinning and M i l l a r (1992) offered one explanation as to why women are unwilling to re l i n q u i s h c h i l d care t u r f . The authors think women's lack of power i n the occupational domain drives them to estab l i s h a power base i n the home. Thus, some women perceive men's p a r t i c i p a t i o n i n c h i l d care as a threat to t h e i r p o s i t i o n of power i n the family. That the perceived q u a l i t y of the parent r o l e i s sim i l a r for men and women i s not a new finding. Barnett et a l . (1994) studied 180 dual-earner couples, the majority of whom were employed i n management and professional 99 occupations. Eighty-three percent of the couples were rearing preschoolers. The authors reported mothers and fathers obtained almost i d e n t i c a l scores i n the parental rewards and concerns scale. Moreover, reward scores were considerably higher than concern scores, suggesting dual-earners enjoyed q u a l i t y parenting experiences. The REQ i s highly e f f e c t i v e i n capturing dual-earners' experiences i n the parent r o l e . The items include an exhaustive s e l e c t i o n of c h i l d care a c t i v i t i e s i d e n t i f i e d by the l i t e r a t u r e as being i n t e g r a l to the parenting experience. I t i s highly u n l i k e l y a major component of the parent r o l e has been overlooked. However, Hall's (1993) detailed coverage of c h i l d care tasks l i m i t s the u t i l i t y of the REQ with older children. Tasks such as feeding, bathing, dressing, and diapering are relevant to parents with infants and toddlers. Thus, the REQ i s li m i t e d to t e s t i n g dual-earner parents with young children. H9. Dual-earner mothers w i l l obtain higher d i s p a r i t y scores i n the i n d i v i d u a l r o l e than dual-earner fathers. A one-tailed t - t e s t confirmed the hypothesis, t(163) = 3.23, p_<.001. Scores suggested mothers found the i n t e n s i t y of the i n d i v i d u a l r o l e somewhat unacceptable and/or uncontrollable (see Table 7) as opposed to fathers who 100 perceived the in t e n s i t y of the in d i v i d u a l r o l e as being acceptable and/or con t r o l l a b l e . The l i t e r a t u r e consistently demonstrates employed mothers spend less time i n l e i s u r e a c t i v i t i e s than men (Les l i e & Anderson, 1988; Schnittger & Bird, 1990). Women spend more hours a week on combined work and family r e s p o n s i b i l i t i e s and have less free time than men (Duxbury, Higgins, & Lee, 1994). Googins (1991) estimated an employed mother's t o t a l work week as 83.7 hours while an employed father's was 72.1 hours. G i l l i g a n (1982) believed society expects women to subordinate t h e i r personal needs i n order to meet the needs of t h e i r f a m i l i e s . Many women f i n d " i t ... more s o c i a l l y acceptable to cut back on community a c t i v i t i e s or l e i s u r e a c t i v i t i e s than to reduce time i n c h i l d care or for spousal support" (Schnittger & Bird, 1990, p. 202). However, H a l l (1992) suggested women who s a c r i f i c e t h e i r own needs i n order to meet the needs of t h e i r families experience feelings of exhaustion, resentment, and anger (Hall, 1992). Thus, there i s support for the notion that women's lack of involvement i n the in d i v i d u a l r o l e may be a source of discontentment. Men, on the other hand, are not as s o c i a l l y conditioned to deny t h e i r needs. Men i n one study indicated "they d i d not and would not put t h e i r needs l a s t " (Hall, 1992, p. 36). 101 Male dual-earners i n Schnittger and Bird's (1990) study lim i t e d t h e i r avocational a c t i v i t i e s less than women. H10. Dual-earner mothers working f u l l - t i m e who have college or professional preparation w i l l have lower d i s p a r i t y scores i n the paid-work r o l e than dual-earner mothers working f u l l - t i m e who have a high school diploma or less . The hypothesis was rejected using a one-tailed t - t e s t , t(51) = .83, p = .20. However, t h i s finding probably represents a beta error. The power calculated for the entire sample (N = 165) for comparisons using the education variable was only .20. Hypothesis 10 used a subsample of only 55, so the power i s considerably lower than .20. A finding of equivalency between groups i s not supported by the l i t e r a t u r e . Researchers consistently f i n d p o s i t i v e correlations between women's educational attainment and the qu a l i t y of t h e i r work roles (Hibbard & Pope, 1985; Kessler & McRae, 1982). Ulbrich (1988) concluded women are more l i k e l y to prefer t h e i r paid work i f they are highly educated and employed f u l l - t i m e . Moreover, the qu a l i t y of the work r o l e exerts a powerful e f f e c t on an indiv i d u a l ' s attitude (Barnett et a l . , 1992; Grimm-Thomas & Perry-Jenkins, 1994). Research further demonstrates highly educated women experience greater congruence between t h e i r self-concepts 102 and the r o l e demands of paid-work. Woods (1985) suggested women with advanced education are l i k e l y to have modern gender r o l e norms. Women with non-traditional orientations may perceive the paid-work r o l e to be more acceptable than women who espouse t r a d i t i o n a l values. Moreover, Repetti et a l . (1989) maintained employment i s b e n e f i c i a l when there i s "congruity between women's attitudes toward ... employment and t h e i r actual r o l e status" (p. 1397). The preceding section reviewed the findings from the hypothesis t e s t i n g . To further explore the construct v a l i d i t y of the REQ, confirmatory factor analysis (CFA) was conducted. Confirmatory Factor Analysis Three CFAs were performed. Factors were extracted using the P r i n c i p a l Components method with varimax r o t a t i o n and Kaiser normalization. Analysis one (intensity and d i s p a r i t y ) . A two-factor solution was proposed. Intensity items were expected to correlate with factor one and d i s p a r i t y items were expected to correlate with factor two. Three i t e r a t i o n s were necessary. The factors correlated moderately with each other, suggesting non-orthogonality (r =.60). Since i n t e n s i t y and d i s p a r i t y are two aspects of the same construct, one would expect some degree of c o r r e l a t i o n . The t o t a l variance explained by the two factors was 22.4%. 103 Out of 67 items, 46 loaded on factor one. However, the magnitude of the loadings was modest. Only two parent r o l e items achieved correlations greater than .71. Forty-three percent of the items ranged from .45 to .68. Twenty-three percent of item correlations ranged from .31 to .45. Another 23% of the items f a i l e d to correlate adequately with either factor and were dropped from the analysis (r<.30). The non-loading items included: four parent r o l e items; three i n d i v i d u a l r o l e items; two spouse r o l e items, and two paid-work r o l e items. Three parent r o l e items and one spouse r o l e item loaded on both factors. Correlations with factor two were r e l a t i v e l y low, ranging from .30 to .36. Nine items did not load on factor one. Instead they loaded on factor two, displaying correlations ranging from .30 to .49. These items included three items each for the spouse and in d i v i d u a l roles and one paid-work r o l e item. For factor two, 43 out of 59 items loaded. Loadings were modest. For 8% of the items, correlations were greater than .55. Thirty-seven percent achieved correlations ranging from .45 to .55 and 45% of the items loaded from .30 to .44. Twenty-two percent of the items did not achieve the minimal c o r r e l a t i o n of .30 for either factor and were dropped from analysis. They included; 11 paid-work r o l e items; one spouse r o l e item; and one i n d i v i d u a l r o l e item. 104 One parent r o l e item loaded only on factor one (r = .43) and another parent r o l e item loaded on both factors. Correlations with factors one and two for t h i s item were .30 and .54, respectively. The l i t e r a t u r e does not provide a d e f i n i t i v e technique for judging the q u a l i t y of a factor analysis. Johnson and Wichern (1992) suggested investigators use the WOW c r i t e r i o n . " I f while s c r u t i n i z i n g the factor analysis, the investigator can shout, 'wow, I understand these factors,' the application i s deemed successful" (p. 444). Vi s u a l inspection of the rotated factor matrix, revealed that the items had a clear-cut a f f i n i t y for one or the other factor. Items loaded i n the hypothesized manner. Two d i s t i n c t columns emerged, each of which contained, for the most part, either i n t e n s i t y or d i s p a r i t y items. Further examination of the factor analysis prompted concern. Only 22.4% of the variance among t e s t scores was accounted for by the two-factor solution. Limited variance may be related to the magnitude of the item loadings. High loadings are desirable because "the greater the loading, the more the variable i s a pure measure of the factor" (Tabachnik & F i d e l l , 1989, p. 640). In order to determine the q u a l i t y of loadings, Tabachnik and F i d e l l (1989) recommended Comfrey's scale. "Loadings i n excess of .71 (50% overlapping variance) are considered excellent, .63 105 (40% overlapping variance) very good, .55 (30% overlapping variance) good, .45 (20% overlapping variance) f a i r , and .32 (10% overlapping variance) poor" (p. 640). Applying Comfrey's c r i t e r i a (Tabachnik & F i d e l l , 1989) to the rotated factor matrix rendered the following evaluation. For factor one ( i n t e n s i t y ) , 46% of the items either loaded poorly or not at a l l . For factor two ( d i s p a r i t y ) , 68% of the items loaded poorly or not at a l l . Thus, a s i g n i f i c a n t proportion of the t e s t items provided l i t t l e or no information about the factor. Although the two factors were not potent explainers of the variables, the f a i l u r e may l i e with the investigator for hypothesizing too simple a factor solution. I t must be remembered the REQ was generated from q u a l i t a t i v e research; the instrument operationalizes a multidimensional construct. Intensity and d i s p a r i t y are but two dimensions of r o l e enactment. Four r o l e components that are often s p e c i f i c to males or females are also included i n the operationalization. A two-factor solution may not be comprehensive enough to capture a construct as complex as r o l e enactment. Sample homogeneity may also account for the low magnitude of the correlations. Brown (1970) stated c o r r e l a t i o n c o e f f i c i e n t s "are influenced by the d i s t r i b u t i o n of scores within the sample used to calculate the 106 c o e f f i c i e n t " (p. 69). Recall the s i m i l a r i t y i n demographic c h a r a c t e r i s t i c s among dual-earners enrolled i n the study. Couples were predominantly white and shared the same l i f e - c y c l e stage, geographic location, and educational background. A sample of t h i s composition cannot be expected to produce a broad range of scores. Unfortunately, though, "as the v a r i a b i l i t y ... of the scorefs] decrease, the c o r r e l a t i o n c o e f f i c i e n t generally decreases ..." (Brown, 1970, p. 69). The small sample siz e i s another methodological shortcoming that might account for the findings. Nunnally (1978) recommended a minimum sample of 650 indiv i d u a l s for a factor analysis of 126 items. When fewer respondents are tested the range of scores i s limited, further constraining v a r i a b i l i t y . Analysis two (intensity i n four r o l e s ) . A four-factor solution was proposed which accounted for 37.4% of the variance. Sixty-one out of 67 items loaded over .30 and were retained for analysis. S i x t y - s i x percent of the items achieved correlations greater than .45. Although varimax rot a t i o n converged i n eight i t e r a t i o n s , items did not cl u s t e r under the four factors as hypothesized. Due to the tendency of the parent and spouse r o l e items to double and t r i p l e load, these items were over-represented i n the factor matrix. 107 Factor one contained 14 spouse r o l e items and 14 parent r o l e items. The spouse r o l e items showed the highest c o r r e l a t i o n with the factor. A l l 14 items loaded i n excess of .45. Parent r o l e item loadings ranged from .32 to .51. Factor two included 13 parent r o l e items, 4 spouse r o l e items (r = .34 to .47), and 2 in d i v i d u a l r o l e items (r = .32 to .42). The parent r o l e items provided the heaviest loading, with 9 correlations over .45. Factor three contained 12 parent r o l e items, 4 spouse items (r =.33 to .47), and 3 paid-work items (r = .32 to .41). The parent r o l e items c l e a r l y defined the factor and provided 6 correlations over .45. Correlations for factor four included; 11 spouse r o l e items (r =.32 to .66), 4 parent r o l e items (r = .31 to .44), and 1 i n d i v i d u a l r o l e item (r = .32). The spouse items dominated with 8 correlations over .45. Sixteen items loaded on 2 factors and included 10 parent r o l e items and 6 spouse r o l e items. Four parent r o l e items t r i p l e - l o a d e d . Triple-loaders demonstrated s i m i l a r correlations across three factors. The four-factor solution accounted for more variance than the two-factor solution. The two-factor solution may have f a i l e d to adequately represent the balance that dual-earners maintain among t h e i r r o l e demands. In the 108 two-factor solution high scores i n one r o l e may have been cancelled out by low scores i n another r o l e . With the four-factor solution, the ranges i n scores may have been better preserved, thus capturing greater variance. On the other hand, the four-factor solution may have provided more "homes" for items. Low loading items dropped from the two-factor solution may have been more r o l e - s p e c i f i c than i n t e n s i t y - s p e c i f i c . With the four-factor solution, items had an opportunity to c l u s t e r under r o l e factors. Thus, more items were retained i n the analysis. The four-factor solution contained more information thereby allowing the factors to account for a greater proportion of variance i n the scores. However, the four-factor solution f a i l e d to produce the hypothesized structure. Each of the four columns contained items from the parent and spouse ro l e s . Moreover, instead of c l u s t e r i n g together, the i n d i v i d u a l items were randomly d i s t r i b u t e d over three factors. The dominance of the parent and spouse r o l e items i n the analysis i s undoubtedly a function of the REQs construction. The author's q u a l i t a t i v e work informed her decision to oversample items i n the family domain. Hall ' s (1993) sample of dual-earner parents with infants placed a lower p r i o r i t y on the i n d i v i d u a l and paid-work r o l e s . 109 Despite the considerable overlapping of r o l e items, the rotated matrix suggested factors one and four r e l a t e d to the spouse r o l e while factors two and three represented the parent r o l e . Examination of the questionnaire statements associated with the items supported t h i s i n t e r p r e t a t i o n . For the spouse r o l e (factors one and four), analysis suggested items clustered according to the t r a d i t i o n a l gender segregation of tasks. Factor one contained items consistently i d e n t i f i e d i n the l i t e r a t u r e as feminine domestic tasks, such as meal preparation, vacuuming, and buying groceries ( B l a i r & Johnson, 1992, Marat & Finlay, 1984) . Factor four contained items describing tasks t y p i c a l l y performed by men, for instance, house repairs, garbage disposal, and car maintenance (B l a i r & Johnson, 1992; Marat & Finlay, 1984). Items that loaded on both factors may represent tasks that genders share, for example, t i d y i n g the house and laundry (Marat & Finlay, 1984). Non-loading items described r o l e demands that were unrelated to the family domain and included, l e i s u r e a c t i v i t i e s , hobbies, work-related education, and commuting. For the parent r o l e (factors two and three), the pattern of c l u s t e r i n g may also represent gender segregation of tasks. Male-oriented a c t i v i t i e s clustered under factor two and suggested a planning or organizational theme. Items 110 referred to long-term planning for c h i l d ( r e n ) , the organization of the child(ren)*s health care, and the handling of safety concerns. However, factor three described d i r e c t c h i l d care a c t i v i t i e s , usually performed by mothers. A c t i v i t i e s included bathing, feeding, and diapering children. Double-loading items suggested tasks that might be shared by men and women, such as feeding, dressing, and playing with children. Three paid-work r o l e items also loaded on factor three. Items described overtime, paid work beyond regular employment, and amount of t r a v e l . Possibly, these items share intrusiveness on the parent r o l e as a common at t r i b u t e . Interpretation of the factor matrix suggests when i t comes to in t e n s i t y , factors are not e n t i r e l y r o l e - s p e c i f i c . The factors are also defined by the gender s p e c i f i c i t y of household and c h i l d care tasks. This finding i s consistent with the l i t e r a t u r e which demonstrates the tenacity of the t r a d i t i o n a l d i v i s i o n of labour (Benin & A g o n s t i n e l l i , 1988; Berardo, Shehan, & L e s l i e , 1987). Analysis three (disparity i n four r o l e s ) . Analysis three r e p l i c a t e d analysis two but examined the d i s p a r i t y dimension. The t o t a l variance explained by the four factors was 3 6.2%. Varimax rotation converged i n 11 i t e r a t i o n s . A I l l moderate degree of c o r r e l a t i o n existed between factors representing the in d i v i d u a l r o l e and the other roles (r = .47 to .69). The parent and paid-work r o l e factors were orthogonal (r = .26). However, the spouse and paid-work r o l e factors correlated s u b s t a n t i a l l y with each other (r = .90) . Such a high c o r r e l a t i o n suggests a great deal of conceptual overlap between the subscales. However, items i n the paid-work r o l e examine demands such as commuting, work-related education, and overtime. The spouse r o l e looks at housekeeping duties and the spousal i n t e r a c t i o n . I t i s d i f f i c u l t to understand how such diverse a c t i v i t i e s could correlate so highly. A l l 59 items i n the d i s p a r i t y scale achieved correlations greater than .30 and were included i n the analysis. Item loadings were r e l a t i v e l y s a t i s f a c t o r y . Sixty-eight percent of the items correlated over .45 with the factors. Factor one was shared by the parent and spouse roles which contributed 13 items (r = .42 to .66) and 10 items (r =.41 to .60), respectively. Factor two was f a i r l y d i s t i n c t and consisted of 10 ind i v i d u a l r o l e items (r = .36 to .78). Four spouse r o l e items and one paid-work r o l e item were also included, but the loadings were less than .43. 112 In factor three the spouse r o l e dominated with eight items (r = .33 to .73). However, the parent and i n d i v i d u a l r o l e s were also represented. Four parent items loaded (r =.31 to .49) and four i n d i v i d u a l r o l e items loaded (r =.34 to .57). The most clear-cut pattern emerged for factor four. Twelve paid-work r o l e items loaded on the factor (r = .30 to .68). Only one non-paid-work r o l e item loaded, a spouse item (r = .3 0). Ten items loaded on two factors, simultaneously. However, loadings for the alte r n a t i v e factors were low, ranging from .31 to .41. Double-loading items included three items each from the parent, spouse, and i n d i v i d u a l r o l e s . One spouse r o l e item loaded on three factors (r = .30 to .43). V i s u a l l y , the rotated factor matrix looked f a i r l y orderly, with items c l u s t e r i n g together and a l i g n i n g under the four factors. Factors two and four were quite d i s t i n c t and d i f f e r e n t i a t e d between in d i v i d u a l and paid-work r o l e items well. However, factor one contained a mixture of parent and spouse r o l e items. Factor three was somewhat chaotic and also displayed a mixture of parent and spouse items. Notable i s the f a i l u r e of the analysis to separate parent and spousal r o l e items. Undoubtedly both roles share 113 common attr i b u t e s , since they account for the heaviest loadings on factors one and three. Factor one i s biased i n favour of parent r o l e items, while factor three i s biased towards spouse r o l e items. Examination of the questionnaire statements i n the parent and spouse r o l e subscales reveals an i n t e r e s t i n g pattern. Factor one almost exclusively contains task-oriented items such as feeding, s e t t l i n g c h i l d ( r e n ) , dressing c h i l d ( r e n ) , doing the dishes, and maintaining the yard. On the other hand, factor three contains a l l the relationship-oriented items. This pattern suggests that the re l a t i o n s h i p and housework components of the spouse r o l e may represent two d i f f e r e n t factors. Furthermore, the parent r o l e items and the housework component of the spouse r o l e seem to combine and define a common factor which may describe d a i l y family maintenance tasks. Chapter Summary In the preceding chapter, s i g n i f i c a n t findings derived from the r e l i a b i l i t y and v a l i d i t y t e s t i n g of the REQ were presented. The re s u l t s were discussed i n the context of the l i t e r a t u r e . A c r i t i q u e of the instrument was offered, insofar as r e l i a b i l i t y and v a l i d i t y performance was concerned. 114 In chapter f i v e , the conclusions related to the study findings are discussed. The implications of the conclusions for nursing are then developed. The chapter also includes recommendations for the r e v i s i o n of the REQ. 115 CHAPTER FIVE Conclusions and Implications Chapter f i v e focuses on the conclusions drawn from the psychometric t e s t i n g of the REQ. Implications for t o o l development and recommendations for r e v i s i o n of the REQ are included. The REQ's u t i l i t y for nursing i s discussed. Conclusions related to hypothesis t e s t i n g are also described and the implications of these conclusions for nursing are offered. Conclusions from Tool Testing In the following section, conclusions regarding the r e l i a b i l i t y and v a l i d i t y of the REQ are presented. R e l i a b i l i t y and V a l i d i t y of the REQ The REQ behaved well psychometrically. R e l i a b i l i t y t e s t i n g indicated that the REQ has excellent i n t e r n a l consistency, with the exception of i n t e n s i t y i n the paid-work and i n d i v i d u a l r o l e s . Substandard alphas for the two r o l e components may be a function of the homogeneous sample and the limited score variance. However, the brevity of the subscales may have precluded adequate r e l i a b i l i t y estimates. The REQ appeared to be f a i r l y stable over time. Because a high percentage of the respondents i n t h i s author's study delayed returning the second questionnaire, t e s t - r e t e s t estimates were lower than those obtained by H a l l 116 (1993). The REQ compared favourably with more mature scales which also measure r o l e q u a l i t y . Strong support for construct v a l i d i t y was shown i n hypothesis t e s t i n g . Although the sample was small, s i g n i f i c a n t findings characterized t e s t i n g of eight of the ten hypotheses. Seven hypotheses were accepted and one was rejected. The parent and paid-work r o l e subscales best d i f f e r e n t i a t e d between demographic groups for i n t e n s i t y , while the paid-work, spouse, and i n d i v i d u a l r o l e subscales most e f f e c t i v e l y d i f f e r e n t i a t e d between groups for d i s p a r i t y . Differences found among the demographic groups concurred with findings i n the l i t e r a t u r e . Moreover, the REQ demonstrated an i n t e r r e l a t i o n s h i p between the paid-work and parent roles which was also consistent with the l i t e r a t u r e . Construct v a l i d i t y for the REQ was p a r t i a l l y supported by confirmatory factor analysis (CFA). In the f i r s t analysis, the rotated factor matrix demonstrated the hypothesized r o l e dimensions: r o l e i n t e n s i t y and r o l e d i s p a r i t y . Analysis also showed some support for the presence of four r o l e components in the d i s p a r i t y dimension. Unfortunately, the proportion of variance accounted fo r by the solutions was r e l a t i v e l y low. 117 I t i s not surprising, however, that the simple two- and four-factor solutions f a i l e d to adequately account for a reasonable proportion of variance among REQ scores. The REQ operationalizes a construct that i s complex and multidimensional. More factors may have to be considered i n order to capture the construct of ro l e enactment. Contrary to the author's prediction, a four-factor CFA f a i l e d to support the existence of the paid-work and in d i v i d u a l roles i n the in t e n s i t y dimension. Analysis also f a i l e d to d i f f e r e n t i a t e between the parent and spouse r o l e s . The rotated matrix demonstrated that the hypothesized four factors consisted almost exclusively of parent and spouse r o l e items. Each r o l e subscale separated and defined two factors. The factors seemed to r e f l e c t gender segregation of domestic tasks. Furthermore, the four-factor solution for d i s p a r i t y revealed r e l a t i o n s h i p and housework items from the spouse r o l e subscale did not define the same factor. The factor matrix demonstrated the spouse r o l e divided into two factors. Relationship items clustered under one factor and housework items joined with parent r o l e items to define another factor. These findings tend to mitigate against the REQ's construct v a l i d i t y . Items should hang together and c l u s t e r under the hypothesized factors. I t may be necessary to 118 revise the instrument i n order to address conceptual problems i d e n t i f i e d by the CFA. The unexpected finding that i n t e n s i t y factors were s p l i t along gender l i n e s lent i n d i r e c t support for the REQ. The instrument captured strong gender preferences towards r o l e a c t i v i t i e s . The finding i s consistent with research demonstrating the tendency of parents with young childre n to assume a t r a d i t i o n a l d i v i s i o n of labour. Implications for Tool Development Psychometric t e s t i n g revealed that further work must be done on the REQ. In the next section recommendations for t o o l r e v i s i o n and future psychometric t e s t i n g are presented. Tool Revision The following section presents a c r i t i q u e of the four subscales. Suggestions for r e v i s i o n of the REQ are offered. The paid-work r o l e . In order to r e f i n e the REQ's operationalization of intensity, the paid-work r o l e subscale must be revised. As written, the subscale introduces l o g i c a l inconsistency into the REQ. Instead of measuring the i n t e n s i t y of demands associated with job performance, the subscale focuses only on job demands which have the p o t e n t i a l to intrude on the family r o l e . The other subscales do not share t h i s function. They measure the time and e f f o r t individuals invest i n r o l e 119 demands. Thus, in t e n s i t y has a somewhat d i f f e r e n t meaning for the paid-work r o l e than for the other r o l e s . Items that tap into demands related to the performance of paid work should be added to the subscale. The addition of more items would also contribute to the subscale's r e l i a b i l i t y . With only seven items, the subscale i s too short. Using Nunnally's formula (1978) 25 items would have to be added, i n order to render an alpha of .70. However, the addition of 2 5 items increases the REQ from 126 to 151 items. Some research part i c i p a n t s might f i n d a questionnaire of t h i s length somewhat daunting. Instead of adding more items, one might also consider incorporating two more steps into the L i k e r t r a t i n g scale. Nunnally (1978) states that, " i f there are only half a dozen items i n the scale, the r e l i a b i l i t y ... may be markedly increased by an increase i n the number of scale steps" (p. 597). Instead of f i v e response choices, respondents would be offered seven. For example, None A great deal The amount you do dishes. 1 2 3 4 5 6 7 (a) How acceptable i s t h i s to you? 1 2 3 4 5 6 7 (b) How much r e s p o n s i b i l i t y do you 1 2 3 4 5 6 7 take for this? 120 Some paid-work items i n the REQ could be deleted. Item number two, "the amount of time you miss work" i s not consistent with the d e f i n i t i o n of i n t e n s i t y . S t r i c t l y speaking, missing work decreases the time and e f f o r t associated with the paid-work r o l e . Moreover, absence from work, does not have the p o t e n t i a l to intrude on family r o l e s . R e l i a b i l i t y analysis indicates deletion of t h i s item increases Cronbach's alpha from .35 to .41. I t i s also unclear how income adequacy i s re l a t e d to i n t e n s i t y i n the paid-work r o l e . A high income may be related to i n t e n s i t y for dual-earners. However, an adequate income may simply r e f l e c t the family's f r u g a l i t y or s a t i s f a c t i o n with a low-paying job. Removal of the item increases Cronbach's alpha to .38. The spouse r o l e . As previously discussed, the spouse r o l e subscale contains items oriented to the r e l a t i o n s h i p and housework. However, combining these items l i m i t s the u t i l i t y of the subscale. For example, the author had d i f f i c u l t y i n terpreting findings i n hypotheses three and s i x . (Recall that hypothesis three confirmed that parents with one c h i l d experienced greater i n t e n s i t y i n the spouse r o l e than parents with two children. Hypothesis s i x demonstrated that parents with more than one c h i l d experienced greater d i s p a r i t y i n the spouse r o l e than parents with one c h i l d ) . Moreover, CFA suggests that the 121 r e l a t i o n s h i p and housework items define two d i f f e r e n t factors. On an i n t u i t i v e l e v e l , the two components seem poorly suited to share a subscale. The concept of r e l a t i o n s h i p i s q u a l i t a t i v e l y d i f f e r e n t from the concept of housework. Furthermore, the spousal rel a t i o n s h i p and housework probably have almost opposite e f f e c t s on dual-earners. Engaging i n p o s i t i v e interactions with t h e i r partner i s l i k e l y perceived by most spouses as a supportive, regenerative experience. However, the performance of domestic chores might be considered somewhat burdensome. Thus, r e l a t i o n s h i p and housework items probably should not be combined i n the same subscale. The parent r o l e . Curiously, the items included i n the parent r o l e r e s u l t i n a s o l i d subscale but l i m i t the usefulness of the REQ. The comprehensiveness of c h i l d care items attests to the subscale's v a l i d i t y . However, the items apply to infants and toddlers, only. Use of the REQ must be r e s t r i c t e d to dual-parents with young children. Perhaps future REQs might address parental demands associated with school children and adolescents. The i n d i v i d u a l r o l e . The f a i l u r e of the four-factor solution for i n t e n s i t y to i d e n t i f y a factor corresponding to the i n d i v i d u a l r o l e i s probably p a r t i a l l y due to the subscale's brevity. There were not enough items to define a 122 d i s t i n c t factor. Lengthening the subscale w i l l help to improve v a l i d i t y and increase r e l i a b i l i t y . I f the subscale i s increased from 6 to 12 items, Cronbach's alpha r i s e s from .53 to .70 (Nunnally, 1978). To determine the kind of items that should be included, Nunnally (1978) recommends that the investigator review item-total s t a t i s t i c s and i d e n t i f y items with high co r r e l a t i o n s . The themes of these items should form the nucleus for newly constructed items. For the i n d i v i d u a l r o l e , items related to s o c i a l a c t i v i t i e s achieved the highest c o r r e l a t i o n s . Thus, pote n t i a l new items should tap into the amount of time dual-earners spend engaging i n a c t i v i t i e s of a s o c i a l nature. R e l i a b i l i t y may also be increased with the deletion of item 42: the amount dual-earners spend planning and attending health care appointments. Analysis revealed that deletion would increase r e l i a b i l i t y to .58. H a l l suggested the item may r e f l e c t a parent's tendency to arrange appointments for the whole family, not just for him/herself (personal communication, June 19, 1995). Future Psychometric Testing; of the REQ Further t e s t i n g of the REQ would be advisable. I t i s with the accumulation of evidence from multiple studies that an instrument's usefulness i s confirmed (Weinert & Tilden, 1990). Researchers contemplating psychometric t e s t i n g of 123 the REQ are advised to pay careful attention to sampling procedures. A large and heterogeneous sample should be recruited. Since the REQ i s r e s t r i c t e d to parents with young children, investigators must endeavour to sample dual-earners who vary with regards to family s i z e , occupation, income l e v e l , and geographic loca t i o n . Sample respondents should be offered an REQ that i s user-friendly. The questionnaire should be double-spaced. The anchor statements and corresponding numerical responses must be easy to track across the page. Comprehensive instructions would provide clear guidelines for completing the questionnaire and improve the accuracy of p a r t i c i p a n t s * responses. Data analysis might include CFAs that propose 8 or 16 factor solutions. For instance, an 8 factor solution would examine the presence of i n t e n s i t y and d i s p a r i t y i n the 4 subscales for both genders. A 16 factor solution might i d e n t i f y how items function for each gender v i s - a - v i s r o l e i n t e n s i t y and r o l e d i s p a r i t y i n the four r o l e components. U t i l i t y of the REQ for Nursing The REQ may be confidently used to t e s t dual-earners i n the community. Such information would provide a clearer understanding of some demographic variables that predict r o l e stress, as r e f l e c t e d i n high l e v e l s of r o l e i n t e n s i t y and r o l e d i s p a r i t y . Information of t h i s nature would give 124 p r a c t i t i o n e r s a better idea of dual-earner parents who may be at r i s k for some forms of r o l e stress. However, unless revised, the REQ probably should not be used for c o r r e l a t i o n a l research. R e l i a b i l i t y estimates f o r i n t e n s i t y i n the paid-work and i n d i v i d u a l roles are suboptimal. Investigators who use the REQ may not be able to r e l a t e t e s t scores to other variables with any degree of confidence. Conclusions from Hypothesis Testing Despite t h e o r i s t s ' prediction of increasing r o l e convergence, the findings from t h i s study support men dominating i n the work r o l e and women dominating i n the parent r o l e . Moreover, gender segregation continues to inform dual-earners' domestic task assignments. The continuation of the t r a d i t i o n a l d i v i s i o n of labour must be regarded as a testament to the tenacity of gender r o l e norms. Gender strongly predicts parents• contentment with i n d i v i d u a l r o l e a c t i v i t i e s . Mothers f i n d the i n t e n s i t y of the i n d i v i d u a l r o l e unacceptable. Gender s o c i a l i z a t i o n patterns that stress the precedence of the family may be responsible for mothers' tendencies to place a low p r i o r i t y on r o l e demands associated with the i n d i v i d u a l r o l e . The REQ provides an excellent opportunity to observe the interdependence of ro l e s . Dual-earners engage i n 125 trade-offs among t h e i r r o l e commitments i n order to accommodate competing r o l e demands. One strategy involves l i m i t i n g mothers' paid work so that they may invest greater time and e f f o r t i n the parent r o l e . However, i t i s unclear how t h i s a f f e c t s r o l e outcomes for women. Men and women i n t h i s study seem content with f a t h e r s 1 involvement i n c h i l d care. Perhaps, men's desires for an androgynous model of fatherhood are no longer as strong as they once were. Or, these dual-earners may f i n d the bar r i e r s to egalitarianism more formidable than expected. Certainly, the current economic climate cannot be regarded as an id e a l time for dual-earner parents to experiment with the paid-work r o l e . Regardless, these r e s u l t s suggest e g a l i t a r i a n sharing of c h i l d care tasks has yet to occur. Findings suggest that dual-earners with more than one c h i l d may jeopardize marital qu a l i t y i n order to meet additional parent r o l e demands. This finding hints that there may be an upper l i m i t to the number of r o l e demands that an i n d i v i d u a l can e f f e c t i v e l y accommodate. Beyond that l i m i t d e terioration i n ro l e q u a l i t y may occur. Implications for Nursing Conclusions drawn from hypothesis t e s t i n g have implications for nursing practice, education, administration, and research. 126 C l i n i c a l Practice Prenatal classes provide an excellent forum for nurses to share information with parents. A topic that might be placed on the agenda i s the impact of parenthood on dual-earners' r o l e s . Nurses could advise f i r s t - t i m e expectant couples that roles may become more t r a d i t i o n a l following c h i l d b i r t h . Couples may wish to explore the topi c with each spouse i d e n t i f y i n g his/her expectations for task-sharing i n the postnatal period. Nurses are i n a pos i t i o n to encourage expectant couples to f i n d balance among t h e i r r o l e s . Dual-earners should be informed that neglecting the i n d i v i d u a l or spouse ro l e s i n favour of the paid-work and/or parent roles may have negative e f f e c t s on personal well-being and/or marital q u a l i t y . Group discussion might f a c i l i t a t e exploration of t h i s issue. Participants could discuss why women tend to assign a low p r i o r i t y to t h e i r own needs. Couples could i d e n t i f y ways to increase time spent together. In the postnatal period, nurses must respect dual-earners' a l l o c a t i o n of r o l e r e s p o n s i b i l i t i e s . Couples may choose a t r a d i t i o n a l rather than e g a l i t a r i a n d i v i s i o n of labour. Although the t r a d i t i o n a l model i s inconsistent with p r e v a i l i n g feminist ideology, evidence suggests dual-earners are r e l a t i v e l y content with a t r a d i t i o n a l approach to the a l l o c a t i o n of r o l e a c t i v i t i e s . Nurses need to reinf o r c e the 127 importance of couples making decisions that f e e l r i g h t for them. Notwithstanding dual-earners' apparent s a t i s f a c t i o n with the t r a d i t i o n a l d i v i s i o n of labour, nurses should continue to promote "involved" fatherhood. Nurses can encourage fathers to become more invested i n c h i l d care by i n v i t i n g them to p a r t i c i p a t e i n prenatal classes and during labour and delivery. Nurses can also enhance men's parenting behaviours by teaching fathers about infant care. Nursing Education Due to changes i n practice requirements and the growing number of women seeking graduate degrees i n nursing, more and more multiple-role women are enroling i n nursing programs. Educators must be cognizant of the intense demands made on mothers who are also nursing students. Results from t h i s study can be used by educators to id e n t i f y dual-earners who may be challenged by the student r o l e . The findings indicate the ages of the children and the family s i z e have an impact on r o l e i n t e n s i t y and r o l e d i s p a r i t y for parents. Since women often assume primary r e s p o n s i b i l i t y for the care of infants and toddlers, mothers of very young children may experience intense r o l e demands. Educators should not assume the presence of a spouse i n the home means mothers w i l l be rel i e v e d of c h i l d care r e s p o n s i b i l i t i e s . 128 Parents with more than one c h i l d may be overtaxed by ro l e demands. Findings from t h i s study suggest parents with large families perceive that the in t e n s i t y of the spouse r o l e i s unacceptable and uncontrollable. By taking on the student r o l e , individuals may be even more hard pressed to maintain the qu a l i t y of t h e i r r o l e s . Nursing f a c u l t y must be prepared to a s s i s t nursing students who are parents. Counsellors should be aware of the tendency of mothers to put t h e i r own needs l a s t . As part of a comprehensive assessment, counsellors could determine whether the mother has s u f f i c i e n t time and energy to meet r o l e demands associated with the i n d i v i d u a l r o l e . Mothers who f a i l to take time for themselves may be at r i s k for r o l e s t r a i n . Students should be taught to appreciate the complexity inherent i n multiple-role l i f e s t y l e s . Moreover, understanding the demographic predictors of ro l e i n t e n s i t y and r o l e d i s p a r i t y may a s s i s t students to i d e n t i f y those couples who p o t e n t i a l l y f i n d i t d i f f i c u l t to combine parenthood with paid work. Nursing Administration Nursing administrators play an important part i n ensuring organizations respond to employees' needs. Since the majority of health care workers are women, administrators can f a c i l i t a t e women's integration of 129 paid-work and motherhood by lobbying for family-oriented personnel p o l i c i e s . Administrators could use the demographic predictors to i d e n t i f y groups of employees l i k e l y to experience high l e v e l s of r o l e i n t e n s i t y and ro l e d i s p a r i t y . Programs may be developed that address the needs of employees who are most vulnerable to ro l e stress. The study suggests mothers of very young children experience intense parent roles and may receive r e l a t i v e l y l i t t l e c h i l d care assistance from t h e i r husbands. Managers must be cognizant many of these women "put i n a second s h i f t " when they leave work (Hochschild, 1989). Often mothers a r r i v e home and spend the remainder of the day attending to domestic r e s p o n s i b i l i t i e s . I t would behoove managers to adopt a sensi t i v e attitude towards mothers of young children and understand the demands motherhood makes on time and energy reserves. Administrators should be aware d i s p a r i t y i n the ind i v i d u a l and spouse roles i s associated with couples who are rearing two or more children. These couples may be too busy to meet the demands of the in d i v i d u a l r o l e . Administrators may a s s i s t couples by providing on-site f i t n e s s and n u t r i t i o n classes. In addition, administrators could ensure marital counselling programs are avail a b l e to employees. 130 Administrators can be instrumental i n increasing men's p a r t i c i p a t i o n i n c h i l d care. H a l l (1992) suggested that "fathers need to be involved with t h e i r infant's care at the e a r l i e s t possible time" (p. 37). By providing paid leave for new fathers administrators can ensure fathers are allowed an opportunity to become more involved i n the parent r o l e . Nursing Research The study empirically validates theory. Researchers may f e e l confident about generating hypotheses from Hall ' s (1993) modification of Ward's (1986) t h e o r e t i c a l framework. The use of v e r i f i e d integrative frameworks i s heralded as an important prerequisite to understanding the r e l a t i o n s h i p between r o l e stress and ro l e outcomes. This study examined a highly educated and well-paid segment of the dual-earner population. I t would be in t e r e s t i n g to investigate the function of i n t e n s i t y and d i s p a r i t y i n a sample of the working poor or i n other c u l t u r a l groups. Substantial differences i n r o l e enactment may be found among occupational s t r a t a . Researchers should i d e n t i f y s t r u c t u r a l features of society that a f f e c t i n d i v i d u a l s ' perceptions of r o l e i n t e n s i t y and d i s p a r i t y . As postulated by Marks (1977), human energy i s s o c i a l l y constructed and may not be ju s t a function of time and e f f o r t . Human energy i s also directed 131 towards r o l e demands i n the context of employment and p o l i t i c a l and s o c i a l p o l i c i e s . Results from t h i s study suggest the t r a d i t i o n a l d i v i s i o n of labour i s firmly entrenched. Although the dual-earner l i f e s t y l e has been a fac t of s o c i e t a l l i f e f o r over twenty years, gender segregation of roles and tasks continues unabated for t h i s sample. Research could determine why gender r o l e norms continue to dominate i n some dual-earner f a m i l i e s . I t would be illuminating to f i n d out why gender ro l e s do not change more rapidl y . Chapter Summary The f i f t h chapter described conclusions derived from the psychometric t e s t i n g of the REQ. The implications for t o o l development and future t e s t i n g of the REQ were detai l e d . In addition, the chapter discussed the REQ's u t i l i t y for nursing. Conclusions related to hypothesis t e s t i n g were presented and the implications of these conclusions for nursing were described. Concluding Comments The goal of t h i s study was to evaluate the psychometric properties of the REQ. The dearth of r e l i a b l e and v a l i d instruments capable of measuring discrete c h a r a c t e r i s t i c s of roles necessitated the REQ's development. Although the study recommended revisions to the instrument, the REQ 132 promises to be a s i g n i f i c a n t advancement i n the f i e l d of dual-earner research. Gone are the days when ro l e occupancy served as a proxy for r o l e q u a l i t y . Today's investigators require an instrument that can quantify r o l e c h a r a c t e r i s t i c s deemed s a l i e n t to the r o l e outcomes of dual-earner men and women. Analysis of r o l e c h a r a c t e r i s t i c s w i l l enable researchers to better understand the rel a t i o n s h i p between roles and r o l e outcomes. The need for sound measuring devices i s greater than ever. As the proportion of dual-earner families r i s e s , employers and governments w i l l l i k e l y experience pressure to resolve work-family issues. Public p o l i c y decisions must be based on research findings obtained from psychometrically sound instruments. Thesis Summary In t h i s study, the author discussed the importance of a psychometrically t r i e d instrument capable of measuring r o l e q u a l i t y associated with dual-earners. The REQ was i d e n t i f i e d as a promising new t o o l . However, i t required further r e l i a b i l i t y and v a l i d i t y t e s t i n g . After a comprehensive review of the l i t e r a t u r e , a method for t e s t i n g the REQ was devised. Procedures and e t h i c a l concerns were i d e n t i f i e d and a description of the sample provided. The findings generated from the psychometric evaluation were 133 discussed i n the context of the l i t e r a t u r e . Pertinent conclusions, nursing implications, and recommendations for further t o o l development were presented. 134 References Abidin, R. (1980). Parent education and intervention handbook. S p r i n g f i e l d : Thomas. Amatea, E., & Fong, M. (1991). The impact of r o l e stressors and personal resources on the stress of professional women. Psychology of Women Quarterly. 15(3), 419-430. Anderson-Kulman, R., & Paludi, M. (1986). Working mothers and the family context: Predicting p o s i t i v e coping. Journal of Vocational Behaviour, 28, 241-253. Arber, S., Gi l b e r t , G., & Dale. (1985). Paid employment and women's health: A benefit or a source of r o l e strain? Sociology of Health and I l l n e s s . 7(3), 375-400. Avant, K. (1988). Stressors on the childbearing family. Journal of Obs t e t r i c a l Gynaecologic and Neonatal Nursing. 17. 179-185. Bacharach, S., & Bamberger, J. (1992). Causal models of ro l e stressor antecedents and consequences: The importance of occupational differences. Journal of Vocational Behaviour. 41, 13-34. Barling, J ., & Macewen, K. (1992). Linking work experiences to facets of marital functioning. Journal of Organizational Behaviour r 13(6), 573-583. 135 Barnett, R., Biener, L., & Baruch, G. (1987). So c i a l r o l e s , gender, and psychological d i s t r e s s . In R. Barnett, L. Biener, and G. Baruch (Eds.), Gender and stress (pp. 122-143). New York: The Free Press. Barnett, R., Brennan, R., & Marshall, N. (1994). Gender and the r e l a t i o n s h i p between parent r o l e q u a l i t y and psychological d i s t r e s s : A study of men and women i n dual-earner couples. Journal of Family Issues, 15(2), 229-252. Barnett, R., & Marshall, N. (1992). Worker and mother roles , s p i l l o v e r e f f e c t s and psychological d i s t r e s s . Women & Health. 18(2), 9-40. Barnett, R., Marshall, N., & Sayer, A. (1 992). P o s i t i v e - s p i l l o v e r e f f e c t s from job to home: A closer look. Women & Health. 19(2/3), 13-41. Barnett, R., Marshall, N., & Singer, J. (1992). Job experiences over time, multiple rol e s , and women's mental health: A longitudinal study. Journal of Personality and Social Psychology, 63_(4) , 634-644. Baruch, G., & Barnett, R. (1986). Consequences of fathers' p a r t i c i p a t i o n i n family work: Parents r o l e s t r a i n and well-being. Journal of Personality and Social Psychology. 51(5), 983-992. 136 Benin, M., & A g o s t i n e l l i , J . (1988). Husbands' and wives' s a t i s f a c t i o n with the d i v i s i o n of labour. Journal of Marriage and the Family, 50, 349-361. Berardo, D., Shehan, C , & L e s l i e , G. (1987). A residue of t r a d i t i o n : Jobs, careers, and spouses' time i n housework. Journal of Marriage and the Family. 49, 381-390. Berthold, J. (19 64). Theoretical and empirical c l a r i f i c a t i o n of concepts. Nursing Science. 2(5), 406-422. Bielby, D. (1992). Commitment to work and family. Annual Review of Sociology. 18, 281-302. B l a i r , S., & Johnson, M. (1992). Wives' perceptions of the fairness of the d i v i s i o n of household labour: The i n t e r s e c t i o n of housework and ideology. Journal of Marriage and the Family. 54, 570-581. Bolger, E., Delongis, A., Kessler, R., & Wethington, E. (1989). The micro structure of d a i l y r o l e - r e l a t e d stress i n married couples. In J. Eckenrode & S. Gore (Eds.), Stress between work and family (pp. 95-114). New York: Plenum. Brett, J., & Yogev, S. (1988). Restructuring work for family: How dual-earner couples with children manage. Journal of Social Behaviour and Personality. 3.(4) , 159-174. 137 Brown, F. (1970). P r i n c i p l e s of educational and psychological t e s t i n g . Hinsdale, IL: Dryden Press. Burke, R. (1988). Work-family c o n f l i c t . In E. Goldsmith (Ed.), Work and family: Theory, research, and applications (Special Issue). Journal of So c i a l Behaviour and Personality. 3.(4), 287-302. Burke, R., & McKeen, C. (1988). Work and family: What we know and what we need to know. Canadian Journal of Administrative Sciences. 5(4), 30-40. Burr, W. (1976). Symbolic in t e r a c t i o n and the family. In R. Burr, R. H i l l , F. Nye, & I. Reiss (Eds.), Contemporary theories about the family (Vol. I I , pp. 42-112). New York: Free Press. Campbell, M., & Moen, P. (1992). Job-family r o l e s t r a i n among employed mothers of preschoolers. Family Relations, 41, 205-211. Chassin, L., Zeiss, A., Cooper, K., & Reaven, J. (1985). Role perceptions, s e l f - r o l e congruence and marital s a t i s f a c t i o n i n dual-worker couples with preschool children. Social Psychology Quarterly. 48 (4) , 301-311. Cleary, P., & Mechanic, D. (1983). Sex differences i n psychological d i s t r e s s among married people. Journal of Health and Social Behaviour, 24., 111-121. 138 Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived stress. Journal of Health and S o c i a l Behaviour, 24, 385-396. Coverman, S. (1989). Role overload, r o l e c o n f l i c t , and stress: Addressing consequences of multiple r o l e demands. Social Forces. J57 (4) , 965-982. Cowan, C , & Cowan, P. (1988). Who does what when partners become parents: Implications for men, women, and marriage. Marriage & Family Review. 12 (3/4), 105-130. Crocker, L., & Algina, J. (1987). Introduction to c l a s s i c a l and modern tes t theory. Harcourt Brace Jovanovich College: Fort Worth. Crnic, K., & Booth, C. (1991). Mother's and father's perceptions of d a i l y hassles of parenting across early childhood. Journal of Marriage and the Family. 53, 1042-1050. Crompton, S. (1991, Summer). Who's looking a f t e r the kids? Perspectives. 68-76. Crouter, A., Perry-Jenkins, M., Huston, T., & McHale, S. (1987). Processes underlying father involvement i n dual-earner and single-earner f a m i l i e s . Developmental Psychology. 23(3), 431-440. 139 Darling-Fisher, C , & Tiedje, L. (1990). The impact of maternal employment c h a r a c t e r i s t i c s on fathers' p a r t i c i p a t i o n i n c h i l d care. Family Relations, 1, 20-25. Duxbury, L., Higgins, C., & Lee, C. (1994). Work-family c o n f l i c t . Journal of Family Issues. 15(3), 449-466. Elman, M., & Gi l b e r t , L. (1984). Coping strategies for ro l e c o n f l i c t i n married professional women with children. Family Relations. 33, 317-327. Fish, L., New, R., & Van Cleave, N. (1992). Shared parenting i n dual-income families. American Journal of Orthopsychiatry. 67(1), 83-92. Fisher, C., & Gitelson, R. (1983). A meta-analysis of the correlates of r o l e c o n f l i c t and ambiguity. Journal of Applied Psychology. 68.(2), 320-333. Froberg, D., Gjerdngen, D., & Preston, M. (1986). Multiple roles and women's mental and physical health: What have we learned? Women & Health. 1JL(2) , 79-96. Frone, M., Russell, M., & Cooper, M. (1991). Prevalence of work-family c o n f l i c t : Are work and family boundaries asymmetrically permeable? Journal of Organizational Behaviour. 12(7), 723-729. Ghalam, N. (1993/Spring). Women i n the workplace. Canadian Social Trends. 1-5. 140 G i l l i g a n , C. (1982). In a d i f f e r e n t voice. Cambridge, MA: Harvard University Press. Geerken, M., & Gove, W. (1983). At home and at work: The family's a l l o c a t i o n of labor. Beverly H i l l s : Sage. Glendinning, C., & M i l l a r , J. (1992). I t a l l r e a l l y s t a r t s i n the family: Gender d i v i s i o n s and poverty. In C. Glendinning and J. M i l l a r (Eds.), Poverty i n B r i t a i n i n the 1990s, (pp. 3-10). New York: Harvester Wheatsheaf. Goode, W. (1960). A theory of ro l e s t r a i n . American So c i o l o g i c a l Review. 25, 483-496. Googins, B. (1991). Work/family c o n f l i c t s : Private l i v e s — p u b l i c responses. New York: Auburn House. Googins, B., & Burden, D. (1987, July/August). V u l n e r a b i l i t y of working parents: Balancing work and home ro l e s . Social Work. 295-299. Gorsuch, R. (1974). Factor analysis. London: Saunders. Gove, W. , & Geerken, M. (1977). The e f f e c t of children and employment on the mental health of married men and women. Social Forces. 56, 66-76. Greenberger, E., & O'Neil, R. (1993). Spouse, parent, worker: Role commitments and ro l e - r e l a t e d experiences i n the construction of well-being. Developmental Psychology. 29, 181-197. 141 Greenglass, E., Pantony, K. , & Burke, R. (1988). A gender-role perspective on ro l e c o n f l i c t , work stress and s o c i a l support. Journal of So c i a l Behaviour and Personality. 2(4), 317-328. Greenhaus, J . (1988). The interse c t i o n of work and family r o l e s : Individual, interpersonal, and organizational issues. Journal of Social Behaviour and Personality. 2(4), 23-44. Greenhaus, J., Bedeian, A., & Mossholder, K. (1987). Work experiences, job performance, and feelings of personal and family well-being. Journal of Vocational Behaviour, 2 1(2), 200-215. Greenhaus, J., & Beutell, N. (1985). Sources of c o n f l i c t between work and family r o l e s . Academy of Management Review. 10(1), 76-88. Greenhaus, J., & Parasuraman, S. (1987). A work-non-work i n t e r a c t i v e perspective of stress and i t s consequences. Journal of Organizational Behavioral Management, 8(2), 37-59. Grimm-Thomas, K., & Perry-Jenkins, M. (1994, A p r i l ) . A l l i n a day's work: Job experiences, self-esteem, and fathering i n working-class f a m i l i e s . Family Relations. 175-181. 142 Grossman, F. (1988). Strain i n the t r a n s i t i o n to parenthood. Marriage and Family Review. 12.(3/4) , 85-103. Guelzow, M., Bird, G., & Koball, E. (1991). An exploratory path analysis of the stress process for dual-career men and women. Journal of Marriage and the Family. 53(1), 151-164. Gunter, B., & Gunter, N. (1991). Inequities i n household labour: Sex ro l e orientation and the need for cleanliness and r e s p o n s i b i l i t y as predictors. Journal of Social Behaviour and Personality. 6.(3), 559-572. H a l l , L. (1990). Prevalence and correlates of depressive symptoms i n mothers of young children. Public Health Nursing. 7(2), 71-79. H a l l , W. (1987). The experience of women returning to work following the b i r t h of t h e i r f i r s t c h i l d . Midwifery. 3, 187-195. H a l l , W. (1991). The experience of fathers i n dual-earner families following the b i r t h of t h e i r f i r s t infant. Journal of Advanced Nursing. 16(2), 423-430. H a l l , W. (1992). Comparison of the experience of women and men i n dual-earner families following the b i r t h of t h e i r f i r s t infant. IMAGE: Journal of Nursing Scholarship. 24(1)/ 33-38. 143 H a l l , W. (1993). Development and early t e s t i n g of the Role Enactment Questionnaire. Canadian Journal of Nursing Research. 25(4), 57-65. Hanson, S. (1985). Fatherhood: Contextual v a r i a t i o n s . American Behavioral S c i e n t i s t . 29.(1), 55-77. Hanson, S., & Bozett, F. (1987). Fatherhood and changing family r o l e s . Family Community Health. 9(4) , 9-21. Hibbard, J., & Pope, C. (1985). Employment status, employment c h a r a c t e r i s t i c s , and women's health. Women & Health. 10(1), 59-77. Higgins, C., Duxbury, L., & Lee, C. (1994). Impact of l i f e - c y c l e state and gender on the a b i l i t y to balance work and family r e s p o n s i b i l i t i e s . Family Relations. 43, 144-150. Hirsch, B., & Rapkin, B. (1986). Multiple ro l e s , s o c i a l networks, and women's well-being. Journal of Personality and Social Psychology, 51(6), 1237-1247. Hochschild, A. (1989). The second s h i f t : Working parents and the revolution at home. New York: Viking Press. Hughes, D., Galinsky, E., & Morris, A. (1992). The e f f e c t s of job c h a r a c t e r i s t i c s on marital q u a l i t y : Specifying l i n k i n g mechanisms. Journal of Marriage and the Family. 54. 31-42. Ivancevich, J., & Matteson, M. (1980). Stress and work. I l l i n o i s : Scott, Foresman. 144 Jacobson, D. (1989). Context and the s o c i o l o g i c a l study of stress: An i n v i t e d response to P e a r l i n (Vol. 30, No. 3). Journal of Health and Social Behaviour. 30. 257-260. Johnson, J . (1989). Female c l e r i c a l workers' perceived work and nonwork stress and d i s s a t i s f a c t i o n as predictors of psychological d i s t r e s s . Women & Health. Ij5(4), 61-76. Johnson, C., & Johnson, F. (1980). Parenthood, marriage, and careers: S i t u a t i o n a l constraints and r o l e s t r a i n . In F. Pepitone-Rockwell (Ed.), Dual-career Couples (pp. 143-161). Beverly H i l l s : Sage. Johnson, R., & Wichern, D. (1992). Applied multivariate s t a t i s t i c a l analysis. Englewood C l i f f s : Prentice-Hall. Jones, M. (1993) Role c o n f l i c t : Cause of burnout or energizer. Social Work. 38.(2) , 136-141. Kach, J., & McGhee, P. (1982). Adjustment of early parenthood: The r o l e of accuracy of preparenthood experiences. Journal of Family Issues. 3.(3), 375-388. Kandel, D., Davies, M., & Raveis, V. (1985). The stressfulness of d a i l y roles for women: Mar i t a l , occupational and household ro l e s . Journal of Health and Social Behaviour. 26, 64-78. 145 Karasek, R. (1979). Job demands, job decision l a t i t u d e , and mental s t r a i n : Implications for job redesign. Administrative Science Quarterly. 24. 285-308. Kelly, R., & Voydanoff, P. (1986). Work/family r o l e s t r a i n among employed parents. Family Relations. 34(3) , 367-374. Kessler, R., & McRae, J. (1982). The e f f e c t s of wives' employment on the mental health of men and women. American S o c i o l o g i c a l Review. 47., 216-227. Kingston, P., & Nock, S. (1987). Time together among dual-earner couples. American S o c i o l o g i c a l Review. 52., 391-400. Lengacher, C. (1993). Development and study of an instrument to measure r o l e s t r a i n . Journal of Nursing Education. 32,(2), 71-77. Lennon, M., Wasserman, G., & Allen, R. (1991). Infant care and wives' depressive symptoms. Women & Health. 17(2), 1-23. L e s l i e , L., & Anderson, E. (1988). Men's and women's p a r t i c i p a t i o n i n domestic r o l e s : Impact on qu a l i t y of l i f e and marital adjustment. Journal of Family Psychology, 2.(2), 212-226. 146 Levanthal-Belfer, L., Cowan, P., & Cowan, C. (1992). S a t i s f a c t i o n with c h i l d care arrangements: E f f e c t s on adaptation to parenthood. American Journal of Orthopsychiatry, 62(2), 165-177. LoBiondo-Wood, G., & Haber, J. (1986). Nursing Research. Toronto: Mosby. Loerch, K., Russell, J ., & Rush, M. (1989). The relationships among family domain variables and work-family c o n f l i c t for men and women. Journal of Vocational Behaviour. 3J5, 288-308. Marat, E., & Finlay, B. (1984). The d i s t r i b u t i o n of household labor among women i n dual-earner f a m i l i e s . Journal of Marriage and the Family. 12, 357-364. Marks, S. (1977). Multiple roles and r o l e s t r a i n : Some notes on human energy, time and commitment. American Soc i o l o g i c a l Review. 42., 921-936. Marshall, N., & Barnett, R. (1993). Work-family s t r a i n s and gains among two-earner couples. Journal of Community Psychology. 21(1), 64-78. Marshall, N., & Barnett, R. (1991). Race, class, and multiple r o l e strains and gains among women employed i n the service sector. Women & Health. 17(4), 1-19. Matthews, K., & Rodin, J. (1989). Women's changing work ro l e s : Impact on health, family, and public p o l i c y . American Psychologist, 44.(11), 1389-1393. 147 McBride, A. (1990). Mental health e f f e c t s of women's multiple r o l e s . American Psychologist, 4.5(3), 381-384. McLaughlin, M., Cormier, L., & Cormier, W. (1988). Relation between coping strategies and d i s t r e s s , stress, and marital adjustment of multiple-role women. Journal of Counselling Psychology. 25(2), 187-193. Meleis, A., & Stevens, P. (1992). Women i n c l e r i c a l jobs: Spousal r o l e s a t i s f a c t i o n , stress, and coping. Women & Health. 18(1), 23-40. Michelson, W. (1985). From sun to sun. Totowa, NJ: Rowman & Allanheld. Moen, P., & Dempster-McClain, D. (1987). Employed parents: Role s t r a i n , work time, and preferences for working le s s . Journal of Marriage and the Family. 49., 579-590. Nock, S., & Kingston, P. (1982). The family l i f e cycles approach to family analysis. In B.B. Wolman and G.Strickler (Eds.), Handbook of developmental psychology. Englewood C l i f f s , NJ: Prentice-Hall. Nunnally, J. (1978). Psychometric theory. McGraw-Hill: New York. Olds, J., Schwartz, R., Eisen, S., Betcher, R., & Van N i e l , A. (1993). Part-time employment and marital well-being: A hypothesis and p i l o t study. Family Therapy. 20(1), 1-16. 148 Parasuraman, S., Greenhaus, J. , & Granrose, C. (1992). Role stressors, s o c i a l support, and well-being among two-career couples. Journal of Organizational Behaviour. 13(4), 339-356. Parry, G., & Warr, P. (1980). The measurement of mothers 1 work attitudes. Journal of Occupational Psychology. 53. 245-252. Pearlin, L. (1989). The s o c i o l o g i c a l study of stress. Journal of Health and Social Behaviour. 30, 241-256. Pearlin, L. (1975). Sex-roles and depression. In N. Datan and L. Ginsberg (Eds.), Life-span developmental psychology. New York: Academic Press. Pearli n , L., & Schooler, C. (1978). The structure of coping. Journal of Health and Social Behaviour. 19, 2-21. Piechowski, L. (1992). Mental health and women's multiple r o l e s . Families i n Society: The Journal of Contemporary Human Services f 73_(3) , 131-139. Pietromonaco, P., Manis, J., & Frohardt-Lane, K. (1986). The psychological consequences of multiple s o c i a l r o l e s . Psychology of Women Quarterly. 10. 373-382. 149 Pittman, J . , & Orthner, D. (1988). Gender differences i n the prediction of job commitment. In E. Goldsmith (Ed.), Work and family: Theory, research, and applications (Special issue). Journal of Soc i a l Behaviour and Personality. 2(4), 227-248. Pleck, J . (1985). The consequences of ro l e overload. In J. Pleck (Ed.), Working wives/working husbands (pp. 97-117). Sage: Beverly H i l l s . Pleck, J. (1976). The work-family r o l e system. Social Problems. 24, 417-425. P o l i t , D., & Hungler, B. (1991). Nursing research: P r i n c i p l e s and methods (4th ed.). Toronto: Lippincott. Power, T, & Parke, R. (1984). Social network factors and the t r a n s i t i o n to parenthood. Sex Roles. 10(11/12), 949-972. Presser, H. (1988). S h i f t work and c h i l d care among young dual-earner American parents. Journal of Marriage and the Family. 50, 133-148. P u g l i e s i , K. (1989). Social support and self-esteem as intervening variables i n the rel a t i o n s h i p between s o c i a l r o l e s and women's well-being. Community Mental Health Journal. 25(2), 87-97. P u g l i e s i , K. (1988). Employment c h a r a c t e r i s t i c s , s o c i a l support and the well-being of women. Women & Health. 14(1), 35-57. 150 Reifman, A., Biernat, M., & Lang, E. (1991). Stress, s o c i a l support, and health i n married professional women with small children. Psychology of Women Quarterly. 15(3), 431-445. Repetti, R., Matthews, K., & Waldron, I. (1989). Employment and women's health: E f f e c t s of paid employment on women's mental and physical health. American Psychologist, 44 , 1394-1401. Rice, R., Frone, M., & McFarlin, D. (1992). Work-nonwork c o n f l i c t and the perceived q u a l i t y of l i f e . Journal of Organizational Behaviour. (13.), 155-168. Rosenfield, S. (1989). The ef f e c t s of women's employment: Personal control and sex differences i n mental health. Journal of Health and Social Behaviour. 30, 77-91. Ross, C., & Mirowsky, J. (1988). Child care and emotional adjustment to wives employment. Journal of Health and Social Behaviour. 29, 127-138. Rotundo, E. (1985). American fatherhood: A h i s t o r i c a l perspective. American Behavioral S c i e n t i s t . 29.(1) , 7-25. Rudd, N., & McKenry, P. (1986). Family influences on the job s a t i s f a c t i o n of employed mothers. Psychology of Women Quarterly. 10, 363-372. 151 Sarbin, T., & Allen, V. (1968). Role theory. In G. Lindzey & E. Aronson (Eds.), The handbook of s o c i a l psychology: Volume 1 (2nd ed.), (pp. 488-567). Reading, MA: Addison-Wesley. Scanzoni, J. (1978). Sex roles, women's work. and marital c o n f l i c t : A study of family change. Lexington, MA: Lexington Books. Scarr, S., P h i l l i p s , D., & McCartney, K. (1989). Working mothers and t h e i r f a m i l i e s . American Psychologist. 44(H)/ 1402-1409. Schnittger, M., & Bird, G. (1990). Coping among dual-career men and women across the family l i f e cycle. Family Relations. 39(2), 199-205. Schroeder, K., Blood, L., & Maluso, D. (1992). An intergenerational analysis of expectations for women's career and family r o l e s . Sex Roles, 7/8, 273-291. Scott, J . , & Alwin, D. (1989). Gender differences i n parental s t r a i n . Journal of Family Issues. JL0(4), 482-503. Sears, H., & Galambos, N. (1992). Women's work conditions and marital adjustment i n two-earner couples: A st r u c t u r a l model. Journal of Marriage and the Family. 54/ 789-797. 152 Sekaran, U. (1985). The paths to mental health: An explanatory study of husbands and wives i n dual-career fa m i l i e s . Journal of Occupational Psychology, 58, 129-137. Shehan, C. (1984). Wives' work and psychological w e l l -being: An extension of Gove's Social Role Theory of depression. Sex Roles. 11. 881-899. Smith, D. (1985). Wife employment and marital adjustment: A cumulation of r e s u l t s . Family Relations. 34., 483-490. Smith, R. (1979). Hours r i g i d i t y : E f f e c t s on the labour-market status of women. In K. Feinstein (Ed.), Working women and families (pp. 211-218). Beverly H i l l s : Sage. Spitze, G. (1988). Women's employment and family r e l a t i o n s : A review. Journal of Marriage and the Family. (50), 595-618. Staines, G., & Pleck, J. (1986). Work schedule f l e x i b i l i t y and family l i f e . Journal of Occupational Behaviour. 7, 147-153. Suitor, J. (1991). Marital q u a l i t y and s a t i s f a c t i o n with the d i v i s i o n of household labour across the family l i f e cycle. Journal of Marriage and the family. 53.(1) , 221-230. 153 Tabachnick, B., & F i d e l l , L. (1989). Using multivariate s t a t i s t i c s . New York: Harper C o l l i n s . Thoits, P. (1983). Multiple i d e n t i t i e s and psychological well-being: A reformulation and te s t of the s o c i a l i s o l a t i o n hypothesis. American S o c i o l o g i c a l Review. 48(2), 174-187. Ulbrich, P. (1988). The determinants of depression i n two-income marriages. Journal of Marriage and the Family. 50(1), 121-131. Umberson, D., & Gove, W. (1989). Parenthood and psychological well-being. Journal of Family Issues. 10(4), 440-462. Vannoy, D., & P h i l l i b e r , W. (1992). Wife's employment and q u a l i t y of marriage. Journal of Marriage and the Family. 54, 387-398. Ventura, J. (1987). The stresses of parenthood reexamined. Family r e l a t i o n s . 3.6(1) , 26-29. Verbrugge, L. (1986). Role burdens and physical health of women and men. Women & Health. 11(1), 47-77. Verhoef, M., Love, E., & Rose, M. (1992). Women's s o c i a l r o l e s and t h e i r exercise p a r t i c i p a t i o n . Women & Health. 19(4), 15-29. Voydanoff, P. (1988). Work ro l e c h a r a c t e r i s t i c s , family structure demands, and work/family c o n f l i c t . Journal of Marriage and the Family. 50, 749-761. 154 Waldron, I., & Jacobs, J. (1989). Ef f e c t s of multiple r o l e s on women's health — Evidence from a national longitudinal study. Women & Health. 15(1), 3-19. Walker, L., & Best, M. (1991). Well-being of mothers with infant children: A preliminary comparison of employed women and homemakers. Women & Health. 17(1), 71-89. Waltz, C. & Bausell, R. (1981). Nursing research: Design, s t a t i s t i c s and computer analysis. Philadelphia: F.A. Davis. Waltz, C., Strickland, 0., & Lenz, E. (1984). Measurement i n nursing research. Philadelphia: F.A. Davis. Ward, C. (1986). The meaning of rol e s t r a i n . Advances i n Nursing Science. 8, 39-48. Weinert, C , & Tilden, V. (1990). Measures of s o c i a l support: Assessment of v a l i d i t y . Nursing Research. 39(4), 212-217. Wethington, E., & Kessler, R. (1986). Perceived support, received support, and adjustment to s t r e s s f u l l i f e events. Journal of Health and Social Behaviour. 27., 78-89. Wilson, H. (1987). Introducing research i n nursing. Don M i l l s : Addison-Wesley. 155 Woods, N. (1985). Employment, family ro l e s , and mental i l l health i n young married women. Nursing Research, 34(1), 4-10. 156 APPENDIX A The Role Enactment Questionnaire Instructions This questionnaire consists of a series of statements representing various behaviours which may or may not be a part of your r o l e as a parent, a worker, a spouse, or an i n d i v i d u a l . Following each statement are three questions. To answer each question, please c i r c l e the number which best represents what i s happening at t h i s point i n your l i f e . For example: 1 = none 2 = a l i t t l e 3 = somewhat 4 = a f a i r amount 5 = a great deal none 1 a great deal 4 5 The amount of time you spend answering research questionnaires. 1 a) How acceptable i s t h i s to you? b) How much do you control this? 1 1 2 2 4 4 4 These responses would indicate that you answer research questionnaires infrequently; that answering them i s somewhat acceptable to you; and that you are responsible for deciding whether or not you w i l l answer them. 157 1 = none 2 = a l i t t l e 3 = somewhat 4 = a f a i r amount 5 = a great deal YOUR ROLE AS A PAID WORKER In t h i s section, you are required to answer questions about your r o l e as a paid worker. A great None deal 1. The amount of overtime you work. 1 2 3 4 5 (a) How acceptable i s t h i s to you? 1 2 3 4 5 (b) How much do you control this? 1 2 3 4 5 2. The amount of time you miss paid work. 1 2 3 4 5 (a) How acceptable i s t h i s to you? 1 2 3 4 5 (b) How much do you control this? 1 2 3 4 5 3. The amount of time you are required to spend i n work-related education. 1 2 3 4 5 (a) How acceptable i s t h i s to you? 1 2 3 4 5 (b) How much do you control this? 1 2 3 4 5 4. The amount your paid work requires you to t r a v e l away from home. 1 2 3 4 5 (a) How acceptable i s t h i s to you? 1 2 3 4 5 (b) How much do you control this? 1 2 3 4 5 5. The amount of commuting required for your paid work. 1 2 3 4 5 (a) How acceptable i s t h i s to you? 1 2 3 4 5 (b) How much do you control this? 1 2 3 4 5 6. The paid work you do beyond your regular employment. 1 2 3 4 5 (a) How acceptable i s t h i s to you? 1 2 3 4 5 (b) How much do you control this? 1 2 3 4 5 7. The adequacy of your combined incomes to meet your family needs. 1 2 3 4 5 (a) How acceptable i s t h i s to you? 1 2 3 4 5 (b) How much do you control this? 1 2 3 4 5 158 1 = none 2 = a l i t t l e 3 = somewhat 4 = a f a i r amount 5 = a great deal YOUR ROLE AS A PARTNER In t h i s section, you are required to answer questions that deal with your r o l e as a spouse. A great deal The amount of time you spend t a l k i n g with your spouse. (a) How acceptable i s t h i s to you? (b) How much do you control this? None 1 1 1 2 2 2 3 3 3 4 4 4 5 5 5 9. The amount of attention you receive from your spouse. (a) How acceptable i s t h i s to you? (b) How much do you control this? 10. The c o n f l i c t you experience with your spouse. (a) How acceptable i s t h i s to you? (b) How much do you control this? 1 1 1 1 1 1 2 2 2 2 2 2 3 3 3 3 3 3 4 4 4 4 4 4 5 5 5 5 5 5 11. The amount that you are alone with your spouse for fun a c t i v i t i e s . 1 2 3 4 5 (a) How acceptable i s t h i s to you? 1 2 3 4 5 (b) How much do you control this? 1 2 3 4 5 FOR EXAMPLE: The amount you r e f i n i s h f u r n i t u r e . (a) How acceptable i s t h i s to you? 1 2 3 4 (b) How much r e s p o n s i b i l i t y do you take f o r this? 1 2 3 4 These responses would indicate that you never r e f i n i s h your own furniture, that not r e f i n i s h i n g your furniture i s acceptable to you, and that you make the arrangements for a firm to pick up your furniture and r e f i n i s h i t . 12. The amount you prepare meals. 1 2 3 4 5 (a) How acceptable i s t h i s to you? 1 2 3 4 5 (b) How much r e s p o n s i b i l i t y do you take for this? 1 2 3 4 5 159 1 — none 2 — a l i t t l e 3 — somewhat 4 — a f a i r amount 5 — a great deal 13. The amount you do dishes. (a) How acceptable i s t h i s to you? (b) How much r e s p o n s i b i l i t y do you take for this? None 1 2 2 3 3 A great deal 4 5 4 4 5 5 14. The amount you vacuum/sweep/mop. (a) How acceptable i s t h i s to you? (b) How much r e s p o n s i b i l i t y do you take for this? 1 1 2 2 4 4 5 5 15. The amount you do laundry. (a) How acceptable i s t h i s to you? (b) How much r e s p o n s i b i l i t y do you take for this? 1 1 2 2 3 3 4 4 5 5 16. The amount you t i d y the house. (a) How acceptable i s t h i s to you? (b) How much r e s p o n s i b i l i t y do you take for this? 1 1 2 2 3 3 4 4 5 5 17. The amount you clean the bathroom. (a) How acceptable i s t h i s to you? (b) How much r e s p o n s i b i l i t y do you take for this? 1 1 2 2 3 3 4 4 5 5 18. The amount you handle the disposal of garbage. (a) How acceptable i s t h i s to you? (b) How much r e s p o n s i b i l i t y do you take for this? 1 1 2 2 3 3 4 4 5 5 19. The amount you do yard maintenance. (a) How acceptable i s t h i s to you? (b) How much r e s p o n s i b i l i t y do you take for this? 1 1 2 2 3 3 4 4 5 5 20. The amount you do house repairs and renovations. (a) How acceptable i s t h i s to you? (b) How much r e s p o n s i b i l i t y do you take for this? 1 1 2 2 4 4 5 5 160 1 = none 2 = a l i t t l e 3 = somewhat 4 = a f a i r amount 5 = a great deal A great None deal 21. The amount you p a r t i c i p a t e i n car maintenance. 1 2 3 4 5 (a) How acceptable i s t h i s to you? 1 2 3 4 5 (b) How much r e s p o n s i b i l i t y do you take for this? 1 2 3 4 5 22. The amount you shop for groceries or household items. 1 2 3 4 5 (a) How acceptable i s t h i s to you? 1 2 3 4 5 (b) How much r e s p o n s i b i l i t y do you take for this? 1 2 3 4 5 YOUR ROLE AS A PARENT This section examines your r o l e as parent. 23. The amount you bathe your c h i l d ( r e n ) . 1 2 3 4 5 (a) How acceptable i s t h i s to you? 1 2 3 4 5 (b) How much r e s p o n s i b i l i t y do you take for th i s ? 1 2 3 4 5 24. The amount you p a r t i c i p a t e i n feeding your c h i l d ( r e n ) . 1 2 3 4 5 (a) How acceptable i s t h i s to you? 1 2 3 4 5 (b) How much r e s p o n s i b i l i t y do you take for this? 1 2 3 4 5 25. The amount you change your child(ren)'s diapers. 1 2 3 4 5 (a) How acceptable i s t h i s to you? 1 2 3 4 5 (b) How much r e s p o n s i b i l i t y do you take for this? 1 2 3 4 5 26. The amount you dress your c h i l d ( r e n ) . 1 2 3 4 5 (a) How acceptable i s t h i s to you? 1 2 3 4 5 (b) How much r e s p o n s i b i l i t y do you take for this? 1 2 3 4 5 161 1 = none 2 = a l i t t l e 3 = somewhat 4 = a f a i r amount 5 = a great deal 27. The amount of time you spend playing with your c h i l d ( r e n ) . (a) How acceptable i s t h i s to you? (b) How much r e s p o n s i b i l i t y do you take for this? None 1 1 2 2 3 3 A great deal 4 4 5 5 28. The amount that you s e t t l e your child(ren) i n bed at night. (a) How acceptable i s t h i s to you? (b) How much r e s p o n s i b i l i t y do you take for this? 1 1 2 2 3 3 4 4 5 5 29. The amount that you pa r t i c i p a t e i n organizing c h i l d care (daycare, nannies, babysitting). 1 2 3 4 5 (a) How acceptable i s t h i s to you? 1 2 3 4 5 (b) How much r e s p o n s i b i l i t y do you take for this? 1 2 3 4 5 30. The amount of time that you spend transporting your child(ren) to childcare or school. 1 2 3 4 5 (a) How acceptable i s t h i s to you? 1 2 3 4 5 (b) How much r e s p o n s i b i l i t y do you take for this? 1 2 3 4 5 31. The amount of sick care you give your c h i l d ( r e n ) . 1 (a) How acceptable i s t h i s to you? 1 (b) How much r e s p o n s i b i l i t y do you take for this? 1 2 2 3 3 4 4 5 5 32. Your p a r t i c i p a t i o n i n organizing your child(ren)'s health care. 1 (a) How acceptable i s t h i s to you? 1 (b) How much r e s p o n s i b i l i t y do you take for this? 1 2 2 3 3 4 4 5 5 162 1 = none 2 = a l i t t l e 3 = somewhat 4 = a f a i r amount 5 = a great deal 33. Your p a r t i c i p a t i o n i n d i s c i p l i n i n g your c h i l d ( r e n ) . (a) How acceptable i s t h i s to you? (b) How much r e s p o n s i b i l i t y do you take for this? None 1 l 2 2 3 3 A great deal 4 4 5 5 34. Your p a r t i c i p a t i o n i n long-term planning for your c h i l d ( r e n ) . (a) How acceptable i s t h i s to you? (b) How much r e s p o n s i b i l i t y do you take for this? 1 1 2 2 3 3 4 4 5 5 35. The amount you plan family a c t i v i t i e s , (a) How acceptable i s t h i s to you? (b) How much r e s p o n s i b i l i t y do you take for this? 1 1 2 2 3 3 4 4 5 5 36. The amount you handle the safety concerns for your child(ren) (child-proofing your home and car). (a) How acceptable i s t h i s to you? (b) How much r e s p o n s i b i l i t y do you take for this? 1 1 1 2 2 2 3 3 3 4 4 4 5 5 5 YOUR ROLE AS AN INDIVIDUAL This section w i l l examine a c t i v i t i e s that may be part of your ro l e as an i n d i v i d u a l . 37, 38 39, Your involvement i n personal friendships. (a) How acceptable i s t h i s to you? (b) How much do you control this? Your involvement with r e l a t i v e s . (a) How acceptable i s t h i s to you? (b) How much do you control this? 1 1 1 1 1 1 The amount you p a r t i c i p a t e i n exercise a c t i v i t i e s . 1 (a) How acceptable i s t h i s to you? 1 (b) How much do you control this? 1 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 4 4 4 4 4 4 4 4 4 5 5 5 5 5 5 5 5 5 163 1 = none 2 = a l i t t l e 3 = somewhat 4 = a f a i r amount 5 = a great deal 40. Your involvement i n personal projects (e.g., hobbies, night classes). (a) How acceptable i s t h i s to you? (b) How much do you control this? 41. Your personal l e i s u r e a c t i v i t i e s (reading, watching t e l e v i s i o n ) . (a) How acceptable i s t h i s to you? (b) How much do you control this? 42. The amount you spend planning and attending health care appointments. (a) How acceptable i s t h i s to you? (b) How much do you control this? None l 1 1 l 1 l A great deal 1 1 1 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 4 4 4 4 4 4 4 4 4 DEMOGRAPHIC DATA 1. What i s your age? 2. What i s your sex? Male (1) Female (2) 3. How many children do you have l i v i n g f u l l - t i m e i n your household? 4. How many children do you have l i v i n g with you part-time? 5. What ages are these children? F i r s t born c h i l d : Second born c h i l d : Third born c h i l d : Fourth born c h i l d : F i f t h born c h i l d : Sixth born c h i l d : 6. What i s you highest l e v e l of education? 1. Some high school 2. High school completed 3. Some college 4. College completed 5. Diploma course completed 6, 7, 5 5 5 5 5 5 5 5 5 Some un i v e r s i t y course ( Completed u n i v e r s i t y with; , Bachelor's degree ( Master's degree Ph.D Medical degree 7. Number of hours of paid work per week. 164 8. Is flex-time available to you? Yes (1) No (2) 9. What i s your personal income level? (Does not include spouse's income). 1. $10,000 - $19,999 ( ) 4. $40,000 - $49,000 ( ) 2. $20,000 - $29,000 ( ) 5. more than $50,000 ( ) 3. $30,000 - $39,000 ( ) 10. Please l i s t the additional help you have at home ( i . e . , nannies, regular babysitting, cleaning and housekeepers, gardeners, r e l a t i v e s who a s s i s t you). 1. 4. 2. 5. 3. 6. 165 APPENDIX B Cover Letter The Role Enactment Questionnaire: R e l i a b i l i t y and v a l i d i t y t e s t i n g . PRINCIPAL INVESTIGATOR: Wendy H a l l , RN, MSN, Assistant Professor of Nursing. CO-INVESTIGATOR: Kenna Sleigh, RN, Master of Science of Nursing candidate. Dear dual-earner parent, My name i s Kenna Sleigh. I am a Registered Nurse who i s currently enrolled at the University of B r i t i s h Columbia (UBC) i n the Master of Nursing program. As part of my educational requirements I am conducting a research study. The purpose of the study i s to see whether or not a new questionnaire, the Role Enactment Questionnaire, can accurately measure the stress that dual-earner couples experience. The questionnaire was designed by my study supervisor Wendy H a l l , who i s an Assistant Professor at the UBC School of Nursing. As you are undoubtedly aware, r a i s i n g a family and working outside the home i s a challenging experience. Many dual-earner couples with children report f e e l i n g stressed out sometimes. However, i n order to f i n d ways to prevent or reduce stress, researchers need to be able to s c i e n t i f i c a l l y measure the amount of stress people experience. We also need to make sure that questionnaires l i k e the Role Enactment Questionnaire are accurate. In order to evaluate the Role Enactment Questionnaire, I need your assistance. You would f i l l out the questionnaire on two separate occasions approximately 2 weeks apart. The questionnaire i s 6 pages long and w i l l take 1 5 - 2 0 minutes of your time to complete. Thus, the t o t a l time necessary to p a r t i c i p a t e i s 40 minutes over a number of weeks. If you f e e l you would l i k e to pa r t i c i p a t e i n t h i s study, please do the following: 1. Both you and your partner should f i l l out copy #1 of The Role Enactment Questionnaire and the accompanying demographic data sheet. Please answer the questions with your a c t i v i t i e s i n mind. Don't compare your responses with those of your partner. 166 2. Place the two completed copies of the Role Enactment Questionnaire and the demographic sheets i n the self-addressed stamped envelope and mail to me. 3. Two weeks l a t e r you w i l l receive copy #2 of the Role Enactment Questionnaire. 4. Both you and your partner should f i l l out copy #2 of the Role Enactment Questionnaire and mail both forms back to me i n the self-addressed, stamped, envelope. 5. When I have received both completed copies of the Role Enactment Questionnaire and the demographic data sheet I w i l l mail you a check for $5.00. Benefits to p a r t i c i p a t i o n 1. Increased understanding of yourself i n r e l a t i o n to r o l e a c t i v i t i e s . 2. The opportunity to express your personal views regarding r o l e a c t i v i t i e s . 3. The opportunity to p a r t i c i p a t e i n a worthwhile project that may help others adapt to the dual-earner l i f e s t y l e . Your anonymity w i l l be preserved because a l l questionnaires w i l l be i d e n t i f i e d only by a code number. A l i s t , i d e n t i f y i n g subjects* names and code numbers, w i l l be held i n confidence by the investigator. I w i l l assume that you have consented to p a r t i c i p a t e i n the study i f you complete copy #1 of the Role Enactment Questionnaire. However, you have the r i g h t to refuse to p a r t i c i p a t e or withdraw at any time. I f you have any questions, please do not hesitate to c a l l me. Thank-you i n advance for your assistance i n t h i s matter. Kenna Sleigh, RN, BSN 

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.831.1-0086996/manifest

Comment

Related Items