Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The influences of gender, family involvement, and work-family conflict on managers’ preferences for family-oriented… Olsen, Jane Michelle 1996

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

Item Metadata

Download

Media
831-ubc_1996-0592.pdf [ 5.07MB ]
Metadata
JSON: 831-1.0087257.json
JSON-LD: 831-1.0087257-ld.json
RDF/XML (Pretty): 831-1.0087257-rdf.xml
RDF/JSON: 831-1.0087257-rdf.json
Turtle: 831-1.0087257-turtle.txt
N-Triples: 831-1.0087257-rdf-ntriples.txt
Original Record: 831-1.0087257-source.json
Full Text
831-1.0087257-fulltext.txt
Citation
831-1.0087257.ris

Full Text

THE INFLUENCES OF GENDER, FAMILY INVOLVEMENT, ZAND WORK-FAMILY CONFLICT ON MANAGERS' PREFERENCES FOR FAMILY-ORIENTED BENEFITS by JANE MICHELLE OLSEN B.H.E., The University of B r i t i s h Columbia, 1993 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS i n THE FACULTY OF GRADUATE STUDIES The School of Family and N u t r i t i o n a l Sciences Family Studies We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October, 1996 Jane Michelle Olsen, 1996 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 ^ c X m iW " S V u C ^ g S The University of British Columbia Vancouver, Canada •ate OcvV-ober \0 , \99k DE-6 (2/88) Abstract This study examines the preferences of managers for job attributes designed to a s s i s t i n balancing work and family demands. S p e c i f i c a l l y , t h i s study investigates, within a conceptual framework r e l a t i n g work and extra-work domains (Rice, Near, & Hunt, 1979), whether male and female managers exhibit differences i n t h e i r preferences for family-oriented work benefits and i f such preferences are influenced by the l e v e l s of family involvement and work-family c o n f l i c t characterizing the worker. Additionally, t h i s study tests whether work-family c o n f l i c t mediates the relationships between gender and preferences and between family involvement and preferences. Data were from the Career Development Survey of Managers and Professionals (Langton, 1995). The respondents were 264 male and female managers employed i n organizations located throughout the Lower Mainland of B r i t i s h Columbia. Results supported the hypotheses that gender, family involvement, and work-family c o n f l i c t influence managers' preferences for family-oriented benefits. However, the hypothesized gender difference was not supported when p a r t i c u l a r work- and family-related variables were used as s t a t i s t i c a l controls. The study also f a i l e d to f i n d support for the hypotheses concerning the mediator e f f e c t s of work-family c o n f l i c t . Results are discussed i n terms of previous research and model development i n the area of work and family. Implications of t h i s study for employees, employers, and family l i f e educators as well as for future research are also discussed. Table of Contents Abstract i i L i s t of Tables v L i s t of Figures v i Acknowledgement v i i Chapter I. Introduction 1 D e f i n i t i o n of Terms 3 Review of the Literature 7 Employee Preferences 7 Preferences for p a r t i c u l a r job attributes 7 Preferences for family-oriented work benefits.... 12 Summary 23 Work-Family C o n f l i c t 24 Gender differences i n work-family c o n f l i c t 24 Family involvement and work-family c o n f l i c t 25 Antecedents of work-family c o n f l i c t 27 Family-oriented benefits and work-family c o n f l i c t 29 Summary 3 0 Conceptual Model 30 Hypotheses 34 II. Method 39 Sample and Data C o l l e c t i o n 39 Measures 42 Dependent Variable ••• 42 Independent Variables 44 Control Variables 46 III. Results 49 i i i Preliminary Analyses 49 Comparison of the Two Sets of Data 49 Comparison of Respondents and Non-Respondents 49 Univariate Distributions 50 Determination of Control Variables 51 Preferences for S p e c i f i c Family-Oriented Benefits 54 Hypothesis Testing 54 Post Hoc Analyses 65 Gender Difference i n Preferences 65 F i t with the Conceptual Model 70 Summary 70 IV. Discussion 73 Preferences for Family-Oriented Benefits 73 Gender and Preferences 73 Family Involvement and Preferences 76, Work-Family C o n f l i c t and Preferences 76 Work-Family C o n f l i c t 77 Gender and Work-Family C o n f l i c t 77 Family Involvement and Work-Family C o n f l i c t 80 The Conceptual Model 82 Limitations 84 Conclusions and Implications 86 References 92 Appendices 101 i v L i s t of Tables Table 1 . Correlations Between Potential Control Variables and Preferences for Family-Oriented Benefits 52 Table 2. Correlations Between Potential Control Variables and Work-Family C o n f l i c t 53 Table 3. Intercorrelations Between Control Variables 55 Table 4. Summary of Regression of Preferences for Family-Oriented Benefits on Control Variables and Family Involvement 59 Table 5. Summary of Regression of Preferences for Family-Oriented Benefits on Control Variables and Work-Family C o n f l i c t 61 Table 6. Summary of Regression of Work-Family C o n f l i c t on Control Variables and Family Involvement 64 Table 7. Summary of T-Tests Comparing Men and Women on the Control Variables 66 Table 8. Summary of Regression Analysis for Variables Predicting Preferences for Family-Oriented Benefits Among Men. 68 Table 9. Summary of Regression Analysis for Variables Predicting Preferences for Family-Oriented Benefits Among Women 69 v L i s t of Figures Figure 1. The conceptual model adapted from Rice, Near, and Hunt (1979) 33 Figure 2. Hypothesized relationships among research variables 38 Figure 3. Results regarding the hypothesized relationships among research variables 72 v i Acknowledgement I would l i k e to express my sincere appreciation to the supervisor of my thesis committee, Dr. P h y l l i s Johnson, for her continued support and encouragement throughout the completion of th i s project. I would also l i k e to thank the other members of my thesis committee, Dr. Daniel Perlman and Dr. Nancy Langton, for the guidance they generously provided. Special thanks are due to Dr. Nancy Langton for giving me the opportunity to work with her i n c o l l e c t i n g the data for t h i s project. My most profound thanks, however, are due to my family, whose t i r e l e s s and u n s e l f i s h support made the completion of t h i s project possible. v i i 1 Chapter I Introduction There exists a plethora of research investigating the work-related preferences of employees. Some studies have explored individuals' preferences for p a r t i c u l a r types of occupations (e.g., Bartol, 1976a; Holmstrom & Beach, 1973; Sheard, 1970). Other studies have examined the extent to which individuals prefer to spend time at work as opposed to with family (e.g., Karambayya & R e i l l y , 1992; Konrad & Langton, 1993; Yogev & Brett, 1985). However, the majority of preference studies have investigated s p e c i f i c aspects of the work s i t u a t i o n desired by workers (e.g., Bigoness, 1988; Jurgensen, 1978; Wiersma, 1990). Although there are numerous studies investigating preferences for s p e c i f i c work aspects, few researchers have s p e c i f i c a l l y examined preferences for job attributes that make i t possible for the worker to successfully balance the demands of work and family. Many studies have asked individuals to indicate how important "benefits" are to them i n t h e i r jobs, but such benefits t y p i c a l l y refer to employer-provided discounts, services, cars, and so f o r t h rather than family-oriented benefits such as on- or near-site c h i l d care, extended leaves for personal or family reasons, and information and r e f e r r a l services to a s s i s t workers i n coping with work and family issues. S i m i l a r l y , researchers have investigated how important "work hours" are to employees, but have f a i l e d to l i n k the job a t t r i b u t e to the family i n terms of i t s implications. Absent from the l i t e r a t u r e are studies which examine employees' preferences for work arrangements which would enable them to better meet family 2 obligations. With the continued increase i n the number of women entering the labour force, and even more dramatic increases i n the numbers of dual-career and single-parent families, there i s a growing number of workers--both male and female--who must manage work and family r e s p o n s i b i l i t i e s . Consequently, an increasing number of organizations are beginning to provide family-oriented benefits for t h e i r employees (Friedman & Galinsky, 1992). However, l i t t l e e f f o r t i s being made on the part of employers to assess how highly such benefits are rated by the workers for whom they are provided. A study by the Conference Board of Canada of approximately 400 organizations employing over one m i l l i o n Canadians found that nearly 60% of the organizations had not conducted a survey of t h e i r employees' benefit preferences (Paris, 1989). As Bowen (1988) states, there i s a ...lack of rigorous and systematic research examining the forces behind and processes and impacts of implementing family-oriented p o l i c i e s and services i n work organizations. ...To date, research has l a r g e l y consisted of anecdotal and descriptive accounts of various types of family support innovations, (p. 183) Indeed, i t i s surprising that there i s a dearth of research examining the needs and desires of workers for p a r t i c u l a r family-oriented benefits given that such research would be b e n e f i c i a l for both the employee, who would benefit from the provision of p o l i c i e s and programs which are t r u l y needed, and the employer, who would p r o f i t from providing only those services which are used by workers. The purpose of t h i s study i s to examine the preferences of managers for s p e c i f i c job attributes which are purposely 3 implemented to help them ease the c o n f l i c t i n g demands between work and family. S p e c i f i c a l l y , t h i s study investigates whether men and women managers exhibit differences i n t h e i r preferences for family-oriented benefits and i f such preferences are influenced by the lev e l s of family involvement and work-family c o n f l i c t characterizing the worker. D e f i n i t i o n of Terms The following concepts are defined for the purposes of t h i s study. The defined terms r e f l e c t those used on the study's measurement instrument. Preferences The concept of preferences used i n t h i s study derives from the work of Pryor (1979, 1981, 1982). According to Pryor (1979), a preference for an aspect of work i s "a statement of the r e l a t i o n between a person (the subject of the relation) and a p a r t i c u l a r q u a l i t y of work (the object of the r e l a t i o n ) . The relationship between the two i s that of a greater or lesser l i k i n g when the person has the opportunity to make a choice" (p. 254). Although the concept of preferences has been used interchangeably with other concepts such as values, needs, work ethics, and orientations to work, Pryor (1982) maintains that these concepts are q u a l i t a t i v e l y s i m i l a r categories of relations between humans and work which d i f f e r primarily i n the l e v e l of generality of the c h a r a c t e r i s t i c s of work considered. Furthermore, he states that each of these related concepts i s "a d i s t i n c t way of conceptualizing occurrences i n the world of vocational psychology. ...[and] i s equally v a l i d as an explanation of some aspects of vocational behavior" (1982, p. 4 42) . Family-Oriented Benefits Family-oriented benefits are workplace p o l i c i e s , programs, and practices aimed at improving employees' a b i l i t y to balance t h e i r work and family r e s p o n s i b i l i t i e s . They represent [an] organization's e f f o r t s to support workers with family r e s p o n s i b i l i t i e s - those who must provide for the d a i l y care and the management of immediate, dependent, or blended family members, including children, parents, grandparents, spouses, in-laws, and steprelations. ...They refe r to services that make the everyday management of family r e s p o n s i b i l i t i e s easier. (Thomas & Ganster, 1995, p. 7) F l e x i b l e work hours. This type of benefit i s an arrangement whereby employees are granted l i m i t e d control over the s t a r t i n g and f i n i s h i n g times of t h e i r work day, provided that they work a standard number of hours during a mandatory time period. Part-time employment with pro-rated benefits. The option to work part-time with pro-rated benefits means that an employee may choose to work less than 35 hours per week and "receive benefits i n proportion to the number of hours worked" (Paris, 1989, p. 12) . Work-location f l e x i b i l i t y . Often termed " f l e x i p l a c e " , work-location f l e x i b i l i t y "includes t r a d i t i o n a l home-based work as well as newer arrangements, such as s a t e l l i t e o f f i c e s and remote or neighborhood centers" (Ferber, O'Farrell, & A l l e n , 1991, p. 128-129). Employees are t y p i c a l l y expected to spend some time on-site at the workplace, but t h e i r home or s a t e l l i t e o f f i c e i s i d e n t i f i e d as the main place of work. On- or near-site daycare f a c i l i t i e s . On- or near-site daycare refers to c h i l d care centres or nurseries set up at or 5 near to the workplace. F l e x i b l e benefit programs. Usually referred to as "cafeteria plans", these programs are ...arrangements i n which employees t a i l o r t h e i r benefits package to t h e i r s p e c i f i c needs. Employees can select the benefits they value most and may forgo benefits of lesser importance to them. Under a f l e x i b l e arrangement, an employer allocates a s p e c i f i e d amount of money to each employee to "purchase" benefits. In t h i s way, employers control the amount they spend...for benefits, while the employee selects the benefits. (Meisenheimer & Wiatrowski, 1989, p. 17) Leaves for personal/family reasons. Personal/family leaves are arrangements whereby employees are enabled to take time off to deal with personal or family matters, such as the b i r t h of a c h i l d , a sick family member, or the pursuit of education, without a f f e c t i n g t h e i r career progress. Workplace counselling and information services. Counselling and information services r e f e r to employee assistance programs which are s p e c i f i c a l l y designed to help employees manage t h e i r work and family l i v e s . Examples of such services include resource and r e f e r r a l services for c h i l d and elder care; marriage and parent-child counselling; retirement and f i n a n c i a l counselling; and counselling for employees experiencing stress-related problems (Paris, 1989). Relocation assistance programs. Relocation assistance primarily consists of the payment of costs associated with t r a n s f e r r i n g to a d i f f e r e n t work s i t e . These costs may include t r i p s to look for housing, temporary l i v i n g expenses, storage, d i r e c t moving costs,...the purchase of furniture and appliances... as well as job assistance placement f o r the spouse. (Hughes & Galinsky, 19 88, p. 259) 6 Organizational culture supportive and understanding of family r e s p o n s i b i l i t i e s . Organizational culture i s "the set of norms, values, and informal mechanisms that shape day-to-day l i f e i n an organization [which] are l i k e l y to be important i n determining both the a v a i l a b i l i t y and the use of benefits" (Ferber et a l . , 1991, p. 140). An organizational culture which i s supportive and understanding of family r e s p o n s i b i l i t i e s - - a "family-friendly" organizational culture--characterizes an organization which openly acknowledges employees' family or personal situations (Friedman & Galinsky, 1992). A family-f r i e n d l y culture supports the adoption of needed p o l i c i e s and programs designed to ease work-family tensions and, most importantly, encourages a l l employees to make use of such p o l i c i e s and programs. A supervisor who i s f l e x i b l e when family needs a r i s e . A f l e x i b l e supervisor i s one who ...empathizes with the employee's desire to seek balance between work and family r e s p o n s i b i l i t i e s . This... [ f l e x i b i l i t y ] might include accommodating an employee's f l e x i b l e schedule, being tolerant of short personal phone c a l l s . . . , granting a time trade so that new elder-care arrangements can be monitored, [or] allowing one to bring a c h i l d to work [when care arrangements cannot be made]. (Thomas & Ganster, 1995, p. 7) Family Involvement As i n Duxbury and Higgins' (1991) study, family involvement i s conceptualized as "the degree to which a person i d e n t i f i e s psychologically with family roles, the importance of the family to the person's self-image and self-concept, and the individual's commitment to family roles" (p. 62). Family involvement, i n t h i s study, refers to the ov e r a l l psychological importance of the 7 family role to an individual rather than merely to the t o t a l amount of time devoted to the family r o l e . Work-Family C o n f l i c t Work-family c o n f l i c t i s "a form of i n t e r r o l e c o n f l i c t i n which the role pressures from the work and family domains are mutually incompatible i n some respect. That i s , p a r t i c i p a t i o n i n the work (family) role i s made more d i f f i c u l t by p a r t i c i p a t i o n i n the family (work) role" (Greenhaus & Beutell, 1985, p. 77). Work-family c o n f l i c t can re s u l t from any aspects of work or family roles that af f e c t one's time involvement, s t r a i n , or behaviour within a role (Burke, 1988). Review of the Literature The l i t e r a t u r e pertaining to two major areas was reviewed: employee preferences and work-family c o n f l i c t . Employee Preferences For the purposes of t h i s study, the l i t e r a t u r e reviewed on employee preferences concentrated on workers' preferences for p a r t i c u l a r job attributes such as pay, security, and opportunities for advancement, and on workers' preferences for family-oriented work benefits. Preferences for p a r t i c u l a r job a t t r i b u t e s . Over the past f i f t y years, much research has been conducted investigating the r e l a t i v e importance workers place on various attributes of t h e i r occupations. Such studies of job a t t r i b u t e preferences are diverse. Research has examined individual preferences for a wide range of job c h a r a c t e r i s t i c s including autonomy, f i n a n c i a l compensation, and opportunities for interpersonal contact. Researchers have also approached the study of job a t t r i b u t e 8 preferences from several d i f f e r e n t perspectives. They have examined the differences i n worker preferences f o r p a r t i c u l a r job ch a r a c t e r i s t i c s that exist among d i f f e r e n t segments of the general population (e.g., Andrisani & Miljus, 1971; Weaver, 1975, 1976, 1992), the relationship between preferences and job s a t i s f a c t i o n (e.g., Gorn & Kanungo, 1980; Kalleberg & Loscocco, 1983), how preferences have changed over time (e.g., Jurgensen, 1978; Posner & Powell, 1990), and how preferences for p a r t i c u l a r aspects of a job relate to job application (e.g., Posner, 1981; Rowe, 1976). The vast majority of studies of work-related preferences, though, have had as t h e i r primary purpose to determine whether men and women d i f f e r i n terms of the job attributes they prefer. "The rationale underlying t h i s stream of research i s the b e l i e f that should sex differences i n job attri b u t e preferences be found, d i f f e r e n t motivational strategies and reward contingencies for male and female employees may be warranted" (Bigoness, 1988, p. 139). Research comparing the job att r i b u t e preferences of men and women has produced mixed r e s u l t s . On the one hand, studies have found few gender differences and showed men and women to be more simi l a r than d i f f e r e n t with respect to t h e i r opinions about the s p e c i f i c aspects of work. Although many studies of job a t t r i b u t e preferences have been l i m i t e d to small samples of uni v e r s i t y students, results of numerous studies have found few gender differences i n preferences for advancement opportunities, money, job security (e.g., Bri e f , Rose, & Aldag, 1977; Scozzaro & Subich, 1990; Terpstra, 1983), s o c i a l prestige and status, in t e r a c t i o n with others (e.g., Brief & Aldag, 1975), f r i e n d l y co-9 workers, competent supervisors, task variety, feedback and recognition (e.g., Singer, 1974; Terpstra, 1983), work that gives a f e e l i n g of accomplishment and achievement, opportunities to use s k i l l s , autonomy, challenging work (e.g., Manhardt, 1972), important r e s p o n s i b i l i t i e s , employee benefits (e.g., Bigoness, 1988), and short working hours permitting greater l e i s u r e time (e.g., Lacy, Bokemeier, & Shepard, 1983). Simi l a r l y , few gender differences have been found i n studies which have created dimensions of job attr i b u t e preferences and analyzed t h e i r data accordingly. Bartol (1976b), for example, divided 25 job ch a r a c t e r i s t i c s into three groups: long-term career objectives, which contained c h a r a c t e r i s t i c s such as advancement, income, and r e s p o n s i b i l i t y ; working environment and interpersonal relationships, which included such c h a r a c t e r i s t i c s as co-workers, work conditions, and job security; and i n t r i n s i c job aspects, which focussed on c r e a t i v i t y , i n t e l l e c t u a l stimulation and independence afforded by the work. She found that men and women did not d i f f e r i n t h e i r ratings of the importance of long-term career objectives and i n t r i n s i c job aspects. And Walker, Tausky, and Oliver (1982) examined gender di f f e r e n c e s . i n preferences for f i v e dimensions of job att r i b u t e s : e x t r i n s i c , i n t r i n s i c , resource adequacy, convenience, and relations with co-workers. Results revealed that men and women did not d i f f e r i n terms of the importance they placed on e x t r i n s i c or i n t r i n s i c aspects of work nor on having good relationships with co-workers. Furthermore, males and females expressed si m i l a r l e v e l s of desire for having s u f f i c i e n t resources for job performance. However, Saleh and L a l l j e e (1969) 10 and Andrisani and Miljus (1977) caution that analysis of dimensions of job attributes does not permit comparisons between men and women on s p e c i f i c job c h a r a c t e r i s t i c s . On the other hand, i n contrast to studies which have found few gender differences i n preferences for p a r t i c u l a r job attributes, there i s a body of research l i t e r a t u r e which suggests that there are s i g n i f i c a n t gender differences i n job attr i b u t e preferences. A s t r i k i n g number of studies has shown that men, compared to women, ascribe greater importance to salary, advancement opportunity, having important r e s p o n s i b i l i t i e s , a work schedule which provides ample l e i s u r e time off the job, working on central organizational problems (e.g., Beutell & Brenner, 1986; Manhardt, 1972), job security, prestige (e.g., Berger, 1986; Block, Denker, & T i t t l e , 1981), d i r e c t i n g the work of others (e.g., Daymont & Andrisani, 1984), and good fringe benefits (e.g., Jurgensen, 1947). Other studies have shown that women have a greater preference for congenial co-workers, pleasant or comfortable working conditions, opportunities to work for supervisors whom they admire and respect (e.g., Beutell & Brenner, 1986; Jurgensen, 1947; Manhardt, 1972), working with people, and jobs which enable them to help others (e.g., Cherry, 1975; Lyson, 1984). And contrary to the research findings which report that men and women do not d i f f e r i n t h e i r preferences for i n t r i n s i c aspects of work, several studies have found that women more so than men value work which provides feelings of accomplishment and achievement, i s challenging, permits the employee to be r e l a t i v e l y independent and autonomous, affords the opportunity to u t i l i z e one's education and s k i l l s , f a c i l i t a t e s 11 s k i l l enhancement (e.g., Beutell & Brenner, 1986), offers s k i l l and task variety, and i s acknowledged with feedback and recognition (e.g., Scozzaro & Subich, 1990). Moreover, i n a study using a highly representative national sample, Walker et a l . (1982) found that men and women d i f f e r e d i n t h e i r preferences for convenience aspects of work, with women being s i g n i f i c a n t l y more l i k e l y than men to desire convenient t r a v e l to work, good hours, freedom from c o n f l i c t i n g demands, pleasant physical surroundings, a reasonable amount of work, and the opportunity to a i r personal problems. This finding i s consistent with Martin and Hanson's (1985) argument that ...women with combined family and economic r e s p o n s i b i l i t i e s require p a r t i c u l a r c h a r a c t e r i s t i c s of jobs more than t h e i r male counterparts. ...[Working] women with family r e s p o n s i b i l i t i e s respond more favorably than men to comfortable or convenient jobs. Comfort and convenience i n thi s sense refers to jobs that do not make excessive demands on the worker i n terms of hours, overtime, travel...and so on. ...[Women] desire these c h a r a c t e r i s t i c s i n that convenient and comfortable work i s most l i k e l y to be compatible with family r e s p o n s i b i l i t i e s . In other words, e f f e c t i v e performance at work does not necessarily hinder e f f e c t i v e performance at home. (p. 94) While some researchers claim that gender differences i n job attri b u t e preferences disappear i f p o t e n t i a l l y confounding variables such as occupation and organizational l e v e l are held constant (Brief & Aldag, 1975; Bri e f & Oliver, 1976), several studies have demonstrated that differences i n the work aspect preferences between men and women remain even a f t e r c o n t r o l l i n g for a va r i e t y of variables including age, race, organizational position, job experience, education, income, and tenure (Block et a l . , 1981; Hatcher, 1991; Lacy et a l . , 1983; Schuler, 1975). 12 Preferences for family-oriented work benefits. While much research has been devoted to the study of preferences people have for s p e c i f i c aspects of t h e i r work, studies examining preferences for j o b s - - s p e c i f i c a l l y attributes of jobs--that make i t easy to balance work and family are scant. In recent years, an increasing amount of l i t e r a t u r e has focussed on the role of workplace and government p o l i c i e s i n reducing the s t r a i n associated with combining work and family r e s p o n s i b i l i t i e s , on the prevalence of family-related supports, and on the impact corporate involvement i n the work-family domain has on both employers and employees. Yet, l i t t l e attention has been paid to the attitudes and preferences of workers with respect to the provision by employers of family-oriented benefits despite the fact that such attention has been c a l l e d for by scholars (e.g., Covin & Brush, 1993; Raabe, 1990). As Raabe (1990) states, ...policy relevance i s often work-site s p e c i f i c , and surveys are important to disclose the p a r t i c u l a r work family problems and p o l i c y preferences of s p e c i f i c work forces. ...For example, an on-site child-care center may not be the desired or appropriate solution to a given work force's child-care problems. . . . D i f f e r e n t i a l l i f e - c y c l e needs are also important sources of variations i n family problems. ...[The] point here for improved evaluation research i s the importance of r e a l i z i n g possible sources of v a r i a b i l i t y i n both p o l i c y relevance and e f f e c t s , (pp. 482-483) The few empirical studies that have investigated the importance employees place on various job attributes which a s s i s t i n balancing work and family demands are l i m i t e d i n terms of t h e i r methodologies and the number and types of attributes they examine. Posner and Powell (1990), for example, asked respondents to rate the importance of counselling services as a fringe benefit i n t h e i r decision to accept a job. The importance 13 placed on t h i s p a r t i c u l a r benefit was not great and did not vary-over time. Padavic (1992) examined preferences for day work among women i n white c o l l a r occupations and found that the more important i t was to work only days, the less interest a respondent had i n transf e r r i n g to a blue c o l l a r job. And Moen and Dempster-McClain (1987) used a single question to measure employed parents' preferences for working less i n order to spend more time with t h e i r families. They found that women, more so than men, preferred to work less and that actual work time and perceived work-family interference were s i g n i f i c a n t l y p o s i t i v e l y related to the desire to work fewer hours. The occasional study has examined, rather c u r s o r i l y , worker preferences for a limi t e d range of family-oriented job benefits. Quinn and Staines (1979) investigated how wage and sa l a r i e d workers evaluated d i f f e r e n t fringe benefits including maternity leave with f u l l employment rights, maternity leave with pay, and c h i l d care services. They found that few workers rated any of these three benefits as very important and even fewer desired an improvement i n the benefit i f i t was already available to them. And Galinsky and Hughes (as c i t e d i n Galinsky & Stein, 1990), i n a study of two-parent working families, found that, next to pay increases, supervisors who are accommodating when family needs arise was the job attr i b u t e most desired as a means of improving the q u a l i t y of work and family l i f e . Bridges (1989) examined gender differences i n respondent ratings of 18 job ch a r a c t e r i s t i c s , three of which were connected with balancing work and family roles--ease of arranging the work schedule to coincide with the schedule of children and/or spouse, 14 ease of reentering the f i e l d a f t e r an interruption for c h i l d rearing, and a v a i l a b i l i t y of part-time employment. She found that female unive r s i t y students valued a l l three c h a r a c t e r i s t i c s s i g n i f i c a n t l y more than did male students. Higgins, Duxbury, and Lee (1992), i n a study of over 14,000 Canadian private sector employees, questioned workers about the extent to which f i v e d i f f e r e n t work arrangements appealed to them. The f i v e work arrangements--regular, flextime, compressed work weeks, work-at-home, and part-time--were chosen for examination because the l i t e r a t u r e suggests that they "reduce work-family c o n f l i c t by increasing an employee's a b i l i t y to control, predict and absorb change i n work and family roles" (Higgins et a l . , 1992, p. 29). Higgins et a l . (1992) found that both male and female employees had favourable attitudes towards a l l of the f l e x i b l e work arrangements. Managerial/professional and technical workers f e l t flextime, compressed work weeks, and work-at-home arrangements were more appealing than did cl e r i c a l / a d m i n i s t r a t i v e workers. Women i n c l e r i c a l and technical occupations appeared to prefer part-time work more than managerial/professional women did. Women were s i g n i f i c a n t l y more l i k e l y than men to f i n d work-at-home and part-time arrangements very appealing. Managerial/professional women exhibited stronger preferences for compressed work weeks, work-at-home arrangements, and part-time work than did managerial/professional men. Furthermore, dual-career mothers and fathers and married-c h i l d l e s s individuals found flextime, compressed work weeks, and work-at-home arrangements more appealing than dual-earner parents and individuals i n t r a d i t i o n a l and single-parent families. Women 15 from dual-career and dual-earner families preferred part-time work more than women from other family types. As well, t r a d i t i o n a l and dual-earner fathers were s i g n i f i c a n t l y more l i k e l y than a l l other groups to rate a regular "9-5" work arrangement appealing. Using the same data set as Higgins et a l . (1992) but including information from public sector workers i n t h e i r analyses, Higgins, Duxbury, Lee, and M i l l s (1992) found few differences between private and public sector employees i n terms of t h e i r desire for p a r t i c u l a r work arrangements. One difference was that fewer private sector employees were attracted to work-at-home arrangements. Another difference was that public sector managerial/professional women, unlike t h e i r private sector counterparts, found flextime more appealing than did public sector managerial/professional men. A small number of studies have s p e c i f i c a l l y and systematically investigated the preferences of employees for job attributes which enable them to ease work-family tensions. Mize and Freeman (1989) questioned randomly-selected male and female employees of a large u n i v e r s i t y about, among other things, the r e l a t i v e value of various forms of employer-sponsored c h i l d care including information and r e f e r r a l services, an on-campus daycare center, and f l e x i b l e benefit packages. In addition, employees were asked to indicate how important the existence of an employer-sponsored on-site daycare center would be i n t h e i r decision about whether or not to continue employment at the University. While on-site daycare was perceived as the most valuable form of employer c h i l d care support, the majority of 16 respondents rated the existence of an on-site center as unimportant i n the decision to continue employment. The Canadian Aging Research Network (1993) surveyed over 5000 individuals employed i n eight Canadian organizations representing f i v e employment sectors about, among other things, t h e i r use of and preferences for f l e x i b l e work options. Eleven work options and three types of employee assistance were presented to respondents: extended leave; short term leave; personal days; time off i n l i e u of overtime pay; short work week; part-time, f u l l benefits; part-time, pro-rated benefits; part-time, no benefits; f l e x i b l e hours; job sharing; work-at-home; elder care information; c h i l d care information; and an employee assistance program. Survey results indicated that shorter work weeks, f l e x i b l e hours, and work-at-home arrangements were the most desired options, while part-time work with pro-rated or no benefits and employee assistance programs were the least preferred options. Results also revealed that those individuals who were engaged i n both c h i l d care and elder care were more l i k e l y to desire work-at-home arrangements than were individuals without such r e s p o n s i b i l i t i e s . Using employees of a Midwestern u t i l i t y company as a sample, Kossek (1990) also assessed the preferences for various c h i l d care assistance options--job-sharing or part-time work, information r e f e r r a l assistance, family daycare, sick care, on-or near-site c h i l d care, a voucher system or f i n a n c i a l assistance i n purchasing care, f l e x i b l e spending accounts, parental leave, and educational seminars or discussion groups. Respondents were asked to rank these options i n order of preference. Results of 17 data analyses i l l u s t r a t e d a d i v e r s i t y i n preferences. Sick care and on-site care were the most preferred options, e s p e c i a l l y among single-parents as opposed to dual-career and two parent-t r a d i t i o n a l families, among parents with infants as opposed to parents with older children, among employees using non-familial as opposed to f a m i l i a l or mixed care, and among male employees i n either single-parent or dual-career households as opposed to t h e i r female counterparts. Single-parents also had s i g n i f i c a n t l y stronger preferences for information and r e f e r r a l assistance and voucher systems than did other family types. Only one gender difference was found--female employees preferred job sharing and part-time work to a greater extent than males. A three-way int e r a c t i o n e f f e c t between gender, age of c h i l d , and household employment configuration on preferences for family daycare was noted, suggesting that female employees i n dual-career households with infant children were es p e c i a l l y l i k e l y to prefer employer support of a family daycare network.. Non-managers were more interested i n family daycare whereas managers were more interested i n f l e x i b l e spending accounts. And f i n a l l y , parents with children i n non-familial care preferred educational or discussion support groups more so than parents using f a m i l i a l care. Building on an e x i s t i n g scale developed and used by Manhardt (1972) to measure preferences for 25 job attributes, Wiersma (1990) added items measuring preferences for eight parental support job attributes meant to reduce role c o n f l i c t . The parental support attributes included a four-day work week, the a b i l i t y to s h i f t between part-time and f u l l - t i m e work, flextime 18 around an eight-hour core, f i v e days o f f f o r sick children, c h i l d care education money, c h i l d rearing leave of absence, a company daycare center, and a company daycare lunch room. Although i t s sample was li m i t e d to families with young children who use daycare centres, Wiersma's (1990) study found that women rated a l l of the parental support attributes as s i g n i f i c a n t l y more important than men did and that role c o n f l i c t was a mediator between gender and preferences for parental support. In a study by Goldberg, Greenberger, Koch-Jones, O'Neil, and Hamill (19 89), the attractiveness of ten workplace benefits and p o l i c i e s to employed married and single parents of preschoolers was examined. Respondents were asked i f being offered a p a r t i c u l a r plan or benefit would motivate them to leave t h e i r current employer. The benefits and p o l i c i e s studied were grouped into four categories: c h i l d care benefits, parental leave benefits, scheduling benefits, and programs and services dealing with issues relevant to working parents. Results revealed that women were more interested than men i n c h i l d care benefits. Single women were s i g n i f i c a n t l y more interested than married men and women i n two s p e c i f i c c h i l d care benefits--employer contributions to c h i l d care and on- or near-site c h i l d care f a c i l i t i e s . Women were also more interested than men i n parental leave p o l i c i e s , with single mothers being the most interested. Two scheduling benefits--a shorter work week and f l e x i b l e work hours--were the most appealing benefits o v e r a l l . In general, respondents' preferences for work scheduling alternatives did not vary by marital status, gender, or occupational category. Rel a t i v e l y few respondents expressed an interest i n programs and 19 services aimed at a s s i s t i n g with balancing the demands of work and family and interest i n such programs did not vary according to gender, marital status, or occupational status. Employing a si m i l a r method to that of Goldberg et a l . (1989), Galinsky, Bond, and Friedman (1993), measured preferences for family-oriented benefits by assessing the trade o f f s subjects were w i l l i n g to make to obtain them. Analyses of trade of f s indicated that a r e l a t i v e l y small number of respondents without access to the benefits was w i l l i n g to change employers or trade job advancement to obtain time off for c h i l d b i r t h and parenting, time off to care for sick family members, part-time work, the opportunity to set t h e i r own s t a r t i n g and q u i t t i n g times, the opportunity to work fewer hours one day and make up the time l a t e r , extended lunch breaks, the opportunity to regularly work at home, f l e x i b l e spending accounts for dependent care, c h i l d care resources and r e f e r r a l s , elder care resources and r e f e r r a l s , employer-sponsored c h i l d care at or near the work s i t e , and vouchers to purchase c h i l d care services. Over one-third of respondents, however, were w i l l i n g to trade salary and other benefits for time off for parenting, and nearly half would do the same to obtain time off to care for sick family members. Women with children under thirteen years of age were s i g n i f i c a n t l y more l i k e l y than men with children under thirteen to make trade o f f s to gain access to part-time work, flextime, and work-at-home options. There was no difference between men and women, however, i n the trade of f s they were w i l l i n g to make to gain access to dependent care assistance. Parents were s i g n i f i c a n t l y more l i k e l y than non-parents to indicate a willingness to switch jobs 20 i n order to have flextime and be able to work at home. Parents were also s i g n i f i c a n t l y more l i k e l y to indicate a willingness to trade salary and other benefits for dependent care assistance--for c h i l d and elder care resource and r e f e r r a l services, on- or near-site c h i l d care and c h i l d care vouchers. Lee, Duxbury, Higgins, and M i l l s (1992) employed a single item measure to assess public sector workers' preferences i n terms of what t h e i r employers could do to help them balance work and family demands. Information gathered from interviews with over 800 employed parents suggested that men and women d i f f e r i n th e i r ranking of important job at t r i b u t e s . Women mentioned f l e x i b l e work hours, family leaves, and on-site c h i l d care most often while men tended to mention working at home, f l e x i b l e work hours, and a supportive supervisor. Single mothers indicated a stronger preference for part-time work and job sharing than did married mothers, and both mothers and fathers with preschoolers desired on-site c h i l d care more than did parents with older children. Managerial/professional mothers were more l i k e l y than managerial/professional fathers to indicate a preference for f l e x i b l e hours, increased family leave, on-site daycare, part-time work, job sharing, and an understanding supervisor. One of the more extensive examinations of employee preferences for family-oriented job c h a r a c t e r i s t i c s i s that of Louis Harris and Associates (1981). Using both personal interview and questionnaire techniques of data c o l l e c t i o n , they investigated preferences for working fu l l - t i m e , working part-time, doing volunteer work, and working at home and analysed the data according to gender, employment status, and occupation. 21 Furthermore, they questioned workers about the degree to which a var i e t y of family-oriented benefits provided by employers would help them i n balancing work and family r e s p o n s i b i l i t i e s and whether or not they would trade off salary to obtain each benefit. The benefits examined included paid personal days s p e c i f i c a l l y for c h i l d and family r e s p o n s i b i l i t i e s ; paid personal leave for mothers for more than s i x weeks for pregnancy and maternity; on-site c h i l d care; paid personal leave for fathers for paternity; freedom to pick and choose the employment benefits that are best for family needs; part-time work for two or three days a week with f u l l - t i m e employment benefits; job counselling and job hunting services for the spouse i f the employer transferred the other to a new community; the right to resume work at the same pay and s e n i o r i t y a f t e r a personal leave of absence; a choice between a 7 to 3, 8 to 4, or 9 to 5 work day; freedom to set a work schedule; work schedules that allow one day to work at home; a four-day work week; the right to refuse relocation with no career penalty; job sharing; and a shorter work week with less pay. The study showed that most employed individuals--and men more so than women--did not f e e l that c h i l d care benefits, a f l e x i b l e benefit package, part-time work, leaves, f l e x i b l e s t a r t i n g and q u i t t i n g times, the right to refuse relocation, job sharing, or a shortened work week would be very b e n e f i c i a l i n helping to balance work and family r e s p o n s i b i l i t i e s . Employment status was, for the most part, unrelated to benefit preferences. Professionals/managers preferred to work f u l l - and part-time more than did individuals i n white c o l l a r occupations, whereas a greater number of 22 individuals i n white c o l l a r occupations preferred to work at home. Freedom to select an ind i v i d u a l package of benefits, part-time work, job counselling, the right to refuse relocation, and the remaining p o l i c i e s aimed at increasing work schedule f l e x i b i l i t y were regarded by respondents as being more b e n e f i c i a l than the c h i l d care benefits i n a s s i s t i n g with balancing work and family demands. While t h i s study examines a wide range of family-oriented benefits and i s based on a na t i o n a l l y representative sample, i t i s dated, possibly l i m i t e d by a non-response bias, and neglects to examine important factors which may influence preferences for family-oriented benefits. And f i n a l l y , Lero, Brockman, Pence, Goelman, and Johnson (1993), u t i l i z i n g data from the 1988 Canadian National Child Care Survey, examined parents' preferences f o r family-supportive benefits and work arrangements. Parents were asked, i n a single open-ended question, to i d e n t i f y one c h i l d - r e l a t e d benefit or arrangement that would best support them i n t h e i r role as parents. A workplace c h i l d care f a c i l i t y and f l e x i b l e work hours were the most preferred arrangements. Other preferred benefits i d e n t i f i e d included the option to work part-time, paid maternity leave, and family-related leaves. Employment c h a r a c t e r i s t i c s - -employment status, employment sector, s k i l l l e v e l , and occupational category--were not found to be s i g n i f i c a n t l y related to preferences for family-oriented benefits. Certain family c h a r a c t e r i s t i c s , however, were associated with such preferences. Preferences were related to the age of the youngest c h i l d i n the home. The most preferred benefit of parents with children under age s i x was a workplace c h i l d care f a c i l i t y while the most 23 preferred benefit for parents with children aged s i x to twelve was f l e x i b l e work hours. Also, the importance of family-oriented benefits to parents declined as the age of the youngest c h i l d increased. Preferences were related to family structure. Single parents preferred f l e x i b l e hours and part-time work more and paid maternity leave less than did parents i n two-parent families. It i s important to note several l i m i t a t i o n s of t h i s study. F i r s t , the vast majority (94%) of study participants were women who were s e l f - i d e n t i f i e d as having primary r e s p o n s i b i l i t y for arranging c h i l d care, which meant that the sampling procedure was possibly biased and that gender differences could not be tested. Second, by asking parents to name the one benefit they f e l t was most important to them, the study was li m i t e d i n terms of the range of benefits that were explored. Nonetheless, t h i s study represents an important step towards the understanding of employee preferences for family-oriented benefits. Summary. The l i t e r a t u r e reviewed involving employee preferences found mixed results concerning gender differences i n preferences for p a r t i c u l a r job attributes such as pay, security, and opportunity for advancement. While research studies of employee preferences for family-oriented work benefits are scant and d i f f e r i n t h e i r methods of assessing preferences, they have yielded results which suggest that women, more than men, prefer job attributes which a s s i s t i n balancing work and family demands. S p e c i f i c a l l y , women tend to have stronger preferences for alternate work arrangements, leave-related p o l i c i e s , c h i l d care benefits, f l e x i b l e benefit packages, and supervisors who are f l e x i b l e i n r e l a t i o n to family needs. Preferences for family-24 oriented benefits were found to be related to family structure, marital status, parental status, household employment configuration, l i f e cycle stage, type of c h i l d care arrangements, dependent care r e s p o n s i b i l i t i e s , work hours, employment sector, job l e v e l , and role c o n f l i c t . The few studies which have examined preferences f o r family-oriented benefits among managers have revealed that female managers f i n d family-oriented benefits more appealing than do male managers. More research i s needed which examines employee preferences for a range of family-oriented benefits and, p a r t i c u l a r l y , the aspects of individuals, families, and organizations which influence them. Work-Family C o n f l i c t Greenhaus (1988, p.25) states that "the multiple role pressures experienced by women and men i n today's society render work-family c o n f l i c t v i r t u a l l y i n e v i t a b l e " . E x i s t i n g l i t e r a t u r e which examines i n depth the c o n f l i c t that occurs between work and family roles i s extensive. For the purposes of t h i s study, the l i t e r a t u r e reviewed on work-family c o n f l i c t concentrated on the differences between men and women i n the amount of c o n f l i c t experienced, the relationship between family involvement and work-family c o n f l i c t , the various antecedents of work-family c o n f l i c t , and family-oriented benefits as means of managing work-family c o n f l i c t . Gender differences i n work-family c o n f l i c t . Several studies have shown work-family c o n f l i c t to be a prevalent phenomenon. For example, Quinn and Staines (1979) found that 43% of women and nearly 3 8% of men who were married with children reported that t h e i r job and family l i v e s c o n f l i c t e d "a l o t " or "somewhat". 25 And, using two d i f f e r e n t but large samples of Canadian employees, Lero (as c i t e d i n Lero & Johnson, 1994) found that 57% of employed mothers reported experiencing either a great deal or a moderate amount of tension i n balancing work and family demands, while MacBride-King (1990) found that nearly 84% of female employees and 76% of male employees reported some stress as a result of balancing the r e s p o n s i b i l i t i e s of home and work. Numerous studies of work-family c o n f l i c t have explored gender differences i n the l e v e l s of c o n f l i c t experienced. These studies have consistently found that women report greater work-family c o n f l i c t than men (Duxbury, Higgins, & Lee, 1994; Duxbury, Higgins, Lee, & M i l l s , 1992; Greenglass, Pantony, & Burke, 1988; Herman & Gyllstrom, 1977; Higgins, Duxbury, & Lee, 1992, 1994; Keith & Schafer, 1980; Wiersma, 1990; Wiersma & Van Den Berg, 1991). Two studies--one by Ottaway and Bhatnagar (as ci t e d i n Wiersma, 1990) and one by Duxbury et a l . (1992)--revealed the same gender difference among managers. One possible reason for such findings i s that women assume more f a m i l i a l r e s p o n s i b i l i t i e s than do men, and women who take on job r e s p o n s i b i l i t i e s tend not to s i g n i f i c a n t l y decrease t h e i r family r e s p o n s i b i l i t i e s (Duxbury et a l . , 19'92) . Another reason why women experience greater work-family c o n f l i c t than men may stem from the facts that the female work role i s more vulnerable to family demands than i s the male work role and "the sex-role norm that women take r e s p o n s i b i l i t y for the family c o n f l i c t s with the norms of the job role " (Duxbury et a l . , 1992, p. 55). Family involvement and work-family c o n f l i c t . Empirical studies of family involvement are r e l a t i v e l y few i n number and 26 li m i t e d i n scope. Those studies which have examined the saliency or c e n t r a l i t y of the family role to one's self-concept have tended to focus on the relationship between l e v e l of family involvement and extent of work-family c o n f l i c t experienced. S p e c i f i c a l l y , high lev e l s of family involvement have been found to be associated with high l e v e l s of work-family c o n f l i c t (Frone, Russell, & Cooper, 1992; Higgins, Duxbury, & Irving, 1992; Loerch, Russell, & Rush, 1989). Moreover, Duxbury and Higgins (1991) found that the relationship between family involvement and work-family c o n f l i c t was moderated by gender; that i s , the relationship was stronger for men than for women. Greenhaus and Beutell (1985) claim that a high degree of family involvement may lead to a high l e v e l of work-family c o n f l i c t i n two ways. F i r s t , high lev e l s of psychological involvement i n the family role may lead to an increase i n the amount of time devoted to that role, thereby making i t d i f f i c u l t to comply with the expectations associated with the work role. Second, high lev e l s of psychological involvement i n the family role may lead one to become mentally preoccupied with the role when attempting to f u l f i l l the demands of the work role. S i m i l a r l y , Hall and Richter (1988) suggest that high work boundary permeability--which characterizes individuals who are highly involved with family--is what leads to increased work-family c o n f l i c t . Such high boundary permeability f a c i l i t a t e s the intrusion of family demands into the work domain. Karambayya and R e i l l y (1992) posited that individuals experiencing work-family c o n f l i c t would respond to the c o n f l i c t by restructuring t h e i r work. That i s , they would make both 27 permanent and temporary arrangements at work, such as working part-time instead of f u l l - t i m e or job sharing, i n order to accommodate the needs of family. They found that psychological involvement i n family roles among dual-earner couples was s i g n i f i c a n t l y p o s i t i v e l y related to work restructuring. Antecedents of work-family c o n f l i c t . S t i l l other studies of work-family c o n f l i c t have documented the antecedents and consequences of such c o n f l i c t . A number of factors which influence the degree to which work and family roles c o n f l i c t have been i d e n t i f i e d . Individual, family, and work setting c h a r a c t e r i s t i c s commonly found to be p o s i t i v e l y related to work-family c o n f l i c t include the number of hours worked (Campbell & Moen, 1992; Gutek, Searle, & Klepa, 1991; Holahan & Gi l b e r t , 1979; Hughes & Galinsky, 1988; Keith & Schafer, 1980; Staines & Pleck, 1984; Voydanoff, 1988; Voydanoff & Kelly, 1984; Wiersma & Van Den Berg, 1991); frequency of overtime (Kelly & Voydanoff, 1985; Voydanoff & Kelly, 1984); working s h i f t s (Burke, 1988; Staines & Pleck, 19 84); time spent commuting to work (Bohen & Viveros-Long, 1981; Wiersma & Van Den Berg, 1991); workload pressure (Galinsky et a l . , 1993; Hughes & Galinsky, 1988; Voydanoff, 1988); work role ambiguity (Voydanoff, 1988); job l e v e l (Higgins, Duxbury, & Lee, 1992); being employed i n the public relations sector (Kelly & Voydanoff, 1985); job tenure (Kelly & Voydanoff, 1985); work-related role stress (Bedeian, Burke, & Moffett, 1988; Burke, 1988; Frone et a l . , 1992); l e v e l of work involvement (Greenhaus & Kopelman, 1981; Greenhaus, Parasuraman, Granrose, Rabinowitz, & Beutell, 1989); the number of hours spent i n family work (Gutek et a l . , 1991; Wiersma & Van 28 Den Berg, 1991); age (Greenhaus et a l . , 1989); education (Greenhaus et a l . , 1989; Voydanoff, 1988); being married (Burke, 1988; Herman & Gyllstrom, 1977); non-work stress (Burke, 1988; Frone et a l . , 1992); the number of children at home (Campbell & Moen, 1992; Duxbury et a l . , 1992; Hughes & Galinsky, 1988; Katz & Piotrkowski, 1983; Keith & Schafer, 1980; Voydanoff, 1988); dependent care demands (MacBride-King, 1990); the number of hours worked by the spouse (Keith & Schafer, 1980); being a single-parent (Kelly & Voydanoff, 1985); being a member of a dual-earner (Kelly & Voydanoff, 1985) or dual-career (Higgins, Duxbury, & Lee, 1992) family; wife's occupational status (Greenhaus & Kopelman, 1981; Kelly & Voydanoff, 1985); and l e v e l of family involvement (Frone et a l . , 1992; Pleck, as c i t e d i n Duxbury & Higgins, 1991). Characteristics of work and family domains found to be negatively related to work-family c o n f l i c t include schedule f l e x i b i l i t y (Bohen & Viveros-Long, 1981; Greenhaus et a l . , 1989; Staines & Pleck, 1984); task autonomy (Galinsky et a l . , 1993; Greenhaus et a l . , 1989); support from supervisors (Galinsky et a l . , 1993; Goff, Mount, & Jamison, 1990; Hughes & Galinsky, 1988; Thomas & Ganster, 1995) and co-workers (Galinsky et a l . , 1993); a supportive organizational culture (Galinsky et a l . , 1993; Warren & Johnson, 1995); supervisor f l e x i b i l i t y (Warren & Johnson, 1995); job tenure (Greenhaus et a l . , 1989); age (Keith & Schafer, 1980; Voydanoff, 1988; Wiersma & Van Den Berg, 1991); amount of free time spent alone, s o c i a l i z i n g with friends, i n family a c t i v i t i e s , and i n community a c t i v i t i e s (Wiersma & Van Den Berg, 1991); family support and cohesion (Holahan & Gi l b e r t , 1979; 29 Wiersma & Van Den Berg, 1991) ; spouse p a r t i c i p a t i o n i n family-work (Hughes & Galinsky, 1988); r e l i a b i l i t y of c h i l d care (Hughes & Galinsky, 1988); s a t i s f a c t i o n with c h i l d care arrangements (Goff et a l . , 1990); as well as the age of the youngest c h i l d l i v i n g at home (Higgins et a l . , 1994; Hughes & Galinsky, 1988; Katz & Piotrkowski, 1983; Kelly & Voydanoff, 1985; Voydanoff & Kelly, 1984). Furthermore, Holahan and G i l b e r t (1979) suggest that assistance i n home maintenance and c h i l d care from outside sources may be inversely related to work-family c o n f l i c t . Family-oriented benefits and work-family c o n f l i c t . Empirical and non-empirical l i t e r a t u r e has also i d e n t i f i e d a number of family-oriented benefits which may help individuals and families manage work-family c o n f l i c t s . Higgins et a l . (1994) suggest that workplace interventions such as f l e x i b l e work arrangements, part-time work, on-site c h i l d care, s i c k - c h i l d care, as well as time management and parenting courses could be of great benefit to employees attempting to balance multiple role r e s p o n s i b i l i t i e s i n that they can increase the control workers have over the work and family interface. A study by Warren and Johnson (1995) demonstrated that aspects of the workplace--specifically a work environment supportive for employees with work-family d i f f i c u l t i e s , and a supervisor who i s f l e x i b l e when family needs a r i s e - - f a c i l i t a t e d employees' a b i l i t y to coordinate work and family demands. Furthermore, the use of family-oriented benefits was found to be associated with fewer worries about adequately f u l f i l l i n g work and family r e s p o n s i b i l i t i e s . S i m i l a r l y , Lero and Johnson (1994), c i t i n g findings from a 30 study conducted by the Canadian Aging Research Network, noted that t a l k i n g with a supervisor, taking a leave, working at home, and a l t e r i n g hours of work were the strategies most frequently used by employed Canadian men and women to deal with work-family c o n f l i c t s . L i t t l e attention, however, has been focussed on assessing employees' preferences for such resources and coping mechanisms and, i n p a r t i c u l a r , on whether or not these preferences vary as a function of the degree of work-family c o n f l i c t experienced. Summary. The reviewed l i t e r a t u r e on work-family c o n f l i c t indicated that c o n f l i c t between work and family roles i s a problem for both men and women; however, women experience such c o n f l i c t to a greater extent than do men. It also revealed the l e v e l of psychological involvement i n the family role, as well as c h a r a c t e r i s t i c s of home and work, to be related to the l e v e l of work-family c o n f l i c t experienced. Although e x i s t i n g l i t e r a t u r e has pointed to workplace family-oriented benefits as possible ways to manage work-family c o n f l i c t , research regarding the relationship between work-family c o n f l i c t and employees' preferences for such benefits was noticeably absent. This i s an area which i s much i n need of further systematic investigation. Conceptual Model A conceptual model propounded by Rice, Near, and Hunt (1979) forms the conceptual basis for the present study. The model i s grounded i n empirical work from the f i e l d s of i n d u s t r i a l / organizational psychology and s o c i a l psychology. Furthermore, the model's delineation of the i n t e r r e l a t i o n s h i p s among work and extra-work domains of l i f e f a c i l i t a t e s an understanding of how 31 work and family environments are intimately interfaced and influence an individual's preferences for family-oriented benefits. Thus, the model provides an appropriate framework within which to investigate the influences of s p e c i f i c i n d i v i d u a l , family, and work c h a r a c t e r i s t i c s on managers' preferences for family-oriented workplace benefits. According to Near, Rice, and Hunt (1980), the relat i o n s h i p between work and extra-work domains i s most e f f e c t i v e l y conceptualized by di v i d i n g each domain into two components--an objective component and a subjective component. The f i r s t component refers to "objective" aspects of the work and extra-work domains - that i s , to the s i t u a t i o n a l conditions or structures associated .with each. The second "subjective" component then refers to individuals' reactions to these structures, including t h e i r attitudes and behaviors. (Near et a l . , 1980, p. 416) Objective s t r u c t u r a l conditions associated with the workplace include job conditions such as pay, work hours, and physical surroundings as well as the nature of the work performed. Objective s t r u c t u r a l conditions associated with the extra-work domain include, among other things, demographic c h a r a c t e r i s t i c s , community variables, and workers' health (Near et a l . , 1980). The work and extra-work s t r u c t u r a l conditions influence each other and also influence behavioural reactions to work and extra-work conditions (Near et a l . , 1980). Subjective behavioural reactions to work include ind i v i d u a l s ' attitudes toward or feelings about work such as the degree of s a t i s f a c t i o n with the job, and work behaviours such as rates of absenteeism and turnover. Subjective behavioural reactions of the extra-work type are workers' attitudes and behaviours outside the workplace and include factors such as ove r a l l l i f e s a t i s f a c t i o n , the degree to which work and family-are central l i f e i n terests, and p a r t i c i p a t i o n i n s o c i a l roles (Near et a l . , 1980). Work and extra-work behavioural reactions influence one another (see Figure 1) . An important and underlying assumption of the model i s that s o c i a l structures are more l i k e l y to influence i n d i v i d u a l behaviour and attitudes than vice versa (Near et a l . , 1980). Very few researchers examining preferences for family-oriented benefits have s p e c i f i e d a conceptual framework for t h e i r studies. Of the studies reviewed, only three (Higgins, Duxbury, & Lee, 1992; Kossek, 1990; Wiersma, 1990) employed conceptual models and each one developed i t s own model from hypothesized relationships. Based on the review of the l i t e r a t u r e pertaining to job att r i b u t e preferences and work-family c o n f l i c t , i t can be concluded that further studies are needed which examine relationships among objective aspects of work and non-work domains, individuals' attitudes and behaviours outside of the workplace, and individuals' feelings about t h e i r work. Near et a l . (1980) contend that an examination of these relationships w i l l benefit from reference to a single overarching scheme or conceptual model of work and family linkages, the use of which w i l l contribute to greater understanding and deeper insights about the nature of the work/non-work relationship. Many of the relationships depicted by Rice et a l . (1979) i n th e i r model are examined i n t h i s study. The extra-work domain examined i s the family. Within the family domain, gender i s the primary objective condition studied while family involvement and 33 Objective Structural Conditions Extra-work variables * Gender * Control variables Objective Structural Conditions Work variables * Control variables Subjective Behavioural Reactions Extra-work variables * Family involvement * Work-family c o n f l i c t * Control variables 77 Subjective Behavioural Reactions Work variables * Preferences for family-oriented benefits Figure 1. The conceptual model adapted from Rice, Near, and Hunt (1979) . 34 work-family c o n f l i c t are the subjective conditions examined. Subjective conditions of the work s i t u a t i o n are conceptualized as preferences for family-oriented benefits. Additionally, a number of i n d i v i d u a l , family, and work c h a r a c t e r i s t i c s comprising the potent i a l control variables for the study represent objective conditions of work and family domains as well as subjective conditions of the family domain (see Figure 1). This research lends to the development of the Rice et a l . (1979) model i n that i t explores the relationships that exist among factors within a p a r t i c u l a r group of variables (e.g., between two d i f f e r e n t subjective extra-work v a r i a b l e s ) . Hypotheses Few studies have s p e c i f i c a l l y investigated the preferences of employees for job attributes which enable them to ease work-family tensions and even fewer have examined gender differences i n such preferences. Of those studies which have compared men and women i n terms of the importance they ascribe to p a r t i c u l a r family-oriented work benefits, most have found that women prefer such benefits more than men. For example, women have been shown to prefer alternate work arrangements such as part-time and at-home work (Galinsky et a l . , 1993; Higgins, Duxbury, & Lee, 1992; Kossek, 1990; Louis Harris & Associates, 1981; Wiersma, 1990), c h i l d care benefits such as on-site daycare (Goldberg et a l . , 1989; Lee et a l . , 1992; Wiersma, 1990), family-related leaves (Goldberg et a l . , 1989; Lee et a l . , 1992; Louis Harris & Associates, 1981), and f l e x i b l e benefit packages (Louis Harris & Associates, 1981) more than men. Studies of preferences for family-oriented benefits which have compared male and female 35 managers have yielded s i m i l a r gender differences (e.g., Higgins, Duxbury, & Lee, 1992; Lee et a l . , 1992). Konrad and Langton (1993) further support the contention that gender differences exist with respect to work aspect preferences by stating that ...both the s o c i o l o g i c a l and the psychological perspectives on preference formation lead to the conclusion that men and women are l i k e l y to have d i f f e r e n t job preferences. The so c i o l o g i c a l perspective predicts t h i s to the extent that men and women are s o c i a l i z e d d i f f e r e n t l y while they are growing up. The s o c i a l psychological perspective predicts t h i s to the extent that men and women have d i f f e r e n t jobs within the workplace, (p. 7) Thus, i t i s hypothesized that: HI: Female managers w i l l show a greater preference for family-oriented benefits than t h e i r male counterparts. The reviewed l i t e r a t u r e revealed no studies s p e c i f i c a l l y examining the impact of family involvement on preferences f o r job att r i b u t e s . However, Brett and Yogev (1988) suggest that individuals are more l i k e l y to want to restructure work--to make both permanent and temporary arrangements at work i n order to accommodate family--when they are more involved i n t h e i r family roles. Indeed, Karambayya and R e i l l y (1992) found family involvement to be p o s i t i v e l y related to work restructuring for both men and women. Only one study was located that explored the relati o n s h i p between work-family c o n f l i c t and job attr i b u t e preferences ( i . e . , Wiersma, 1990). Wiersma (1990) found that men and women who reported higher l e v e l s of role c o n f l i c t expressed a greater desire for parental support job att r i b u t e s . Therefore, i t i s hypothesized that: H2: Family involvement w i l l be p o s i t i v e l y associated with preferences for family-oriented benefits. 36 H3: Work-family c o n f l i c t w i l l be p o s i t i v e l y associated with preferences for family-oriented benefits. Results from Wiersma's (1990) study suggest that differences i n the job attr i b u t e preferences of men and women may be explained by p a r t i c u l a r mediating variables. She found that role c o n f l i c t mediated the relationship between gender and preferences for parental support a t t r i b u t e s . That i s , role c o n f l i c t accounted for the r e l a t i o n between gender and preferences for family-oriented personnel p o l i c i e s . In other words, men and women s i g n i f i c a n t l y d i f f e r e d i n the extent to which they preferred parental support attributes; t h i s difference was due, in part, to the fact that men and women d i f f e r e d i n the amount of role c o n f l i c t experienced which was, i t s e l f , related to preferences. Wiersma (1990) notes that [t]he r e l a t i v e l y higher l e v e l s of work-home role c o n f l i c t reported by women may explain why they have been known to value c e r t a i n job a t t r i b u t e s d i f f e r e n t l y from men. Women, more aware of c o n f l i c t between work and non-work spheres of l i f e , may set d i f f e r e n t p r i o r i t i e s for themselves from men. (p. 233) Likewise, the higher l e v e l s of work-family c o n f l i c t reported by individuals who are highly involved i n t h e i r family roles may explain t h e i r greater preference for family-oriented job attribu t e s . Sarbin and A l l e n (1968) contend that increases i n lev e l s of role c o n f l i c t lead to increases i n behaviours directed toward resolving such c o n f l i c t . Consistent with t h i s contention, individuals experiencing high l e v e l s of work-family c o n f l i c t should want to reduce that c o n f l i c t more so than individuals experiencing less work-family c o n f l i c t . Individuals experiencing high lev e l s of c o n f l i c t , therefore, may express a stronger preference for family-oriented benefits since such benefits are often regarded as possible ways to manage or reduce work-family c o n f l i c t (e.g., Higgins et a l . , 1994). Thus, the difference between men and women i n terms of t h e i r preferences for family-oriented benefits may be explained by the fact that men and women d i f f e r i n the amount of work-family c o n f l i c t they experience. And the difference between individuals with high l e v e l s of family involvement and individuals with low le v e l s of family involvement i n terms of t h e i r preferences for family-oriented benefits may be explained by the fact that individuals of varying l e v e l s of family involvement d i f f e r i n the amount of work-family c o n f l i c t they experience. To test whether work-family c o n f l i c t i s a mediating variable and, therefore, whether i t explains the relationships between gender and preferences for family-oriented benefits and between family involvement and preferences for family-oriented b e n e f i t s - - i f such differences are found to e x i s t - - i t i s hypothesized that: H4: Female managers w i l l experience greater work-family c o n f l i c t than t h e i r male counterparts. H 5 : When the ef f e c t s of gender on work-family c o n f l i c t (a) and of work-family c o n f l i c t on preferences for family-oriented benefits (b) are controlled, the s i g n i f i c a n t r e l a t i o n s h i p between gender and preferences (c) w i l l f a i l to be s i g n i f i c a n t (see Figure 2 ) . In other words, work-family c o n f l i c t w i l l mediate the relationship between gender and preferences for family-oriented benefits. H6: Family involvement w i l l be p o s i t i v e l y associated with work-family c o n f l i c t . H 7 : When the ef f e c t s of family involvement on work-family c o n f l i c t (d) and of work-family c o n f l i c t on preferences for family-oriented benefits (e) are controlled, the s i g n i f i c a n t r e l a t i o n s h i p between family involvement and preferences (f) w i l l f a i l to be s i g n i f i c a n t (see Figure 2). In other words, work-family c o n f l i c t w i l l mediate the re l a t i o n s h i p between family involvement and preferences for family-oriented benefits. Work-Family C o n f l i c t Work-Family C o n f l i c t Figure 2. Hypothesized relationships among research variables. 39 Chapter II Method Sample and Data C o l l e c t i o n The present study was part of a larger study e n t i t l e d The Career Development Survey of Managers and Professionals (Langton, 1995). The data were co l l e c t e d from male and female managers employed i n organizations located throughout the Lower Mainland of B r i t i s h Columbia. The managers were selected from a large data base purchased from a Vancouver marketing firm. Information used to construct the data base was gathered from licensed B.C. businesses through telephone and fax contact i n mid-1994. The information about the organizations provided by the data base included, among other things, the names, business addresses, and job positions of potential respondents. The entire data base contained the names of 111,306 company owners and managers, 83,770 (75%) of which were men. Over 85,000 firms were represented i n the data base. While the l i s t of firms provided by the data base was diverse and the l i s t of managers sizable, the data base was not in c l u s i v e of a l l Lower Mainland firms or managers. The c h a r a c t e r i s t i c s of the excluded firms and managers are not known. The sampling frame for t h i s study was constructed to include those individuals i d e n t i f i e d by the data base as working for organizations employing 11 or more individuals and as holding middle-level or senior management positions within t h e i r places of employment. These selection c r i t e r i a were designed to i d e n t i f y those individuals who were l i k e l y to occupy a supervisory r o l e . Approximately equal numbers of men (N=499) and 40 women (N=488) managers were randomly drawn from the sampling frame. The selected managers were mailed a questionnaire which included questions about preferences for various job attributes, current job attributes, career history, family l i f e , and demographic information. Of the 9 87 surveys that were mailed out, 178 were returned completed, 47 were not deliverable and 9 were mailed to managers who were eithe r no longer with t h e i r companies or away f o r an extended period of time. Of the surveys which were completed and returned, 91 (51%) were from female managers and 87 (49%) were from male managers. The response rate, then, was about 19% of those e l i g i b l e to pa r t i c i p a t e i n the study. Because the response rate to the i n i t i a l mail-out (conducted i n March, 1995) was low, a second mail-out was conducted s i x months l a t e r i n an attempt to augment the sample size and improve the rate of response. A sampling frame for the second mail-out was constructed using the same data base and selection c r i t e r i a as was used for the f i r s t mail-out. Those individuals who were mailed a questionnaire i n the f i r s t mail-out were excluded from the sampling frame. An additional 249 male and 146 female managers were selected and mailed a questionnaire. The c r i t e r i a for s e l e ction from the o r i g i n a l data base coupled with the fact that there were fewer women i n the data base resulted i n unequal numbers of male and female managers being drawn from the frame. Of the 395 questionnaires that were mailed out, 87 were returned completed, 8 were not deliverable, and 4 were mailed to managers who were no longer with t h e i r companies. One questionnaire was unusable due to substantial missing data. Thirty-seven (43%) of 41 the usable surveys were from female managers and 49 (57%) were from male managers. The response rate to the second wave of data c o l l e c t i o n was about 23% of those e l i g i b l e to p a r t i c i p a t e i n the study. A post-card follow-up to the second mail-out yielded no additional responses. In t o t a l , 264 managers comprised the sample for th i s study. The o v e r a l l response rate was about 20%. While t h i s response rate i s low, i t i s not at y p i c a l of studies which have sought to examine s i m i l a r populations (e.g., Kamerman, Kahn, & Kingston, 1983) . Fifty-one percent (n = 135) of the respondents were male and 49% (n = 129) were female. The ages of respondents ranged from 23.42 years to 65.17 years, with the average age being 45.11 years. Thirty-two percent of the sample reported that they held no post-secondary degrees or c e r t i f i c a t e s ; 26% reported that the highest l e v e l of education attained was a Bachelor's degree and 18% reported that they had attained a Master's degree. The sample was predominantly Caucasian i n e t h n i c i t y (92%). Most of the respondents were married (80%) and had children (73%). Of those respondents who had children, the average number of children was 2.25 and the average age of the f i r s t c h i l d was 18.42 years. The annual s a l a r i e s of respondents ranged from $20,000 to $420,000; the average annual salary was $90,509 (Mdn = $79,000). The average annual salary for male managers was $109,660 (Mdn = $90,000) and for female managers, $71,044 (Mdn = $62,000). Two-thirds of the respondents had an employed spouse or partner. The average t o t a l household income for t h i s sample was $127,814 (Mdn = $115,000). Respondents worked an average of - 42 50.28 hours per week. A comparison of the demographic ch a r a c t e r i s t i c s of male and female managers revealed differences and s i m i l a r i t i e s which p a r a l l e l those documented i n the management l i t e r a t u r e (see Parasuraman & Greenhaus, 1993, for a review). For example, the majority of female managers earned lower s a l a r i e s and were less l i k e l y to be married and have children than t h e i r male counterparts. For a detailed demographic p r o f i l e see Appendix B. Measures The quasi-experimental design of t h i s study i s a f a c t o r i a l design, involving three independent variables and one dependent variable. The variables of the study were constructed from information regarding subjects' preferences for family-oriented benefits, l e v e l s of family involvement and work-family c o n f l i c t , as well as from demographic information concerning work and family provided by the questionnaires. A l i s t i n g of the questions used to gather such information can be found i n Appendix A. Dependent Variable Preferences for family-oriented benefits were measured with a 10-item question which asked respondents to indicate how important the a v a i l a b i l i t y of p a r t i c u l a r job conditions i s to them. The job conditions examined were work schedule f l e x i b i l i t y , part-time employment with pro-rated benefits, work location f l e x i b i l i t y , on- or near-site daycare f a c i l i t i e s , f l e x i b l e benefit programs, leaves for personal/family reasons, workplace counselling and information services designed to a s s i s t i n managing work and family l i f e , r elocation assistance programs, 43 an organizational culture supportive and understanding of family-r e s p o n s i b i l i t i e s , and a supervisor who i s f l e x i b l e when family-needs a r i s e . Each of these job conditions represents one way organizations can a s s i s t employees i n balancing work and family demands. Because no scale of items measuring preferences for family-oriented benefits could be located, a group of items was selected which represented the wide range of family-related job benefits outlined by Paris (19 89). Furthermore, the family-oriented benefits selected correspond to those work-family issues which, according to Galinsky and Stein (1990), are of greatest concern to workers with family r e s p o n s i b i l i t i e s . The items comprising the preference measure were pre-tested for c l a r i t y , conciseness, and relevance to the target sample among faculty members and graduate students i n the areas of Commerce and Business Administration and Family Studies at a western Canadian university. Experts from these p a r t i c u l a r f i e l d s of study were asked to pre-test the preference measure as family-oriented benefits represent both a business and a family concern. Written comments were returned by 20 i n d i v i d u a l s . Pre-test results indicated that the items were clear, relevant, and encompassed the range of benefits available to employees. The importance of each job condition was assessed with a 4-point Likert-type response scale ranging from very important (1) to very unimportant (4). The items were recoded (1=4, 4=1) and summed to y i e l d a t o t a l score, with a high score i n d i c a t i n g a strong preference for family-oriented benefits. Internal consistency, as measured by Cronbach's alpha, was .81 for the 10-item index. In order to determine whether the 44 items constituted a single factor or whether d i f f e r e n t dimensions of preferences f o r family-oriented benefits were being measured, a p r i n c i p a l components factor analysis with varimax rotation was performed on the set of 10 items. A l l items loaded on a single factor at or above .50 with minimal cross loadings before the rotation was performed. Varimax rotation produced three d i s t i n c t factors or dimensions. The item measuring preferences for on- or near-site c h i l d care d i d not load cleanly on any factor. The three factors were not interpretable conceptually and, when examined i n d i v i d u a l l y for r e l i a b i l i t y , were not as i n t e r n a l l y consistent as the set of 10 items. Moreover, because an o v e r a l l measure of preferences was desired as opposed to measures of di f f e r e n t dimensions of preferences, a l l of the items were compiled to form a single index of preferences for family-oriented benefits. Although the item measuring preferences for on- or near-site c h i l d care did not have a negative ef f e c t on the r e l i a b i l i t y of the preference index nor did i t f a i l to correlate with the entire scale, i t was dropped from the index as i t was not applicable to a l l subjects (e.g., those without children or those with older children). Cronbach's alpha for the remaining nine items of the index was .79 for t h i s sample. Independent Variables The f i r s t independent variable of the study--gender--was measured by a single item asking "What i s your sex?". This item provided two response categories: male and female. The second independent variable--family involvement--was measured by a scale consisting of three items taken from the 45 Quality of Employment Survey (Quinn & Shepard, 1974; Quinn & Staines, 19 79). The scale included items such as "The most important things that happen to me involve my family" which measured the o v e r a l l psychological importance of the family role rather than the amount of time devoted to i t . Respondents were asked to indicate, u t i l i z i n g a 4-point Likert-type response format ranging from strongly agree (1) to strongly disagree (4), t h e i r l e v e l of agreement with each item. The r e l i a b i l i t y of t h i s 3-item scale, when used i n a study of Stanford MBA alumni (Konrad & Langton, 1993), was good (alpha=.82). A l l three items were recoded (1=4, 4=1) so that high scores suggested that the respondent was highly involved i n family roles. Scores for each item were computed and summed to y i e l d an o v e r a l l family involvement score. The r e l i a b i l i t y score for t h i s scale, as measured by Cronbach's alpha, was .72. A p r i n c i p l e components factor analysis with varimax rotation revealed the scale to be unidimensional with a l l of the items loading on a single factor at or above .63. The t h i r d independent variable--work-family conflict--was measured by a 6-item scale consisting of four items from Kopelman, Greenhaus, and Connolly's (1983) 8-item scale of i n t e r r o l e c o n f l i c t and two items from Bohen and Viveros-Long's (1981) Job-Family Role Str a i n scale. Sample items included "My work takes up time that I'd l i k e to spend with my family" and "I have a good balance between my job and my family time". These items were selected for use on the survey instrument because they were deemed most applicable to the target sample. For instance, several scales of work-family c o n f l i c t (e.g., Bohen & Viveros-46 Long, 1981; Holahan & G i l b e r t , 1979) include multiple items which refer to s p e c i f i c roles, such as "spouse" and "parent"; such roles may not be held by many individuals i n t h i s study's sample. Respondents were asked to indicate, on a 5-point Likert-type response scale ranging from strongly agree (1) to strongly disagree (5) , the extent to which they agree that each statement r e f l e c t s t h e i r current job s i t u a t i o n . With respect to t h e i r 8-item scale, Kopelman et a l . (19 83) have reported evidence of construct v a l i d i t y and the r e l i a b i l i t y estimate (alpha) of t h i s scale was reported as .88 by Beutell and O'Hare (1987). Negatively worded items (2, 3, 4, 6) were recoded (1=5, 5=1) so that high scores indicated high lev e l s of work-family c o n f l i c t . An o v e r a l l work-family c o n f l i c t score was calculated by summing the scale items. Internal consistency of the scale, as measured by Cronbach's alpha, was .89. A p r i n c i p l e components factor analysis with varimax rotation revealed that the scale was unidimensional with a l l of the items loading on a single factor at or above .78. A factor analysis performed on a l l of the items comprising the scales for preferences, family involvement, and work-family c o n f l i c t revealed that the scales were d i s t i n c t . That i s , the three groups of items forming the scales loaded on separate factors. Control Variables Many of the remaining items on the questionnaire measured demographic c h a r a c t e r i s t i c s of the respondents, some of which were used as control variables. A va r i e t y of i n d i v i d u a l , family, and work c h a r a c t e r i s t i c s suggested by the l i t e r a t u r e to be important predictors of work-family c o n f l i c t and preferences were analysed to see i f they were correlated with each of these two variables i n t h i s study and, therefore, i f they needed to be included as controls i n hypothesis testing. The variables entered into the c o r r e l a t i o n analysis included l e v e l of work involvement, the number of hours worked per week, the number of hours worked at home per week, employment sector, job tenure, salary, age, marital status, the presence of an employed spouse/partner, the number of weekly hours worked by the spouse/partner, number of children, the number of weekly hours spent performing household tasks, the number of weekly hours devoted to elder care and c h i l d care r e s p o n s i b i l i t i e s , r e s p o n s i b i l i t y for household labour, and whether or not the respondent has paid help with housework r e s p o n s i b i l i t i e s . Work involvement was measured by a scale consisting of three items taken from the Quality of Employment Survey (Quinn & Shepard, 1974; Quinn & Staines, 1979). The scale included items such as "The most important things that happen to me involve my work" which measured the o v e r a l l psychological importance of the work role. Respondents were asked to indicate, u t i l i z i n g a 4-point Likert-type response format ranging from strongly agree (1) to strongly disagree (4), t h e i r l e v e l of agreement with each item. The r e l i a b i l i t y of t h i s three item scale, when used i n a study of Stanford MBA alumni (Konrad & Langton, 1993), was adequate (alpha=.65). A l l three items were recoded (1=4, 4=1) so that high scores suggested that the respondent was highly involved i n work roles. Scores for each item were computed and summed to y i e l d an o v e r a l l work involvement score. An item 48 analysis performed on the three items resulted i n the second item being dropped from the scale as the r e l i a b i l i t y c o e f f i c i e n t increased from .58 to .73 when the second item was deleted from analysis. A p r i n c i p l e components factor analysis with varimax rotation revealed that the scale was unidimensional with the items loading on a single factor at or above .82. Those factors found to be s i g n i f i c a n t l y correlated with work-family c o n f l i c t were held constant for two reasons: (a) i n order to di s t i n g u i s h the e f f e c t of work-family c o n f l i c t per se on the dependent variable from the effects of the antecedents or correlates of work-family c o n f l i c t , and (b) to examine the impact of gender and family involvement alone on the amount of work-family c o n f l i c t experienced. Those factors found to be s i g n i f i c a n t l y correlated with preferences for family-oriented benefits were held constant so that the d i s t i n c t influences of gender, family involvement, and work-family c o n f l i c t on preferences could be examined. Job l e v e l , a variable found i n a p r i o r study (Higgins, Duxbury, & Lee, 1992) to influence preferences for family-oriented benefits, was controlled for through the design of the study which focussed on managerial/professional men and women only. 49 Chapter III Results Preliminary Analyses Before the hypotheses of the study were tested, several analyses were performed i n order to assess: (a) whether the information from the two d i f f e r e n t waves of data c o l l e c t i o n could be merged into one data set; (b) whether the respondents d i f f e r e d s u b s t a n t i a l l y from the non-respondents; (c) the univariate d i s t r i b u t i o n s of the dependent and independent variables; and (d) which, i f any, of the potential control variables needed to be included as s t a t i s t i c a l controls i n the hypothesis testing. Comparison of the Two Sets of Data To ascertain whether the respondents from the two d i f f e r e n t waves of data c o l l e c t i o n were si m i l a r enough to be combined into a single sample, t-tests and chi-square analyses were conducted to compare the two groups on a l l of the study's variables as well as on additional demographic variables such as race, education l e v e l , and job industry. The comparison indicated that there was s t a t i s t i c a l support for combining the two groups i n subsequent analyses. Only one s i g n i f i c a n t difference was found between the two groups of respondents at the p_ < .05 l e v e l : respondents i n the f i r s t wave worked at smaller firms than did respondents i n the second wave (t = 2.64, p_ < .01) . Comparison of Respondents and Non-Respondents Information from the data base regarding gender, business size, type of business ( i . e . , l o c a l o f f i c e , branch o f f i c e , head o f f i c e , regional head o f f i c e ) , industry, and business status ( i . e . , corporation, partnership, independent, franchise, home-50 based) about respondents was compared to s i m i l a r information about non-respondents to see whether a non-response bias may-ex i s t . A t o t a l of f i v e analyses (t-tests and chi-square analyses as appropriate) were used to compare the two populations. A Bonferroni procedure was performed and a p r o b a b i l i t y l e v e l of .01 was applied to the test r e s u l t s . The sample used for analysis did not d i f f e r s u b s t a n t i a l l y from those that were excluded from analysis. Only one s i g n i f i c a n t difference was found between the groups: individuals who responded worked at larger organizations (t = -3.34, p_ < .01) than those who did not respond. Thus, based on the analysis of select information, i t appears that a non-response bias does not pose a threat to the v a l i d i t y of t h i s study (see Appendix C). Univariate D i s t r i b u t i o n s The univariate d i s t r i b u t i o n s of the i n t e r v a l l e v e l dependent and independent variables were examined p r i o r to hypothesis testing. The average score on the 9-item scale measuring the dependent variable--preferences for family-oriented benefits--was 2.71 (SD = .52), where the possible range of responses was from 1 to 4. There was v a r i a t i o n i n the d i s t r i b u t i o n of scores around the mean. The d i s t r i b u t i o n of the dependent variable was neither s i g n i f i c a n t l y skewed (skewness = .22, SE skewness = .15) nor s i g n i f i c a n t l y k u r totic (kurtosis = .06, SE kurtosis = .30) and i t approximated a normal d i s t r i b u t i o n . The average score on the 3-item scale measuring the f i r s t i n t e r v a l l e v e l independent variable--family involvement--was 3.03 (SD = .49), where possible responses ranged from 1 to 4. The d i s t r i b u t i o n was neither s i g n i f i c a n t l y skewed (skewness = .06, SE 51 skewness = .15) nor s i g n i f i c a n t l y k u r t o t i c (kurtosis = -.34, SE kurtosis = .30) and i t approximated a normal d i s t r i b u t i o n . The average score on the 6-item scale measuring the second i n t e r v a l l e v e l independent variable--work-family conflict--was 2.94 (SD = .86), where possible responses ranged from 1 to 5. The d i s t r i b u t i o n was neither s i g n i f i c a n t l y skewed (skewness = .20, SE skewness = .15) nor s i g n i f i c a n t l y k u r t o t i c (kurtosis = -.43, SE kurtosis = .30) and i t approximated a normal d i s t r i b u t i o n . Determination of Control Variables Correlations between the potential control variables and both preferences and work-family c o n f l i c t were computed to ascertain which variables needed to be included as s t a t i s t i c a l controls i n the te s t i n g of the hypotheses. Nominal l e v e l variables such as marital status were recoded as dichotomous variables p r i o r to c o r r e l a t i o n analysis. The Pearson c o r r e l a t i o n c o e f f i c i e n t s that resulted revealed that seven variables were s i g n i f i c a n t l y related to preferences for family-oriented benefits at a p_ <, .01 l e v e l : work hours, salary, age, time spent i n dependent care a c t i v i t i e s , hours spent performing household tasks, r e s p o n s i b i l i t y for household labour, and spouse's work hours. In order to reduce the number of control variables, only those variables correlated with preferences at a p_ s .001 l e v e l were controlled for i n subsequent analyses involving preferences for family-oriented benefits (see Table 1). Three variables were found to be s i g n i f i c a n t l y related to work-family c o n f l i c t at a p. s .01 l e v e l : work hours, hours spent working at home, and hours spent i n dependent care a c t i v i t i e s (see Table 2). These 52 Table 1 Correlations Between Potential Control Variables and Preferences for Family-Oriented Benefits Potential Control Variables n Work Involvement Work Hours Hours Worked at Home Employment Sector Job Tenure Salary Age Marital Status Presence of Employed Spouse/Partner Spouses Work Hours Number of Children Hours Spent i n Household Tasks Hours Devoted to Dependent Care Responsibility for Household Labour Help with Housework 256 258 249 252 238 241 256 258 253 225 258 252 251 245 254 - . 04 - .20*** . 03 . 01 - .11 - . 23*** - .23*** - .03 . 14 .21** .04 . 16** 27*** - .26*** - .01 a Number varies due to missing data. **p_ < .01. ***p_ < .001. 53 Table 2 Correlations Between Potential Control Variables and Work-Family C o n f l i c t Potential Control Variables n a r Work Involvement 259 . 01 Work Hours 260 .33*** Hours Worked at Home 251 _ 24*** Employment Sector 254 - .00 Job Tenure 241 .03 Salary 244 . 04 Age 258 - .04 Marital Status 260 . 11 Presence of Employed Spouse/Partner 255 - .05 Spouses Work Hours 228 - .09 Number of Children 260 .10 Hours Spent i n Household Tasks 254 - .02 Hours Devoted to Dependent Care 253 _ 17* * Responsibility for Household Labour .' 252 .03 Help with Housework 255 - .13 a Number varies due to missing data. **p_ < .01. ***p_ < .001. 54 variables were controlled for i n subsequent analyses involving work-family c o n f l i c t . The co r r e l a t i o n matrix was also examined for possible m u l t i c o l l i n e a r i t y among control variables. Although some of the control variables were s i g n i f i c a n t l y intercorrelated, none were so highly i n t e r c o r r e l a t e d as to warrant t h e i r exclusion from further analysis (see Table 3). Preferences for S p e c i f i c Family-Oriented Benefits The family-oriented benefit most preferred by t h i s sample was a supervisor who i s f l e x i b l e when family needs aris e (M = 3.30, SD = .67). This benefit was also the most desired benefit among male managers (M = 3.24, SD = .64) and among female managers (M = 3.36, SD = .70). The family-oriented benefit preferred the least by t h i s sample was part-time employment with pro-rated benefits (M = 2.04, SD = .96). The least preferred benefit among men was, s i m i l a r l y , part-time employment with pro-rated benefits (M = 1.79, SD = .86) while, among women, i t was relocation assistance programs (M = 2.27, SD = .94). The preferences of male and female managers f o r family-oriented benefits are detailed i n Appendix D. Hypothesis Testing Analyses of variance (ANOVAs) and analyses of covariance (ANCOVAs) were used to test hypotheses 1 and 4. These procedures were run once i t was determined that--with a single exception--the underlying assumptions of ANOVA--that the dependent variable be approximately normally distributed, the c e l l sizes be equal, the variances be equal, the independent variable be nominal, and the dependent variable be interval--had not been s i g n i f i c a n t l y v i o l a t e d . Only i n the case of hypothesis 1 was an assumption of 55 Table 3 Intercorrelations Between Control Variables Control Variables 2 3 4 1. Work Hours 29*** .06 14 18** .34*** 2 . Salary 35*** -.11 .36*** .18** 3 . Age 37*** 27*** .08 4. Hours Devoted to Dependent Care 04 .09 5. Responsibility for Household Labour 02 6. Hours Worked at Home **p < .01. ***p_ < .001 56 ANOVA--that of equality of variance--not met for the dependent variable. However, because ANOVA i s a robust s t a t i s t i c a l test and because multiple assumptions were not violated/ heterogeneity of variance does not pose a serious threat to the v a l i d i t y of the study (Kerlinger, 1986). Bivariate and multiple regression analyses were used to test hypotheses 2, 3 and 6. Regression analyses were used i n the process of testing the hypotheses s t a t i s t i c a l l y once i t was determined that the assumptions of regression--that the d i s t r i b u t i o n s of the dependent variable be normal for each value of the independent variables, that the variances of these d i s t r i b u t i o n s be equal, and that the independent and dependent variables be l i n e a r l y related--had not been s i g n i f i c a n t l y v i o l a t e d . The assumption of regression that a l l observations be independently selected was not v i o l a t e d by t h i s study. HI: Female managers w i l l show a greater preference for family-oriented benefits than t h e i r male counterparts. An ANOVA and an ANCOVA were used to test the f i r s t hypothesis. The mean preference score for men was 2.59 (SD = .45, n = 134) and for women, 2.84 (SD = .55, n = 124). Gender had a s i g n i f i c a n t main e f f e c t on preferences for family-oriented benefits, F (1, 256) = 16.02, p_ < .001; i t explained nearly 6% of the variance i n the dependent variable. Thus, the results provide support for the hypothesis that female managers prefer family-oriented benefits more than do male managers. Upon concluding that the relationship between gender and preferences was s i g n i f i c a n t , those control variables found to be s i g n i f i c a n t l y correlated with preferences for family-oriented 57 benefits were added to the analysis i n order to see i f the relationship between gender and preferences was spurious. The mean preference score for men was 2.61 (SD = .45, n = 118) and for women, 2.86 (SD = .54, n = 111). The mean preference score adjusted for the entry of the covariates was 2.69 for men and 2.78 for women. The F (1, 222) ra t i o s for the covariates measuring time spent i n dependent care a c t i v i t i e s , r e s p o n s i b i l i t y for household labour, age, salary, and work hours were 8.98 (p_ < .01), 7.73 (p < .01), 1.01 (p = .32), .74 (p = .39), and 4.01 (p < .05) respectively. The entire group of covariates had a si g n i f i c a n t e f f e c t on the dependent variable, F (5, 222) = 8.62, p < .001. Examination of the significance l e v e l of the main ef f e c t showed that the s i g n i f i c a n t e f f e c t gender had on preferences before the controls were introduced, disappeared a f t e r the controls were introduced, F (1, 222) = .91, p = .34. Further analyses indicated that the covariates explained approximately 16% of the variance i n the dependent variable, while gender explained a mere .34%. Therefore, when control variables are entered into the analysis, the hypothesis that female managers prefer family-oriented benefits more than do male managers i s not supported. It appears that the rel a t i o n s h i p between gender and preferences may be spurious. H2: Family involvement w i l l be p o s i t i v e l y associated with preferences for family-oriented benefits. Hypothesis 2 was tested with b i v a r i a t e and multiple regression analyses. Regressing preferences for family-oriented benefits on family involvement indicated that family involvement explained a s i g n i f i c a n t portion of the variance i n the dependent 58 variable, F (1, 254) = 9.12, R2 = .03, p < .01. In other words, the higher the l e v e l of family involvement, the stronger the preference for family-oriented benefits. This finding provides support for the hypothesized relationship between family involvement and preferences. After examining the relationship between family involvement and preferences, the control variables measuring time spent i n dependent care a c t i v i t i e s , r e s p o n s i b i l i t y for household labour, age, salary, and work hours were added to the analysis. As indicated by the R square change, family involvement continued to make a s i g n i f i c a n t contribution to preferences when entered into the multiple regression equation a f t e r the covariates, R2 change = .02, p < .05. The variables entered into the regression equation accounted for a s i g n i f i c a n t proportion of explained variance i n preferences, R2 = .16, F (6, 232) = 8.52, p < .001. Only three of the variables entered into the equation--r e s p o n s i b i l i t y for household labour (Beta = -.19, p < .01), time spent i n dependent care a c t i v i t i e s (Beta = .16, p < .05), and family involvement (Beta = .15, p < .05)--had s i g n i f i c a n t Beta c o e f f i c i e n t s , of which r e s p o n s i b i l i t y for household labour was the most important i n influencing preferences (see Table 4). Hence, the results provide support for the po s i t i v e relationship between family involvement and preferences for family-oriented benefits. H3: Work-family c o n f l i c t w i l l be p o s i t i v e l y associated with preferences for family-oriented benefits. Hypothesis 3 was tested using b i v a r i a t e and multiple regression analyses. Regressing preferences for family-oriented 59 Table 4 Summary of Regression of Preferences for Family-Oriented Benefits on Control Variables and Family Involvement Variable 3 Work Hours -.10 Age -.08 Responsibility for -.19** Household Labour Hours Devoted to .16* Dependent Care Salary -.09 Family Involvement .15* Note. Work hours, age, r e s p o n s i b i l i t y for household labour, hours devoted to dependent care, and salary were entered together i n the f i r s t step. Family involvement was entered i n the next step. R 2 = .14 for Step 1; A R 2 = .02 for Step 2 (p < .001 and p_ < .05 respectively) . *p_ < . 05 . **p < . 01. 60 benefits on work-family c o n f l i c t showed there to be a s i g n i f i c a n t l i n e a r association between work-family c o n f l i c t and the dependent variable, F (1, 253) =9.51, R2 = .03, p < .01. In other words, the higher the l e v e l of work-family c o n f l i c t experienced, the stronger the preference for family-oriented benefits. This finding provides support for the hypothesized rela t i o n s h i p between work-family c o n f l i c t and preferences. After examining the relationship between work-family c o n f l i c t and preferences, the control variables measuring time spent i n dependent care a c t i v i t i e s , r e s p o n s i b i l i t y for household labour, age, salary, work hours, and hours spent working at home were added to the analysis. Even a f t e r the addition of control variables, work-family c o n f l i c t accounted for a s i g n i f i c a n t portion of variance i n preferences, R2 change = .04, p < .001. A l l of the variables entered into the regression equation accounted for a s i g n i f i c a n t amount of explained variance i n the dependent variable, R2 = .19, F (7, 229) = 8.70, p < .001. Four of the variables entered into the equation--work-family c o n f l i c t (Beta = .23, p < .001), work hours (Beta = -.21, p_ < .01), r e s p o n s i b i l i t y for household labour (Beta = -.16, p < .05), and time spent i n dependent care a c t i v i t i e s (Beta = .15, p < .05)--had s i g n i f i c a n t Beta c o e f f i c i e n t s , of which work-family c o n f l i c t was the most important i n influencing preferences (see Table 5). These results provide support for the p o s i t i v e relationship between work-family c o n f l i c t and preferences for family-oriented benefits. 6 1 Table 5 Summary of Regression of Preferences for Family-Oriented Benefits on Control Variables and Work-Family C o n f l i c t Variable 3 Hours Worked at Home . 0 6 Responsibility for - . 1 6 * Household Labour Hours Devoted to . 1 5 * Dependent Care Work Hours - . 2 1 * * Salary - . 0 8 Age - . 0 9 Work-Family C o n f l i c t . 2 3 * * * Note. Hours worked at home, r e s p o n s i b i l i t y f or household labour, hours devoted to dependent care, work hours, salary, and age were entered together i n the f i r s t step. Work-family c o n f l i c t was entered i n the next step. R 2 = . 1 5 for Step 1 ; A R 2 = . 0 4 for Step 2 (p_s < . 0 0 1 ) . *p_ < . 0 5 . **p_ < . 0 1 . ***p_ < . 0 0 1 . 62 H 4 : Female managers w i l l experience greater work-family c o n f l i c t than t h e i r male counterparts . An ANOVA and an ANCOVA were used to test the fourth hypothesis. The mean work-family c o n f l i c t score for men was 2.96 (SD = .88, n = 135) and for women, 2.93 (SD = .85, n = 125). Gender f a i l e d to have a s i g n i f i c a n t main e f f e c t on work-family c o n f l i c t , F (1, 258) = .07, p = .79; and i t explained only .03% of the variance i n work-family c o n f l i c t . When the three control variables found to be s i g n i f i c a n t l y correlated with work-family c o n f l i c t were added to the ANOVA testi n g the eff e c t s of gender on work-family c o n f l i c t , the mean work-family c o n f l i c t score for men became 2.95 (SD = .89, n = 126) and for women, 2.88 (SD = .82, n = 118). The mean,work-family c o n f l i c t score adjusted for the entry of covariates was 2.91 for men and 2.93 for women. The F (1, 239) ra t i o s for the covariates measuring work hours, hours spent working at home, and time spent i n dependent care a c t i v i t i e s were 33.43 (p < .001), 3.83 (p_ = .05), and 13.55 (p_ < .001) respectively. For gender, the F (1, 239) r a t i o was .03 (p = .86). Further analyses indicated that the covariates explained almost 2 0% of the variance i n work-family c o n f l i c t , while gender explained .01%. Thus, the hypothesis that female managers experience greater work-family c o n f l i c t than do male managers was not supported. H5: When the e f f e c t s of gender on work-family c o n f l i c t and of work-family c o n f l i c t on preferences for fami ly-or iented benef i ts are c o n t r o l l e d , the s i g n i f i c a n t r e l a t i o n s h i p between gender and preferences w i l l f a i l to be s i g n i f i c a n t . In other words, work-family c o n f l i c t w i l l mediate the r e l a t i o n s h i p between gender and preferences for fami ly -or ien ted b e n e f i t s . In order for work-family c o n f l i c t to be considered a 63 mediator between gender and preferences, gender must af f e c t work-family c o n f l i c t , gender must af f e c t preferences, and work-family c o n f l i c t must af f e c t preferences. A l l of these conditions must hold and do so i n the predicted directions (Baron & Kenny, 1986). Because gender did not have an ef f e c t on work-family c o n f l i c t and because i t did not have an eff e c t on preferences when covariates were included i n the analysis--as was demonstrated i n the test i n g of hypotheses 1 and 4--hypothesis 5 could not be tested. H6: Family involvement w i l l be p o s i t i v e l y associated with work-family c o n f l i c t . Bivariate and multiple regression analyses were u t i l i z e d to test the sixth hypothesis. Regressing work-family c o n f l i c t on family involvement revealed that family involvement did not contribute to the explained variance i n work-family c o n f l i c t , F (1, 255) = .03, R2 = -.004, p = .87. Thus, the hypothesis that increases i n family involvement correspond to increases i n work-family c o n f l i c t was not supported by the regression analysis. When the covariates measuring work hours, time spent working at home, and time spent i n dependent care a c t i v i t i e s were entered into a regression of work-family c o n f l i c t on family involvement, a s i g n i f i c a n t proportion of the variance i n work-family c o n f l i c t was explained, R2 = .15, F (4, 243) = 12.18, p < .001. Only work hours (Beta = .31, p < .001) and time spent i n dependent care a c t i v i t i e s (Beta = .21, p < .001) were s i g n i f i c a n t predictors of work-family c o n f l i c t i n thi s regression equation (see Table 6). Table 6 Summary of Regression of Work-Family C o n f l i c t on Control Variables and Family Involvement Variable B Hours Devoted to .21*** Dependent Care Hours Worked at Home .12 Work Hours .31*** Family Involvement -.05 Note. Hours devoted to dependent care, hours worked at home, and work hours were entered together i n the f i r s t step. Family involvement was entered i n the next step. R 2 = .15 for Step 1 (p < .001); A R 2 = .002 for Step 2. ***P < .001. 65 H 7 : When the e f f e c t s of family involvement on work-family c o n f l i c t and of work-family c o n f l i c t on preferences for family-oriented benefits are controlled, the s i g n i f i c a n t r e l a t i o n s h i p between family involvement and preferences w i l l f a i l to be s i g n i f i c a n t . In other words, work-family c o n f l i c t w i l l mediate the relat i o n s h i p between family involvement and preferences for family-oriented benefits. As was the case with hypothesis 5, work-family c o n f l i c t could only be considered a mediator between family involvement and preferences i f three conditions held: (a) i f family involvement affected work-family c o n f l i c t ; (b) i f family involvement affected preferences; and (c) i f work-family c o n f l i c t affected preferences. Even though family involvement and work-family c o n f l i c t did aff e c t preferences, hypothesis 7 could not be tested because family involvement d i d not have an e f f e c t on work-family c o n f l i c t - - a s was demonstrated i n the te s t i n g of hypothesis 6. Post Hoc Analyses Gender Difference i n Preferences Because the hypothesized relationship between gender and preferences for family-oriented benefits was not supported when control variables were included i n the analysis, additional analyses were performed i n order to provide a more det a i l e d picture of the relationship between gender and preferences. T-tests were employed to see whether male and female managers d i f f e r e d s i g n i f i c a n t l y i n terms of the hours they work per week, t h e i r s a l a r i e s , t h e i r ages, the time they devote to dependent care r e s p o n s i b i l i t i e s , and the amount of r e s p o n s i b i l i t y they have for household labour. Table 7 presents the results of the t-tests. S i g n i f i c a n t gender differences were found for a l l of the 66 W a) r H xi rd • H r l (0 > r H o r l -P C 0 u a) xi -p c o c QJ S o c rd c a) s 0 c • H r l (C Q e 0 u -P CO a) EH 1  EH <tH 0 > <D rc r H 43 (0 E H CO -PI c a) o 9 SI Cl c QJ 2 9 Cl o r l -p c o u rH XI • r l U > * * * * * * * * * * * * CO CO H in CO CO CO VO VO H • • • • • • 1 co in CO VO H ' o in CO CO H in r o H CM in in • • • • • • CM o CO H H CM o VO r o O l r o r - CO O in H CM • • • • • • 00 co CO H -a* o H t -VO VTi CM CO CM in CM CM CM CM CM CM r H H H H H r H H CO co in CO in CO CO CM in in CO * • • • • • cn o CO CO H CO VO • rd -P in H o VO o CM rd VO CO T J • • • • • • VO CM o VO CM CO IT in VO c VO • r H CTl IQ O CO H • r l s H in •<* CO o CO CO CM CO CO CO u H H H H H H -P a) T J -P >1 rd CO T J -P T j T J 0) 0) -P • H r H • H •P C r H O a) ' u O 0) lfl •H X! M • cd > T J XI cu fc > a) c •H W 0 a QJ 0 10 fc « c o u 0) 10 0) O « lfl XI >H Q 0) M cd o fc 0) g 3 >H U H ai (0 fc X) P B O O cd O CO IT Q) O cd O 0 S5 K -P U CO «< « <H i-q X X 67 control variables except time devoted to dependent care r e s p o n s i b i l i t i e s . Male managers worked more hours per week, earned more money, and were older than t h e i r female counterparts. Female managers, however, assumed greater r e s p o n s i b i l i t y for household labour than did male managers. A t-te s t also revealed there to be no s i g n i f i c a n t difference between men and women with respect to the number of hours spent working at home. Male and female managers were also compared with respect to the factors which influence t h e i r preferences for family-oriented benefits. Separate regression analyses were performed for men and women whereby preferences for family-oriented benefits were regressed on family involvement, work-family c o n f l i c t , work hours, salary, age, time spent i n dependent care a c t i v i t i e s , and r e s p o n s i b i l i t y for household labour. The r e l a t i v e impact each variable had on preferences was assessed. Results revealed that, f o r men, only two variables--work-family c o n f l i c t and family involvement--had s i g n i f i c a n t Beta c o e f f i c i e n t s (see Table 8). Of the two, work-family c o n f l i c t was the best predictor of preferences for family-oriented benefits. Thus, men who experienced higher lev e l s of work-family c o n f l i c t preferred family-oriented benefits more than men who experienced less c o n f l i c t . And, s i m i l a r l y , men who had higher lev e l s of family involvement preferred family-oriented benefits more than men who were less involved i n t h e i r family roles. For the male sample, the entire group of variables accounted for about 9% of the variance i n the dependent variable. Among the women, results indicated that three variables--work-family c o n f l i c t , age, and work hours--had s i g n i f i c a n t c o e f f i c i e n t s (see Table 9). As was Table 8 Summary of Regression Analysis for Variables Predicting Preferences for Family-Oriented Benefits Among Men (n=135) Variable (3 Work Hours -.13 Age .07 Responsibility for -.15 Household Labour Salary -.13 Hours Devoted to .01 Dependent Care Family Involvement .19* Work-Family C o n f l i c t .30** Note. A l l variables were entered together i n one step. R2 = .09 (p < .05) . *p_ < . 05 . **p < . 01. Table 9 Summary of Regression Analysis for Variables Predicting Preferences for Family-Oriented Benefits Among Women (n=l29) Variable (3 Work Hours -.24* Age -.25** Responsibility for -.07 Household Labour Salary -.04 Hours Devoted to .17 Dependent Care Family Involvement .12 Work-Family C o n f l i c t .26** Note. A l l variables were entered together i n one step. R2 = .24 (p < .001) . *p_ < . 05 . **p_ < . 01. 70 the case with the men, work-family c o n f l i c t was the best predictor of preferences. Thus, women who experienced more work-family c o n f l i c t , who were younger, and who worked fewer hours preferred family-oriented benefits more than did women with the converse c h a r a c t e r i s t i c s . For the female sample, a l l of the variables together accounted for over 24% of the variance i n the dependent variable. F i t with the Conceptual Model In order to explore whether the variables employed i n th i s study f i t the model empirically as they were believed to conceptually, a p r i n c i p a l components factor analysis was performed on the set of variables comprising the dependent variable, the independent variables, and the poten t i a l control variables. The variables did not group p r e c i s e l y as conceptualized by the model. They did not load on factors according to whether they were objective work, objective family, subjective work, or subjective family variables. The variables did, however, load according to whether they were subjective or objective variables. That i s , with a single exception ( i . e . , the variable measuring time devoted to dependent care), subjective variables and objective variables loaded on separate factors--objective and subjective components appeared to be d i s t i n c t (see Appendix E). Summary The results of t h i s study support the hypotheses that gender, family involvement, and work-family c o n f l i c t influence managers' preferences for family-oriented benefits. The hypotheses stating that family involvement and work-family 71 c o n f l i c t influence preferences for family-oriented benefits were supported even a f t e r the addition of control variables. Because gender and family involvement were not found to influence the degree of work-family c o n f l i c t experienced, the hypotheses that work-family c o n f l i c t mediates the relationships between gender and preferences and between family involvement and preferences were not supported (see Figure 3). The findings of the post hoc analyses revealed that there were s i g n i f i c a n t gender differences i n most of the variables used as s t a t i s t i c a l controls i n the test of the hypothesis r e l a t i n g gender and preferences for family-oriented benefits. The analyses also suggested that the factors which influence preferences for family-oriented benefits are somewhat d i f f e r e n t for men and women. Even though work-family c o n f l i c t was the most important predictor of preferences for both men and women, i t was the only common s i g n i f i c a n t predictor. 7And those variables related to preferences among the entire sample were better predictors of preferences among women than among men. 72 Work-Family C o n f l i c t Gender Preferences for Family-Oriented Benefits Work-Family C o n f l i c t Family Involvement _^ Preferences for Family-Oriented Benefits S u p p o r t e d P a r t i a l l y S u p p o r t e d Not S u p p o r t e d F i g u r e 3. R e s u l t s r e g a r d i n g the h y p o t h e s i z e d r e l a t i o n s h i p s among r e s e a r c h v a r i a b l e s . 73 Chapter IV Discussion The purpose of t h i s study was to examine the preferences of managers for job attributes designed to a s s i s t i n balancing the demands of work and family. S p e c i f i c a l l y , managers' preferences for family-oriented workplace benefits were examined to see i f they were influenced by gender or by the le v e l s of family involvement and work-family c o n f l i c t characterizing the worker. In addition, the relationships between gender and work-family c o n f l i c t and between family involvement and work-family c o n f l i c t were assessed to see i f work-family c o n f l i c t might mediate the relationships of gender and family involvement to preferences. A conceptual model r e l a t i n g work and family domains (Rice et a l . , 1979) was used as a framework for exploring and understanding the many influences on managers' preferences for family-oriented benefits. Preferences for Family-Oriented Benefits Gender and Preferences As expected, gender did influence preferences for family-oriented benefits; female managers preferred such benefits more than did male managers. This finding i s consistent with previous research which suggests that women, more than men, desire workplace benefits that enable them to devote more time to family while continuing to work outside of the home (e.g., Higgins, Duxbury, & Lee, 1992; Lee et a l . , 1992). Unlike previous research, however, t h i s gender difference i n preferences disappeared when p a r t i c u l a r work and family variables were controlled for. That i s , when the amount of time devoted to 74 dependent care r e s p o n s i b i l i t i e s , the extent of r e s p o n s i b i l i t y for household labour, the number of hours worked, age, and salary were taken into account, gender i t s e l f had l i t t l e e f f e c t on preferences. Among the sample as a whole, individuals who devoted more time to dependent care, who had the main r e s p o n s i b i l i t y for performing household tasks, who worked fewer hours, who were younger, and who earned smaller s a l a r i e s preferred family-oriented benefits more than did individuals with the converse c h a r a c t e r i s t i c s . The fact that male and female managers d i f f e r e d with respect to many of these c h a r a c t e r i s t i c s l i k e l y resulted i n the observed gender difference i n preferences for family-oriented benefits. Indeed, i t seems that the gender difference i n preferences for family-oriented benefits was rather a gender difference i n the factors related to preferences for family-oriented benefits. Female managers had more r e s p o n s i b i l i t y for household labour than did male managers. Assuming primary r e s p o n s i b i l i t y for household work while maintaining employment outside of the home contributes to the number of work and family demands which must be met and, therefore, may increase the desire for workplace p o l i c i e s which help to balance these demands. Female managers also worked fewer hours than male managers. Spending less time at work (and more time at home) may not necessarily reduce the s t r a i n associated with balancing work and family demands, but rather increase the array of pressures to which the worker i s exposed as work outside of the home i s performed i n addition to work i n the home (Greenhaus & Beutell, 1985). Consequently, the 75 necessity or importance of workplace benefits which help ease the pressures a r i s i n g from p a r t i c i p a t i o n i n d i f f e r e n t roles i s heightened. As well, individuals who work fewer hours may have less access to family-oriented benefits and thus may have a greater desire for them. Female managers were younger than male managers. Younger individuals l i k e l y have less experience i n t h e i r work and family roles and i n balancing these roles; therefore, they may desire, more than older individuals, workplace benefits which help reduce the stress caused by such inexperience. And f i n a l l y , female managers earned less money than t h e i r male counterparts. It i s possible that income i s a resource which serves the same purpose as family-oriented benefits--that i s , i t eases the s t r a i n associated with balancing work and family demands; hence, having a lower income would increase the importance placed on such benefits. Thus, i t appears that s p e c i f i c c h a r a c t e r i s t i c s associated with gender (e.g., r e s p o n s i b i l i t y for household labour) influence the strength of preferences for family-oriented benefits rather than gender i t s e l f . The finding that the gender difference i n preferences disappeared when s p e c i f i c work and family c h a r a c t e r i s t i c s were taken into account i s p a r t i c u l a r l y noteworthy i n that previous studies of preferences for family-oriented benefits have neglected to use a range of control variables i n t h e i r examinations of gender differences (e.g., Greenglass et a l . , 1988; Keith & Schafer, 1980; Wiersma, 1990). The results of t h i s study suggest that previous findings of gender differences i n preferences for family-oriented benefits be more cl o s e l y t 76 scrutinized. Family Involvement and Preferences The results of t h i s study concerning the influence of family involvement on preferences for family-oriented benefits were as expected. Even a f t e r c o n t r o l l i n g for a v a r i e t y of work and family c h a r a c t e r i s t i c s , family involvement had a s i g n i f i c a n t influence on preferences. Individuals characterized by a strong commitment to t h e i r family roles ascribed greater importance to family-oriented benefits than did those who were less psychologically involved i n t h e i r family roles. No previous empirical research has documented the influence of family involvement on preferences for family-oriented job a t t r i b u t e s . The finding of t h i s study does, however, p a r a l l e l that of a previous research study which found high l e v e l s of family involvement to be associated with the desire to restructure work i n order to accommodate family demands (Karambayya & R e i l l y , 1992) . Work-Family C o n f l i c t and Preferences Not surprisingly, work-family c o n f l i c t was found to influence managers' preferences for family-oriented benefits. In fact, i t was the most important predictor of preferences for family-oriented benefits among men and women. That higher l e v e l s of work-family c o n f l i c t were associated with stronger preferences for family-oriented benefits signals that family-oriented benefits were perceived by respondents as ways to r e l i e v e the s t r a i n associated with combining work and family roles. Thus, those individuals who were experiencing greater role s t r a i n desired, more strongly than others, means of reducing such 77 s t r a i n . This finding lends support to previous research (Wiersma, 1990) that has linked role c o n f l i c t to preferences for parental support job att r i b u t e s . The finding also reinforces the association between work-family c o n f l i c t and workplace-sponsored ways of managing such c o n f l i c t ( i . e . , family-oriented p o l i c i e s ) . Work-Family C o n f l i c t Gender and Work-Family C o n f l i c t Perhaps the most int e r e s t i n g r e s u l t of t h i s study was that men and women did not d i f f e r i n terms of the work-family c o n f l i c t they experience. This finding i s i n stark contrast to the findings of numerous studies which have consistently found that women experience greater work-family c o n f l i c t than men (e.g., Greenglass et a l . , 1988; Higgins et a l . , 1994; Wiersma, 1990). One possible explanation for t h i s rather unique finding may l i e i n the type of sample used. This study focussed on men and women i n managerial/professional l e v e l jobs only. However, much of the research which has found women to experience more work-family c o n f l i c t than men has u t i l i z e d samples of individuals employed i n a range of job lev e l s (e.g., Greenglass et a l . , 1988; Herman & Gyllstrom, 1977; Higgins et a l . , 1994; Wiersma, 1990; Wiersma & Van Den Berg, 1991). And many such studies have f a i l e d to control for the eff e c t of job l e v e l when examining gender differences i n work-family c o n f l i c t . Job l e v e l has been shown to be related to work-family c o n f l i c t (Higgins, Duxbury, & Lee, 1992); therefore, the gender differences i n work-family c o n f l i c t found i n previous studies may be due to the varying job le v e l s of the respondents rather than to t h e i r gender. Men and women at sim i l a r job leve l s , l i k e the men and women i n th i s study, may 78 experience comparable le v e l s of work-family c o n f l i c t . /Another possible explanation for the absence of a gender eff e c t on work-family c o n f l i c t may concern the s i m i l a r i t y between men and women with respect to p a r t i c u l a r variables related to work-family c o n f l i c t . Although the men i n thi s sample work more hours than the women--a c h a r a c t e r i s t i c which was found i n th i s study and which has been found i n previous studies to relate to greater perceived work-family c o n f l i c t (e.g., Campbell & Moen, 1992; Voydanoff, 1988; Wiersma & Van Den Berg, 1991)--no s i g n i f i c a n t differences between men and women were found with respect to the amount of time they devote to dependent care r e s p o n s i b i l i t i e s or the number of hours they spend working at home--two variables found i n thi s study to relate s i g n i f i c a n t l y and p o s i t i v e l y to work-family c o n f l i c t . Dependent care demands (MacBride-King, 1990) and the number of hours spent working at home (Silver, 1993) have been shown to be related to higher le v e l s of perceived work-family c o n f l i c t i n other studies as well. Thus, the fact that men and women i n th i s sample do not d i f f e r noticeably with respect to the amount of time they devote to dependent care or to working at home may explain why they do not demonstrate d i f f e r i n g l e v e l s of work-family c o n f l i c t . It may be that the potential for work-family c o n f l i c t i s r e l a t i v e l y equal for men and women and thi s explains why they experience sim i l a r l e v e l s of work-family c o n f l i c t . In other words, r e l a t i v e equality i n terms of objective predictors of work-family c o n f l i c t appears to indicate r e l a t i v e equality i n terms of the subjective perceptions of work-family c o n f l i c t . Yet another explanation can be offered for t h i s study's 79 f a i l u r e to f i n d a gender difference i n the amount of work-family-c o n f l i c t experienced by managers. Perhaps men and women view or define " c o n f l i c t " or "balance" d i f f e r e n t l y . That i s , men and women may be l i k e l y to experience and report d i f f e r e n t forms of work-family c o n f l i c t . Greenhaus and Beutell (1985) contend that there are three major forms of work-family c o n f l i c t : time-based c o n f l i c t ; strain-based c o n f l i c t ; and behaviour-based c o n f l i c t . Time-based c o n f l i c t occurs when time devoted to one role cannot be devoted to another role or when pressures within one role produce a preoccupation with the role which interferes with performance i n another role. Strain-based c o n f l i c t exists when s t r a i n (e.g., tension, anxiety, fatigue) i n one role makes i t d i f f i c u l t to meet the demands of another role. And behaviour-based c o n f l i c t results when " s p e c i f i c patterns of in - r o l e behavior may be incompatible with expectations regarding behavior i n another role" (Greenhaus & Beutell, 1985, p. 81). The majority (5 of 6) of items used to measure work-family c o n f l i c t i n t h i s study measured time-based work-family c o n f l i c t . Perhaps men and women i n thi s study do not d i f f e r s i g n i f i c a n t l y i n terms of the time-based c o n f l i c t they experience but rather d i f f e r with respect to the amount of s t r a i n - or behaviour-based c o n f l i c t they experience. Because the vast majority of studies of work-family c o n f l i c t have not distinguished between time-, s t r a i n - , and behaviour-based forms of c o n f l i c t (see Loerch et a l . , 1989, for an exception), i t i s possible that previously found gender differences i n work-family c o n f l i c t may r e f l e c t differences i n st r a i n - or behaviour-based c o n f l i c t instead of differences i n time-based c o n f l i c t . Then, the results of the present study 80 which revealed no gender difference i n work-family c o n f l i c t - -p rimarily time-based work-family conflict--may not be unusual. Future research should seek to v e r i f y the existence of the d i f f e r e n t forms of work-family c o n f l i c t i d e n t i f i e d by Greenhaus and Beutell (19 85) and investigate whether gender plays a role i n the type of work-family c o n f l i c t experienced. Family Involvement and Work-Family C o n f l i c t This study rejected the hypothesis that family involvement i s p o s i t i v e l y associated with work-family c o n f l i c t . While other empirical studies which have examined the saliency or c e n t r a l i t y of the family role to one's self-concept have found i t to influence work-family c o n f l i c t (Frone et a l . , 1992; Higgins, Duxbury, & Irving, 1992; Loerch et a l . , 1989), th i s study did not. Thus, contrary to the suggestions put forth by some l i t e r a t u r e (e.g., Greenhaus & Beutell, 1985; Hall & Richter, 1988), i t appears that individuals characterized by high l e v e l s of psychological involvement i n t h e i r family roles may not be more l i k e l y than others to experience the intrusion of family demands into the work domain. Greenhaus and Beutell (1985) posit that high l e v e l s of family involvement may lead one to become mentally preoccupied with the family role while performing a c t i v i t i e s related to the work role and that t h i s preoccupation, i n turn, leads to a high l e v e l of c o n f l i c t between family and work roles. S i m i l a r l y , Hall and Richter (1988) argue that high work boundary permeability--a c h a r a c t e r i s t i c of individuals who are highly involved i n t h e i r family r o l e s - - f a c i l i t a t e s the intrusion of family demands into the work domain. What may be the case, however, i s that male and female managers mentally 81 compartmentalize or segregate t h e i r family and work roles. That i s , i n an e f f o r t to a l l e v i a t e the c o n f l i c t caused by the intrusion of family demands into the work domain and vice versa, men and women may attempt to keep family and work roles separate, focussing on work while at work and family while at home. The use of such a role-management strategy as compartmentalization may explain why the l e v e l of family involvement characterizing the worker was not associated with the amount of work-family c o n f l i c t experienced. The finding that family involvement did not influence work-family c o n f l i c t may also suggest that a strong commitment to family roles alone i s not enough to foster higher lev e l s of work-family c o n f l i c t . Rather, the degree of commitment to other non-family roles (e.g., work, community, and r e l i g i o u s roles) may need to be considered together with the l e v e l of family involvement as a high l e v e l of involvement i n non-family roles combined with a high l e v e l of family involvement may be a better predictor of work-family c o n f l i c t than a high l e v e l of family involvement alone. And f i n a l l y , the finding that psychological involvement i n family roles was not related to work-family c o n f l i c t may be attributed to the type of work-family c o n f l i c t that was measured. Greenhaus and Beutell (19 85) contend that work-family c o n f l i c t i s b i d i r e c t i o n a l i n nature. That i s , work may in t e r f e r e with family (work-to-family-conflict) and/or family may in t e r f e r e with work (family-to-work-conflict). Greenhaus and Beutell (1985, p. 84) further note that "a d i r e c t i o n a l assumption of role interference (usually work i n t e r f e r i n g with family) often i s i m p l i c i t i n the 82 theory and the measurement of c o n f l i c t . " Moreover, based on empirical evidence, Frone et a l . (1992) state that mixed d i r e c t i o n a l measures of work-family c o n f l i c t (e.g., "I have a good balance between my work and my family time") primarily assess work-to-family-conflict. The measure of work-family c o n f l i c t employed i n t h i s study was l a r g e l y a measure of the degree to which work interfered with family ( i . e . , work-to-family-conflict) . Frone et a l . (1992) hypothesized and found that family involvement was d i r e c t l y and p o s i t i v e l y related to family-to-work-conflict as opposed to work-to-family-conflict. Thus, i t may be that family involvement i s associated with greater family-to-work-conflict rather than greater work-to-f a m i l y - c o n f l i c t . The Conceptual Model Unlike most studies which have examined preferences for family-oriented benefits, t h i s study employed a conceptual model to guide the testing of hypotheses. The model, propounded by Rice et a l . (1979) conceptualizes work and family domains as each having an objective or structural component and a subjective or behavioural component. The i n t e r r e l a t i o n s h i p s among a l l of these components were examined i n the present study. The results of t h i s study support the notion that work and family environments are intimately interfaced. A l l of the linkages of the conceptual model were strengthened. That i s , each relationship between components of the model was supported, a l b e i t inconsistently. As Rice et a l . (1979) depict i n t h e i r model, objective conditions of the family domain ( i . e . , gender, age, time devoted to dependent care demands, and r e s p o n s i b i l i t y 83 for household labour) and subjective conditions of the family-domain ( i . e . , family involvement and work-family c o n f l i c t ) were related to a subjective condition of the work domain--namely preferences for family-oriented benefits. Likewise, objective conditions of the work domain ( i . e . , work hours and salary) were related to a subjective condition of the work domain ( i . e . , preferences) as well as to an objective condition of the family domain ( i . e . , gender). In addition, one objective work variable ( i . e . , work hours) was related to a subjective family variable ( i . e . , work-family c o n f l i c t ) . However, only one objective condition of the family domain ( i . e . , time devoted to dependent care demands) was related to a subjective condition of the work domain ( i . e . , work-family c o n f l i c t ) ; the primary objective condition of the family domain studied--gender--was not. This research contributes to the development of the model i n that relationships among factors within groups of variables (e.g., between two d i f f e r e n t subjective family variables) were examined. No relationship was found between the two subjective conditions of the family domain ( i . e . , between family involvement and work-family c o n f l i c t ) , but relationships between objective conditions of the family domain ( i . e . , between gender and age and between gender and r e s p o n s i b i l i t y for household labour) were documented. When men and women were examined separately, more of the relationships between work and family variables depicted by the model were supported for women than for men. Nonetheless, the model appears to be an appropriate framework to use i n understanding the inte r r e l a t i o n s h i p s among work and family domains for both men and women. 84 The present findings also have implications for general model development i n the area of work and family. Analyses exploring the f i t of t h i s study's variables with the conceptual model revealed objective and subjective variables to be d i s t i n c t , but not work and family variables. Indeed, several objective and subjective variables examined i n t h i s study (e.g., hours spent working at home, work-family c o n f l i c t , and preferences for family-oriented workplace benefits) are not e a s i l y c l a s s i f i e d as either work or family variables. Given that numerous models employed i n s o c i a l science research (e.g., Brett & Yogev, 1988; Burke, 1988; Greenhaus & Beutell, 1985; Kelly & Voydanoff, 1985; Taylor & Pryor, 1986; Voydanoff & Kelly, 1984) have tended to c l a s s i f y variables as either work or non-work (family) variables, then the findings of t h i s study are intriguing.. Perhaps p a r t i c u l a r variables such as hours worked at home are neither s o l e l y work nor s o l e l y family variables, but instead are "work/family" variables. In th i s case, models such as that developed by Rice et a l . (1979) could plausibly have a work/family domain i n addition to a work domain and a family domain, with each domain having both a subjective and an objective component. Additional research which challenges the conceptualization of variables as either work or family variables would be i n s i g h t f u l . Limitations Several l i m i t a t i o n s of th i s study should be noted. F i r s t , because of the job l e v e l of the respondents chosen for t h i s study, i t i s not possible to generalize the results to individuals who do not occupy managerial/professional employment 85 positions. Men and women occupying higher positions i n t h e i r organizations are more l i k e l y than those i n lower-level positions to have access to family-oriented benefits and work arrangements and to have higher pay, more job autonomy, and greater job security (Lero et a l . , 1993). Hence, they may experience fewer burdens associated with managing both work and family r e s p o n s i b i l i t i e s and t h e i r preferences for benefits aimed at easing such burdens may be notably d i f f e r e n t . Future research should r e p l i c a t e t h i s study on samples of workers at job l e v e l s other than the managerial/professional l e v e l . Second, the g e n e r a l i z a b i l i t y of the study may also be hampered by s e l f - s e l e c t i o n biases, possibly r e f l e c t e d i n the low response rate. P a r t i c i p a t i o n i n t h i s study was voluntary and such s e l f - s e l e c t i o n may influence the r e s u l t s . For instance, only those individuals who f e l t they had time to respond to the survey--perhaps those who had lower l e v e l s of perceived work-family conflict--may have done so. S i m i l a r l y , devoting time to the completion of a lengthy survey may have been inconvenient and d i f f i c u l t for many subjects. Third, survey responses were self-reported and, therefore, may be subject to some bias (e.g., s o c i a l d e s i r a b i l i t y response bi a s ) . In addition, i t i s unclear whether managers can accurately report t h e i r preferences for p o l i c i e s and programs they may have had a hand i n implementing. It i s not known whether managers p a r t i c i p a t i n g i n t h i s study are or were involved i n the implementation of family-oriented benefits i n t h e i r places of work.• Fourth, as mentioned previously, the measure of work-family 86 c o n f l i c t used i n t h i s study did not measure a l l three forms of work-family c o n f l i c t i d e n t i f i e d by Greenhaus and Beutell (1985) . Rather, i t primarily measured time-based work-family c o n f l i c t . Nor did the measure of work-family c o n f l i c t used measure the interference of family with work ( i . e . , family-to-work-conflict); instead, i t primarily measured the interference of work with family ( i . e . , work-to-family-conflict). While the measure of work-family c o n f l i c t employed i n t h i s study was sound ( i . e . , was unidimensional, i n t e r n a l l y consistent, and appropriate for the sample), i t did not tap s t r a i n - and behaviour-based work-family c o n f l i c t or family-to-work-conflict; had i t done so, th i s study might have yielded d i f f e r e n t r e s u l t s . A challenge for researchers of work-family c o n f l i c t w i l l be to create measures of the three forms of c o n f l i c t that are applicable across a range of family situations and job l e v e l s . And, as Greenhaus and Beutell (1985, p. 84) note, " i t i s necessary to develop c o n f l i c t scales that contain a balance of items that r e f l e c t the d i f f e r e n t directions of role interference." F i n a l l y , t h i s study used data col l e c t e d as part of a larger study (Langton, 1995). While several questions included i n the survey instrument were constructed s p e c i f i c a l l y for the present study, space l i m i t a t i o n s precluded the use of more comprehensive measures. Conclusions and Implications A close examination of the l i t e r a t u r e pertaining to job preferences--namely preferences for s p e c i f i c job at t r i b u t e s - -c l e a r l y reveals the neglect on the part of researchers to study preferences for family-related work benefits. Numerous studies 87 of preferences have focussed on a v a r i e t y of job c h a r a c t e r i s t i c s such as salary, advancement opportunities, and security but have excluded from analysis aspects of work which are purposely implemented to ease the c o n f l i c t i n g demands between work and family. The present research study represents an attempt to broaden the base of knowledge concerning job a t t r i b u t e preferences and work-family c o n f l i c t . This study contributes to the job attribute preference l i t e r a t u r e by examining preferences for family-oriented p o l i c i e s and programs and the factors that influence such preferences. By focussing on the i n d i v i d u a l , family, and organizational c h a r a c t e r i s t i c s that influence preferences for family-oriented benefits, a more well-rounded picture of family-oriented workplace benefits and t h e i r implications i s gained. This study found that a number of factors influence managers' preferences for family-oriented benefits: the l e v e l of involvement i n family roles, salary, age, the number of hours worked, the amount of time devoted to dependent care r e s p o n s i b i l i t i e s , the extent of r e s p o n s i b i l i t y for household labour, the number of hours worked by one's spouse/partner, and the amount of time spent i n housework a c t i v i t i e s . These results suggest that d i f f e r e n t groups of workers (e.g., younger vs. older) have d i f f e r e n t preferences. Therefore, gathering information about p a r t i c u l a r work forces i s c r u c i a l to implementing family-oriented p o l i c i e s and programs which su i t employees' preferences and meet employees' needs. This research also contributes to the work-family c o n f l i c t l i t e r a t u r e by examining work-family c o n f l i c t as i t relates to 88 preferences for a range of family-oriented benefits--a topic which has not been studied previously. By better understanding the relationship between work-family c o n f l i c t and preferences for family-oriented benefits that exists for workers, employers and educators are aided i n t h e i r e f f o r t s to a l l e v i a t e such i n t e r r o l e c o n f l i c t . The findings of the present study demonstrate that workers who report higher l e v e l s of work-family c o n f l i c t express a strong desire for family-oriented benefits aimed at helping balance work and family demands. Thus, o f f e r i n g benefits such as f l e x i b l e work schedules, family leaves, and supervisors who are f l e x i b l e when family needs aris e may be one way to reduce work-family c o n f l i c t . This study expands on e x i s t i n g studies of preferences for family-oriented benefits by examining a range of benefits--both formal and informal. Researchers who have studied preferences for family-oriented benefits have tended to focus on formal supports at the workplace such as f l e x i b l e work arrangements and leave-related p o l i c i e s and to neglect informal support provided by supervisors and the organizational culture (e.g., Goldberg et a l . , 1989; Higgins, Duxbury, & Lee, 1992). Other studies of preferences for family-oriented benefits have l i m i t e d t h e i r analyses to a p a r t i c u l a r type of family-oriented benefit such as c h i l d care assistance (e.g., Kossek, 1990; Wiersma, 1990). This study's focus on both formal and informal supports could further pave the way for future studies seeking to expand upon the range of family-oriented benefits examined i n t h i s study. Not only should researchers examine a wider range of benefits, but also develop and employ multiple item measures of preferences for 89 p a r t i c u l a r benefits. Such work could f a c i l i t a t e an investigation into whether d i f f e r e n t dimensions of preferences for family-oriented benefits e x i s t . Burke (1988, p. 301) asserts that "research i n the work-family area must include comprehensive models or frameworks so that a wide var i e t y of relationships (and concepts) can be considered simultaneously". This study represents a step toward t h i s goal. It i s unique i n that i t examines employee preferences for family-oriented benefits within a pre-existing conceptual framework which recognizes that work and family are intimately interconnected. Continued e f f o r t s should be directed towards using and developing conceptual models of work/non-work linkages. As Raabe (1990, p. 481) states, "improved conceptual... understanding of the i n t e r r e l a t i o n s of work and family domains and the complexities of p o l i c y variables and processes ... i s ... c r u c i a l to a t t a i n v a l i d conclusions". This study recognizes that the workforce i s diverse by examining preferences for family-oriented benefits as they are influenced by a host of factors. Future research examining preferences for workplace family supports should examine employee preferences as they are influenced by variables other than those examined i n thi s study. For example, the relationships between access to benefits, actual use of benefits, and preferences for family-oriented benefits should be explored. Wiersma (1990) suggests that future research should focus on changes that need to be made at work to lessen role c o n f l i c t . An important--and often overlooked--way of determining what workplace changes need to be made i s to ask employees themselves. 90 Kossek (1990) notes that employers should conduct systematic needs assessments p r i o r to implementing family-supportive p o l i c i e s and programs i n order to determine the need for and amount of variance i n preferences for assistance i n the workforce. Her comments regarding employer-sponsored c h i l d care assistance are equally applicable to a l l family-related benefits: Child care requirements may not only vary for employee groups within a firm, but also vary s u b s t a n t i a l l y between firms. ...Unfortunately, many companies have jumped on the " c h i l d care corporate bandwagon" without delineation of how such programs f i t with t h e i r human resource strategy or work force p r o f i l e . ... Few...organizations [conduct] formal c h i l d care needs assessments or [survey] t h e i r workforces before adoption. Assuming there are scarce organizational resources available to attack the c h i l d care problem, adding programs based on current labor market data w i l l better ensure that the p o l i c i e s added w i l l be needed and used. (p. 784) The fact that men and women i n the present study ascribed the greatest importance to the same family-oriented benefit--a supervisor who i s f l e x i b l e when family needs arise--has important implications for employers and employees. A f l e x i b l e supervisor i s indeed a benefit that most, i f not a l l , employers can afford to o f f e r t h e i r employees. The widespread d e s i r a b i l i t y of a f l e x i b l e supervisor signals that a large sector of the working population could be s a t i s f i e d and motivated by the implementation of t h i s p a r t i c u l a r family-oriented benefit. The findings of t h i s study have implications for family l i f e educators working with employers, employees, and prospective employees. Work-family c o n f l i c t s and the preferred means of managing such c o n f l i c t s are important topics of family l i f e education. Family l i f e educators could a s s i s t employers i n conducting surveys of t h e i r employees' preferences for family-91 oriented benefits. Such a consideration by employers of the d i v e r s i t y of employee preferences and the factors influencing them may enable employers to o f f e r benefits which would be perceived as equitable and e f f e c t i v e by a l l workers. "Knowing what employees desire from t h e i r work helps organizations design jobs, develop reward systems, promote leadership styles, and design general personnel p o l i c i e s to improve job s a t i s f a c t i o n and motivation" (Wiersma, 1990, p. 231). And i n helping individuals to i d e n t i f y and examine t h e i r p r i o r i t i e s , family l i f e educators may be better able to help them set more attainable goals. "Adolescents and adults", claim Raabe and Gessner (1988, p. 201), "can benefit from more knowledge about today's s i g n i f i c a n t linkages between employment and family domains, the challenge of work-family coordination, possible problem-solving s k i l l s , and organizational supports". As an increasing number of workers indicate a desire to devote more time and attention to t h e i r personal and family l i v e s (Galinsky et a l . , 1993), there i s a growing need for workplaces to address t h i s preference. By asking workers about how best t h e i r needs may be met and acting accordingly, employers w i l l be better able to a l i g n workers' needs with workplace goals and reduce the negative consequences for workers--both male and female--of combining work and family r e s p o n s i b i l i t i e s . 92 References Andrisani, P.J., & Miljus, R.C. (1971). Individual differences i n preferences for i n t r i n s i c versus e x t r i n s i c aspects of work. Journal of Vocational Behavior, 11, 14-30. Baron, R.M., & Kenny, D.A. (1986). The moderator-mediator variable d i s t i n c t i o n i n s o c i a l psychological research: Conceptual, strategic, and s t a t i s t i c a l considerations. Journal of Personality and Social Psychology, 51, 1173-1182. Bartol, K.M. (1976a). Expectancy theory as a predictor of female occupational choice and attitude toward business. Academy of Management Journal, 19. 669-675. Bartol, K.M. (1976b). Relationship of sex and professional t r a i n i n g area to job orientation. Journal of Applied Psychology, 61, 368-370. Bedeian, A.G., Burke, B.G., & Moffett, R.G. (1988). Outcomes of work-family c o n f l i c t among married male and female professionals. Journal of Management. 14, 475-491. Berger, P.S. (1986). Differences i n importance of and s a t i s f a c t i o n from job c h a r a c t e r i s t i c s by sex and occupational type among Mexican-American employees. Journal of Vocational Behavior. 28. 203-213. Beutell, N.J., & Brenner, O.C. (1986). Sex differences i n work values. Journal of Vocational Behavior, 28, 29-41. Beutell, N.J., & O'Hare, M.M. (1987). Work-nonwork c o n f l i c t among MBAs: Sex differences i n role stressors and l i f e s a t i s f a c t i o n . Work and Stress. 1, 35-41. Bigoness, W.J. (1988). Sex differences i n job att r i b u t e preferences. Journal of Organizational Behavior, 9, 139-147. Block, J., Denker, E.R., & T i t t l e , C.K. (1981). Perceived influences on career choices of eleventh graders: Sex, SES, and ethnic group comparisons. Sex Roles, 7, 895-904. Bohen, H.H., & Viveros-Long, A. (1981). Balancing jobs and family l i f e : Do f l e x i b l e work schedules help? Philadelphia: Temple University Press. Bowen, G.L. (1988). Corporate supports for the family l i v e s of employees: A conceptual model for program planning and evaluation. Family Relations. 37, 183-188. Brett, J.M., & Yogev, S. (1988). Restructuring work for family: How dual-earner couples with children manage. In E.B. Goldsmith (Ed.), Work and family: Theory, research, and applications (pp. 159-174). Newbury Park, CA: Sage. 93 Bridges, J.S. (1989) . Sex differences i n occupational values. Sex Roles, 20, 205-211. Brief, A.P., & Aldag, R.J. (1975). Male-female differences i n occupational attitudes within minority groups. Journal of Vocational Behavior, 6, 305-314. Brief, A.P., & Oliver, R.L. (1976). Male-female differences i n work attitudes among r e t a i l sales managers. Journal of Applied Psychology, 61, 526-528. Brief, A.P., Rose, G.L., & Aldag, R.J. (1977). Sex differences i n preferences for job attributes r e v i s i t e d . Journal of Applied Psychology, 62, 645-646. Burke, R.J. (1988). Some antecedents and consequences of work-family c o n f l i c t . In E.B. Goldsmith (Ed.), Work and family: Theory, research, and applications (pp. 287-302). Newbury Park, CA: Sage. Campbell, M.L., & Moen, P. (1992). Job-family role s t r a i n among employed single mothers of preschoolers. Family Relations. 41. 205-211. Canadian Aging Research Network. (1993, March). Work and family. The survey. Guelph, ONT: Author. Cherry, N. (1975). Occupational values and employment: A follow-up study of graduate men and women. Higher Education, 4, 357-368. Covin, T.J., & Brush, C C . (1993) . A comparison of student and human resource professional attitudes toward work and family issues. Group and Organization Management, 18, 29-49. Daymont, T.N., & Andrisani, P.J. (1984). Job preferences, college major, and the gender gap i n earnings. Journal of Human Resources, 19, 408-428. Duxbury, L., Higgins, C , & Lee, C. (1994). Work-family c o n f l i c t . A comparison by gender, family type, and perceived control. Journal of Family Issues. 15, 449-466. Duxbury, L., Higgins, C , Lee, C , & M i l l s , S. (1992). An examination of organizational and individual outcomes. Optimum: The Journal of Public Sector Management. 23, 46-59. Duxbury, L.E., & Higgins, CA. (1991). Gender differences i n work-family c o n f l i c t . Journal of Applied Psychology, 76, 60-74. Ferber, M.A., O'Farrell, B., & Allen, L. (1991). Work and family: P o l i c i e s for a changing work force. Washington, DC: National Academy Press. 94 Friedman, D.E., & Galinsky, E. (1992). Work and family issues: A legitimate business concern. In S. Zedeck (Ed.), Work, families, and organizations (pp. 168-207). San Francisco: Jossey-Bass. Frone, M.R., & Rice, R.W. (1987). Work-family c o n f l i c t : The e f f e c t of job and family involvement. Journal of Occupational Behaviour. 8. 45-53. Frone, M.R., Russell, M., & Cooper, M.L. (1992). Antecedents and outcomes of work-family c o n f l i c t : Testing a model of the work-family interface. Journal of Applied Psychology, 77, 65-78. Galinsky, E., Bond, J.T., & Friedman, D.E. (1993). The changing workforce: Highlights of the national study. New York: Families at Work Institu t e . Galinsky, E., & Stein, P.J. (1990). The impact of human resource p o l i c i e s on employees. Balancing work/family l i f e . Journal of Family Issues. 11, 368-383. Goff, S.J., Mount, M.K., & Jamison, R.L. (1990). Employer supported c h i l d care, work/family c o n f l i c t , and absenteeism: A f i e l d study. Personnel Psychology, 43, 793-809. Goldberg, W.A., Greenberger, E., Koch-Jones, J., O'Neil, R., & Hamill, S. (1989). Attractiveness of c h i l d care and related employer-supported benefits and p o l i c i e s to married and single parents. Child and Youth Care Quarterly, 18, 23-37. Gorn, G.J., & Kanungo, R.N. (1980). Job involvement and motivation: Are i n t r i n s i c a l l y motivated managers more job involved? Organizational Behavior and Human Performance. 26. 265-277 . Greenglass, E.R., Pantony, K., & Burke, R.J. (1988). A gender-role perspective on role c o n f l i c t , work stress and s o c i a l support. In E.B. Goldsmith (Ed.), Work and family: Theory, research, and applications (pp. 317-328). Newbury Park, CA: Sage. Greenhaus, J.H. (19 88). The inters e c t i o n of work and family roles: Individual, interpersonal, and organizational issues. In E.B. Goldsmith (Ed.), Work and family: Theory, research, and applications (pp. 23-44). Newbury Park, CA: Sage. Greenhaus, J.H., & Beutell, N.J. (1985). Sources of c o n f l i c t between work and family roles. Academy of Management Review, 10, 76-88. Greenhaus, J.H., & Kopelman, R.E. (1981). C o n f l i c t between work and nonwork roles: Implications for the career planning process. Human Resource Planning, 4, 1-10. 95 Greenhaus, J.H., Parasuraman, S., Granrose, C.S., Rabinowitz, S., & Beutell, N.J. (1989). Sources of work-family c o n f l i c t among two-career couples. Journal of Vocational Behavior, 34, 133-153. Gutek, B.A., Searle, S., & Klepa, L. (1991). Rational versus gender role explanations for work-family c o n f l i c t . Journal of Applied Psychology, 76. 560-568. Hal l , D., & Richter, J. (1988). Balancing work l i f e and home l i f e : What can organizations do to help? Academy of Management Executive. 11, 213-223. Hatcher, M.A. (1991). The corporate woman of the 1990s: Maverick or innovator? Psychology of Women Quarterly. 15. 251-260 . Herman, J., & Gyllstrom, K. (1977). Working men and women: Inter- and i n t r a - r o l e c o n f l i c t . Psychology of Women Quarterly. 1, 319-333. Higgins, C , Duxbury, L., & Lee, C. (1992). Balancing work and family: A study of Canadian private sector employees. Ottawa, ONT: The University of Western Ontario, National Centre for Management Research and Development. Higgins, C , Duxbury, L., & Lee, C. (1994). Impact of l i f e -cycle stage 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. Higgins, C , Duxbury, L. , Lee, C , & M i l l s , S. (1992). An examination of work-time and work-location f l e x i b i l i t y . Optimum: The Journal of Public Sector Management, 23, 29-37. Higgins, C.A., Duxbury, L.E., & Irving, R.H. (1992). Work-family c o n f l i c t i n the dual-career family. Organizational Behavior and Human Decision Processes. 51. 51-75. Holahan, C.K., & Gil b e r t , L.A. (1979). C o n f l i c t between major l i f e r oles: Women and men i n dual-career couples. Human Relations. 32, 451-467. Holmstrom, V.L., & Beach, L.R. (1973). Subjective expected u t i l i t y and career preferences. Organizational Behavior and Human Performance, 10, 201-207. Hughes, D., & Galinsky, E. (1988). Balancing work and family l i v e s . Research and corporate applications. In A.E. G o t t f r i e d & A.W. Got t f r i e d (Eds.), Maternal employment and children's development: Longitudinal research (pp. 233-268). New York: Plenum Press. Jones, A.P., & Butler, M.C. (1980). A role t r a n s i t i o n approach to the stress of organizationally induced family role disruption. Journal of Marriage and the Family. 42, 367-376. 96 Jurgensen, C E . (1978). Job preferences (What makes a job good or bad?). Journal of Applied Psychology. 63, 267-276. Jurgensen, C E . (1947). Selected factors which influence job preferences. Journal of Applied Psychology, 31, 553-564. Kalleberg, A.L., & Loscocco, K.A. (1983). Aging, values, and rewards: Explaining age differences i n job s a t i s f a c t i o n . American Socio l o g i c a l Review, 48, 78-90. Kamerman, S.B., Kahn, A.J., & Kingston, P.W. (1983). Maternal p o l i c i e s and working women. New York: Columbia University Press. Karambayya, R., & R e i l l y , A.H. (1992). Dual earner couples: Attitudes and actions i n restructuring work for family. Journal of Organizational Behavior, 13, 585-601. Katz, M.H., & Piotrkowski, C S . (1983). Correlates of family role s t r a i n among employed black women. Family Relations, 32, 331-339. Keith, P.M., & Schafer, R.B. (1980). Role s t r a i n and depression i n two-job families. Family Relations, 29. 483-488. Kelly, R.F., & Voydanoff, P. (1985). Work/family role s t r a i n among employed parents. Family Relations. 34, 367-374. Kerlinger, F.N. (1986). Foundations of Behavioral Research (3rd ed.). Toronto, ONT: Holt, Rinehart and Winston. Konrad, A.M., & Langton, N. (1993). What do women want? The impact of preferences on earnings. Unpublished paper. Kopelman, R.E., Greenhaus, J.H., & Connolly, T.F. (1983). A model of work, family, and i n t e r r o l e c o n f l i c t : A construct v a l i d a t i o n study. Organizational Behavior and Human Performance. 32. 198-215. Kossek, E.E. (1990). D i v e r s i t y i n c h i l d care assistance needs: Employee problems, preferences, and work-related outcomes. Personnel Psychology. 43. 769-791. Lacy, W.B., Bokemeier, J.L., & Shepard, J.M. (1983). Job attribute preferences and work commitment of men and women i n the United States. Personnel Psychology, 36. 315-329. Langton, N. (1995) . Career development survey of managers and professionals. Unpublished data. Lee, C , Duxbury, L., Higgins, C , & M i l l s , S. (1992). Strategies used by employed parents to balance the demands of work and family. Optimum: The Journal of Public Sector Management. 23, 60-69. 97 Lero, D.S., Brockman, L.M., Pence, A.R., Goelman, H., & Johnson, K.L. (1993). Workplace benefits and f l e x i b i l i t y : A perspective on parents' experiences (Catalogue No. 89-530E). Ottawa, ONT: S t a t i s t i c s Canada. Lero, D.S., & Johnson, K.L. (1994). 110 Canadian s t a t i s t i c s on work and family. Ottawa, ONT: The Canadian Advisory Council on the Status of Women. Loerch, K.J., Russell, J.E.A., & Rush, M.C. (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 Behavior. 35, 288-308. Louis Harris & Associates, Inc. (1981). Families at work: Strengths and str a i n s . Minneapolis, MN: General M i l l s . Lyson, T.A. (1984). Sex differences i n the choice of a male or female career l i n e : An analysis of background c h a r a c t e r i s t i c s and work values. Work and Occupations, 11. 131-146. MacBride-King, J.L. (1990). Work and family: Employment challenge of the 90's (Report 59-90). Ottawa, ONT: Conference Board of Canada. Manhardt, P.J. (1972). Job orientation of male and female college graduates i n business. Personnel Psychology, 25. 361-368. Martin, J.K., & Hanson, S.L. (1985). Sex, family wage-earning status, and s a t i s f a c t i o n with work. Work and Occupations, 12, 91-109. Meisenheimer, J.R., & Wiatrowski, W.J. (1989). F l e x i b l e benefit plans: Employees who have a choice. Monthly Labor Review, 112. 17-23. Mize, J., & Freeman, L.C. (1989). Employer-supported c h i l d care: Assessing the need and potential support. Child and Youth Care Quarterly, 18. 289-301. Moen, P., & Dempster-McClain, D.I. (1987). Employed parents: Role s t r a i n , work time, and preferences for working l e s s . Journal of Marriage and the Family. 49. 579-590. Near, J.P., Rice, R.W., & Hunt, R.G. (1980). The relationship between work and nonwork domains: A review of empirical research. Academy of Management Review. 5, 415-429. Padavic, I. (1992). White-collar work values and women's interest i n b l u e - c o l l a r jobs. Gender and Society. 6. 215-230. 98 Parasuraman, S., & Greenhaus, J.H. (1993). Personal p o r t r a i t : The l i f e - s t y l e of the woman manager. In E.A. Fagenson (Ed.), Women and work: A research and p o l i c y series Vol. 4. Women in management: Trends, issues, and challenges i n managerial d i v e r s i t y (pp. 186-211). Newbury Park, CA: Sage. Paris, H. (1989). The corporate response to workers with family r e s p o n s i b i l i t i e s (Report 43-89) . Ottawa, ONT: The Conference Board of Canada. Pleck, J. (1979). Work-family c o n f l i c t : A national assessment. Paper presented at the annual meeting of the Society for the Study of Social Problems, Boston, MA. Posner, B.Z. (1981). Comparing r e c r u i t e r , student, and faculty perceptions of important applicant and job ch a r a c t e r i s t i c s . Personnel Psychology. 34, 329-339. Posner, B.Z., & Powell, G.N. (1990). A longitudinal investigation of work preferences among college graduates. Psychological Reports. 66, 1125-1134. Pryor, R. (1982). Values, preferences, needs, work ethics, and orientations to work: Toward a conceptual and empirical integration. Journal of Vocational Behavior, 20, 40-52. Pryor, R.G.L. (1979). In search of a concept: Work values. Vocational Guidance Quarterly, 27, 250-258. Pryor, R.G.L. (1981). Interests and values as preferences: A va l i d a t i o n of the Work Aspect Preference Scale. Australian Psychologist. 16, 258-272. Quinn, R.P., & Shepard, L.J. (1974). The 1972-73 Quality of Employment survey: Descriptive s t a t i s t i c s with comparison data from the 1969-70 Survey of Working Conditions. Ann Arbor, MI: University of Michigan, Institute for Social Research Survey Research Center. Quinn, R.P., & Staines, G.L. (1979). The 1977 Quality of Employment Survey: Descriptive s t a t i s t i c s with comparison data from the 1969-70 and the 1972-1973 surveys. Ann Arbor, MI: University of Michigan Press. Raabe, P.H. (1990). The organizational e f f e c t s of workplace family p o l i c i e s . Past weaknesses and recent progress toward improved research. Journal of Family Issues, 11, 477-491. Raabe, P.H., & Gessner, J.C. (1988). Employer family-supportive p o l i c i e s : Diverse variations on the theme. Family Relations. 37. 196-202. Rice, R.W., Near, J.P., & Hunt, R.G. (1979). Unique variance i n job and l i f e s a t i s f a c t i o n associated with work-related and extra-workplace variables. Human Relations, 32, 605-623. 99 Rowe, P.M. (1976). Effects of expected job c h a r a c t e r i s t i c s and interview factors on organizational choice. Psychological Reports. 38, 1011-1018. Saleh, S.D., & L a l l j e e , M. (1969). Sex and job orientation. Personnel Psychology, 22, 465-471. Sarbin, T.R., & Allen, V.L. (1968). Role theory. In G. Linzey & E. Aronson (Eds.), Handbook of s o c i a l psychology: Volume 1 (2nd ed., pp. 488-567). Reading, MA: Addison-Wesley. Schuler, R.S. (1975). Sex, organizational l e v e l , and outcome importance: Where the differences are. Personnel Psychology. 28. 365-375. Scozzaro, P.P., & Subich, L.M. (1990). Gender and occupational sex-type differences i n job outcome' factor perceptions. Journal of Vocational Behavior. 36, 109-119. Sheard, J.L. (1970). Intrasubject prediction of preferences for organization types. Journal of Applied Psychology, 54, 248-252 . S i l v e r , H. (1993). Homework and domestic work. Soc i o l o g i c a l Forum, 8, 181-204. Singer, J.N. (1974). Sex d i f f e r e n c e s - s i m i l a r i t i e s i n job preference factors. Journal of Vocational Behavior, 5, 357-365. Staines, G.L., & Pleck, J.H. (1984). Nonstandard work schedules and family l i f e . Journal of Applied Psychology, 69, 515-523. Taylor, N.B., & Pryor, R.G.L. (1986). The conceptualisation and measurement of vocational and work aspect preferences. B r i t i s h Journal of Guidance and Counselling, 14, 66-77. Terpstra, D.E. (1983). An investigation of job-seeker preferences through multiple methodologies. Journal of Employment Counseling, 20, 169-178. Thomas, L.T., & Ganster, D.C. (1995). Impact of family-supportive work variables on work-family c o n f l i c t and s t r a i n : A control perspective. Journal of Applied Psychology, 80, 6-15. Voydanoff, P. (1988). Work role 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. Voydanoff, P., & Kelly, R.F. (1984). Determinants of work-related family problems among employed parents. Journal of Marriage and the Family, 46, 881-892. 100 Walker, J.E., Tausky,' C , & Oliver, D. (1982). Men and women at work: S i m i l a r i t i e s and differences i n work values within occupational groupings. Journal of Vocational Behavior. 21. 17-36. Warren, J.A., & Johnson, P.J. (1995). The impact of workplace support on work-family role s t r a i n . Family Relations. 44. 163-169. Weaver, C.N. (1975). Job preferences of white c o l l a r and blue c o l l a r workers. Academy of Management Journal, 18, 167-175. Weaver, C.N. (19 76). What workers want from t h e i r jobs. Personnel, 53. 48-54. Weaver, C.N. (1992). Black-white differences i n job preferences 16 years l a t e r . Psychological Reports. 71, 1087-1090. Wiersma, U.J. (1990). Gender differences i n job att r i b u t e preferences: Work-home role c o n f l i c t and job l e v e l as mediating variables. Journal of Occupational Psychology, 63. 231-243. Wiersma, U.J., & Van Den Berg, P. (1991). Work-home role c o n f l i c t , family climate, and domestic r e s p o n s i b i l i t i e s among men and women i n dual-earner families. Journal of Applied Social Psychology. 21. 1207-1217. Yogev, S., & Brett, J. (1985). Patterns of work and family involvement among single- and dual-earner couples. Journal of Applied Psychology. 70. 754-768. 101 Appendix A Questionnaire Items Dependent Variable Preferences for Family-Oriented Benefits 1. Each of the following job conditions represents one way organizations can help employees balance work and family demands. Please indicate how important the a v a i l a b i l i t y of each of these i s to you. Ve ry S ome wha t S omewha t Very Important Important Unimportant Unimportant a. F l e x i b i l i t y i n scheduling your work hours b. Part-time employment with pro-rated benefits c. Work location f l e x i b i l i t y (e.g., work at home, telecommuting) d. On- or near-site daycare f a c i l i t i e s e. F l e x i b l e benefit programs (e.g., the opportunity to select a package of benefits) f. Leaves for personal/family reasons g. Workplace counselling and information services designed to a s s i s t i n managing work and family l i f e h. Relocation assistance programs i . An organizational culture supportive and understanding of family r e s p o n s i b i l i t i e s j . A supervisor who i s f l e x i b l e when family needs arise [The items comprising the preference scale were recoded and summed to y i e l d a t o t a l score, with a high score i n d i c a t i n g a strong preference for family-oriented benefits.] Independent Variables Gender l . What i s your sex? a. Male b. Female Family Involvement 1. Please indicate your l e v e l of agreement with the following statements: Strongly Agree Agree Disagree Strongly Disagree 102 a. The most important things that happen to me involve my family b. Even when I'm busy doing other things, I often think about my family c. My main s a t i s f a c t i o n i n l i f e comes from my family [Individual item scores were summed to y i e l d a t o t a l family involvement score. Items were recoded so that high scores on t h i s scale are i n d i c a t i v e of high family involvement.] Work-Family C o n f l i c t 1. Please indicate the extent to which each of the following statements r e f l e c t s your current job s i t u a t i o n : Strongly Agree Neither Agree Disagree Strongly Agree nor Disagree Disagree a. My work schedule rarely c o n f l i c t s with my family l i f e b. My family d i s l i k e s how often I am preoccupied with my work while I am at home c. My work takes up time that I'd l i k e to spend with my family d. My job makes i t d i f f i c u l t to be the kind of spouse/partner or parent I'd l i k e to be e. I have a good balance between my job and my family time f. My job keeps me away from my family too much [Scale items were summed to y i e l d an o v e r a l l work-family c o n f l i c t score. Negatively worded items (b, c, d, f) were recoded i n order that a high score indicate a high l e v e l of work-family c o n f l i c t . ] Control Variables Work Involvement l . Please indicate your l e v e l of agreement with the following statements: Strongly Agree Agree Disagree Strongly Disagree a. The most important things that happen to me involve my job b. Even when I'm busy doing other things I often think about my job c. My main s a t i s f a c t i o n i n l i f e comes from my work 103 [Scores for the items were recoded and summed to y i e l d a t o t a l work involvement score, with a high score i n d i c a t i n g high work involvement.] Work Hours 1. With respect to your paid employment r e s p o n s i b i l i t i e s , how many hours per week, on average: do you work? Hours Worked at Home 1. With respect to your paid employment r e s p o n s i b i l i t i e s , how many hours per week, on average: do you conduct your paid work at home? Employment Sector With respect to the current employer: 1. Starting with your current p o s i t i o n and working backwards to your f i r s t f u l l - t i m e p o s i t i o n (greater than 30 hours per week), please supply the following information about each employer for whom you have worked f u l l - t i m e . Was the firm public, p r i v a t e - f o r - p r o f i t , or private-n o t - f o r - p r o f i t ? Job Tenure With respect to the current employer: 1. Starting with your current p o s i t i o n and working backwards to your f i r s t f u l l - t i m e p o s i t i o n (greater than 30 hours per week), please supply the following information about each employer for whom you have worked f u l l - t i m e . Dates of ful l - t i m e employment (mo/yr to mo/yr) [The t o t a l number of years the respondent has been employed i n his/her current p o s i t i o n was calculated.] Salary 1. (a) What i s your current annual salary (including bonuses) before taxes? Age 1. What i s your date of birth? Month Year 104 Marital Status 1. Which of the following best describes your current marital status? a. Single, not involved i n a relationship b. Involved i n a relationship but not married c. Living with a partner but not married d. Married Presence of an Employed Spouse 1. (a) What i s your spouse's (or partner's) occupation? Please include job t i t l e / p o s i t i o n and industry: Job T i t l e / P o s i t i o n Industry Spouse's Work Hours 1. (c) On average, how many hours per week does your spouse/partner work i n a paid position? Number of Children 1. Please l i s t the ages of your children, the percentage of time they l i v e with you, and the type of c h i l d care they receive. Age Percent of time Type of c h i l d l i v i n g with you care arrangement Child #1 Child #2 Child #3 Child #4 Child #5 : [Only the Age portion of t h i s question was used for t h i s study.] Hours of Housework 1. Please indicate the average number of hours i n a t y p i c a l week (including weekends) that you spend on the following types of a c t i v i t i e s : Housework - including food preparation, dishwashing, regular or periodic housecleaning, clothing care a c t i v i t i e s , etc., or arranging for any of these tasks to be done by others. Hours Devoted to Dependent Care R e s p o n s i b i l i t i e s 1. Please indicate the average number of hours i n a t y p i c a l week (including weekends) that you spend on the following types of a c t i v i t i e s : 105 Child care - including feeding, dressing, bathing, helping with homework, d i s c i p l i n i n g , talking, reading, driving, playing, etc., or arranging for any of these tasks to be done by others. Elder care - including transporting, scheduling a c t i v i t i e s or performing tasks for elder family members or arranging for any of these tasks to be done by others. [The t o t a l number of hours per week devoted to dependent care r e s p o n s i b i l i t i e s was calculated by summing the number of weekly hours spent i n c h i l d and elder care a c t i v i t i e s . ] Responsibility for Household Labour 1. In your family, who has the main r e s p o n s i b i l i t y for seeing that the following tasks get done? (a) Housework, including food preparation, housecleaning, clothing care, etc. or arranging for these types of tasks be done by others. Help with Housework 1. (a) Do you pay someone to help with housework on a regular basis? a. b. d. c. I have Share equally with partner Partner has Not applicable a. b. Yes No If yes, how many hours per week? 106 Appendix B Table 1 Demographic Characteristics of the Sample of Managers (N=264! Characteristic Percent 3 Gender Male 51.1 Female 4 8.9 Racial Background Caucasian 92.4 Chinese 3.0 Indian 0.8 Japanese 1.5 Other Asian 1.1 Education No Degrees/Certificates 31.8 C e r t i f i c a t e / s from Trade, Technical, Vocational or Business School 14.0 Bachelors Degree 25.8 Bachelors Degree with l or more C e r t i f i c a t e s 9.1 Masters Degree 17.8 Doctorate Degree 1.5 T i t l e CEO 35.6 Functional Manager 41.3 Middle Manager 9.5 Senior S p e c i a l i s t 2.3 Unit Manager 6 . 8 F i r s t Line Supervisor 2.3 No Management Position 0.8 Firmsize 1 " 9 1.9 1 0 " 4 9 22.0 50-99 1 4 4 100-499 3 0 3 500-999 6 ' ± 1000-1999 3 ; 8 2 000 or more 20 1 107 Appendix B Table 1 (Continued) Charac t e r i s t i c Percent 3 Number of Employees Supervised None 4 .9 1-9 47 . 0 10 or more 47 .0 Type of Firm Public 33 .0 Pr i v a t e - f o r - P r o f i t 59 . 1 Private-not-for-Profit 5 . 7 Industry Service 81 .5 Manufacturing 17 . 6 Marital Status Single, not involved i n a relationship 9 . 8 Involved i n a relationship but not married 2 .3 Livi n g with a partner but not married 7 . 6 Married 80 .3 Number of Children None 27 .3 One 14 .8 Two 33 .7 Three 17 .4 Four 6 .1 Five 0 .4 Seven 0 .4 Expect to Add Child to Family Yes 10 .6 No 82 .2 Maybe 0 : 4 a Totals do not sum to 100% due to missing data. 108 Appendix B Table 2 Demographic Characteristics of Male (n=135) and Female (n=129) Managers Charac t e r i s t i c Percent Men Women Racial Background Caucasian Chinese Indian Japanese Other Asian 91.9 2.2 1.5 3.0 0.7 93 .0 3.9 0.0 0.0 1.6 Education No Degrees/Certificates C e r t i f i c a t e / s from Trade, Technical, Vocational or Business School Bachelors Degree Bachelors Degree with 1 or more C e r t i f i c a t e s Masters Degree Doctorate Degree 31.1 14 .8 24 .4 12 .6 14 . 8 2.2 32 . 6 13 27 5.4 20.9 0.8 T i t l e CEO Functional Manager Middle Manager Senior S p e c i a l i s t Unit Manager F i r s t Line Supervisor No Management Position 38 42 5 2 7 1 0 32 40 13 2 6. 3 1 Firmsize 1-9 10-49 50-99 100-499 500-999 1000-1999 2000 or more 1.5 2.3 20.7 23.3 17.0 11.6 30.4 30.2 3.0 9.3 3.0 4.7 22.2 17.8 Table 2 (Continued) Appendix B Percent 3 Characteristic M e n Women Number of Employees Supervised None 3.0 7.0 1-9 44.4 49.6 10 or more 51.9 41.9 Type of Firm Public 31.9 34.1 Pr i v a t e - f o r - P r o f i t 63.7 54.3 Private-not-for-Profit 2.2 9.3 Industry Service 75.2 88.2 Manufacturing 22.2 12.6 Marital Status Single, not involved i n a relationship 3.0 17.1 Involved i n a relationship but not married 3.0 1.6 Living with a partner but not married 0.7 14.7 Married 93.3 66.7 Presence of an Employed Spouse No 40.0 22.5 Yes 58.5 75.2 Number of Children None 15.6 39.5 One 11.1 18.6 Two 37.8 29.5 Three 25.2 9.3 Four 8.9 3.1 Five 0.7 0.0 Seven 0.7 0.0 Appendix B Table 2 (Continued) Percent 3 C h a r a c t e r i s t i c Men Women Expect to Add Child to Family-Yes 8.1 13 .2 No 85 .9 78 .3 Maybe 0.0 0.8 Note. Additional descriptive information i s presented i n Table 7 on page 66. a Totals do not sum to 100% due to missing data. co * cn o CM CO H • CM in CM in in in «* in - * vo cn in in cn > H co i > H CM CM ^ O) r l r l co in • • • VO CO CO CO CM C M cn vo co vo cn C M o H in in C M cn vo vo cn C M C O C O H to r> V D ^ J * flflrlrl C M co t-» co H in co co cn cn ^* co co i n vo H i n r ~ co CM H H co co • • • r H i—I VO C O "tf C M vo in C M C O t-v VO ' t f CO VO H cn C O H H co cn cn co o vo 0) U CJ CJ crj a rd a) c a t . CO CJ N •H W CO CJ CJ >i o r H ft a CJ 10 w in CO CM C I • r ( H 10 rH (0 10 CO CO <D CJ >i >i O O H H a ft e co o 0 o in H 1 I VO H C M in 10 co 0) >i 10 (0 o 0) CO r H CO 0) ft O O CJ r H r H ft ft vl) g e r l CO 0) O e r l O o o in o C M in H H H o in o H C M in CO CO CJ 0) c o •H -H co <w cq o m r H o rd o CJ o ftrJ >1 EH CO O •H d) <M O MH -H O <W <(H X! O O C Tf rd rd U CJ pa K 112 oo O C M V D C T » H C J i ' t f ' f f C M H H H H CM i n o ^ v o v o r v r - ^ i n c o r l H r l N r l N O r i K i r i M i m o n H H H H C l i n H O O V O O C T l O ^ (O N f l > H C o • r l •p Ul (0 d) rH -p -H o tn O -p 10 M C 3 CT>-H Tj < g c H u r < -p O CP cd C M-l •P (0 -P rl O ft Ul c cd rl EH 0) (0 TJ C (0 H H +J «• Cd (0 0) rl -P Ul 0) O H CJ Cd (0 •H « 10 0) o C -H •H i—I 10 Xi CO o. i n H CM r » i n H o o o CT> CO 0 0 ^ \D CO H o H i n o o 0 0 O r-l O O in CM H •<* O O u to -p -p CO ft-p •H C TJ (0 (0 c •H c o •rl X IH d) O -p 10 Tj tO Ul cd M C -H cd r - l (1) 0) r H - Q o c ft o ft-P Q) C rl r tTj fl O lU C rl w u H fe m <1) e o c •H TJ TJ r H o c -p 0 c 0) rH rd to rC -p TJ c cd to r-l d) r Q 1 0) TJ o o >1 0) > r-l to 0) .c: •p TJ d) •p 0) rH d) TJ to -p c -p c CO o • H M—I . c o to I c o c TJ Q) c o • H •p (0 r-l o ft r-l o o TJ c • H rd <d _ , CM rl . •p o d) Ul cd £1 cd -p cd TJ a) r H -p >1 A TJ 0) TJ •rH > o u ft -p rd r 3 •p to TJ c cd to - T ' r l 2 * e o <4-l = rl o rl o TJ O 0) -P TJ C 0 - H H O TJ a) N rH s o Q) +J to o a) - H • H TJ rl 4-> Ul to cd TJ c • H e—I 43 rji cd c Q) to -p c a) cu TJ TJ C c o O ft ft to to a) d) rl rl <rl a) o TJ a) -p c a) m Ul rl ^ Q, C £ <d -P - H g rl cd > a) rl xi EH ai •p o Q) 3 £ -P > 3 -P U C •H cd rl O < <4H V c c Q) P I > w * 1 1 3 CO H cd to s i CO vo CTl O CN C M V O CN co CN CN rd -P O E H vo C M C N vo C M V O C M V O C M Q CO o cn cn cn cn co fc CO Tj C CO o > Si CO CO fc o 0 CO CO o c CO fc CO iw CO fc o 10 c o •H -p rd •H > CO TJ fc rd TJ c rd -P CO TJ c rd CO c rd CO c CO g o c CO SI §1 SI Cl TJ CO -p c CO •H fc 0 1 -p •H <M CO c >i CO H cq •H g rd fa n co cn CM co C M C M H C O C M cn C M H 00 CM cn CM in V O C M cn C O • cn C O • cn cn o vo co H C M C M in in in in co C O C O C O H H H H X! C fc -P -H 0 •H C 2s 0 •H >, •P Cn -P •P -P •H C 0) C TJ rd -H H -rH g CO CO CO 0 rH CO CO •H rH •H g -p -p 0 -H rH -P g XI 3 -P rd •H rH Si £1 •H rd •rl TJ CO | 0 fc U •rl •H 4H fc X! CO fc •P H 1 CO * X X CO cn CO X! 3 fc ft 0 c fc 0) CO c 0 H 0 0 rd g fc CO 0 H rH CO fc fa CO si CO ftXl S MH fa XI ft 114 SI co co CO CO cn CN CO a ro SI CM O CO CO CM CM vo CM CM VO cn CM (0 -P O En vo CM VO CM CO vo CM VO CM Tj CO c •H -P c o u co co cn co co fc co TJ c a) CJ > l £ t Ul co u o o to co o c a> u 0) <u CO u cu <4H o to c o •H -p •H > CO a -o u rd TJ c rd -p CO Tj c rd CO c rd CO c CO a o CO s SI TJ CO -P C CO -P •rH - H >H <4H O 0) I C >i 0) r H OQ • r l a rd fa cn co cn CM CM CM cn CM t-> CM CM CO CM If) O CO CM CO o H rH CO cn CO vo vo CO CO H CM CO • • • • CM CM CM CM ID IT) in «tf CO CO CO CO H H H H >1 TJ rH c rH •r l rd cd a 0) C rd Cn C > 0 fc <u C 0 c CO • r l - r l o ^ Q) - H •r l o o -P -P <W rH U H + J CO •r l c to fc rd rd co rd H rd CO •P cd a 0 N CO to c c rH CO a 0 rd -P rd ft-H fc CO o o ft CO fc •r l U CO fc ft C 3 > co cfl X c 0 > O •r l 0> 3 rd +J rd fc rd r l 3 <W fc rH CO 0 CO CP rH CO CO CO O 0 c CO CO CO fc fc 3 r l ftrl at o •r l lfl « cd ft < O 0 115 T j d) c - H •P e o cj u d) T J c <u o >l 4a o> u o o to 01 . o c 0) rH 0) 4-1 0) r-l OH (W O Ul c o •H •P (0 • H > 0) Q T J rH rd TJ c rd -p CO T J c rd ui c rd 0) 0) rH ft rd co rd -P O EH 0) o c 0) g g! SI SI SI TJ 0) -p c 0) • H -p r l <rl 0 0) 1 C > i 0) rH m • H e rd fe o co ro co vo CM o r-vo co ro oo CN vo CN co IT) ro 0) U O Ul - H > u 0) <w ft < Ul 42 • r l X 0) 0) -P rH •rH O Ul VH 00 u ~ • rd •>* H CO CM c u 0 1 c o Ul rd d - p •rt * < r l H <1) o C • 0) • H 43 0) • II <D rd . O X > i • H colCO TJ I" C vo • H • H 0) 43 TJ •P C o r l rd C 0) T J (1) 4-> 0) rd T J 0) 43 -P r l O 4H Q) U O o Ul r l o <u (1) — o 0) ro C T J CO QJ r l 0) 0) Ci r l ft -P • r l CD 43 VO C -p vo rd . Q) u n O II 4-i <D C3I43 rll m l I i r l — o 0 <u Ul ro i—I • ft at H g O rd C Ul Ul d) rd r l > H 0) rd • <H r ^ 4 J 0) Ul O r» M 0) -P in ft-H CM +J Q) C - H 43 II rd H -p d e o u rd O 0) <H fe VO 43 CO EH 0) • a) •P o (3 II o wl 116 Appendix E Factor Loadings for Dependent,Independent and Potential Control Variables Factor Variable Preferences for Family-Oriented Benefits (SW) .48 Family Involvement (SF) .59 Work-Family C o n f l i c t (SF) .51 Gender (OF) .79 Work Involvement (SW) .74 Work Hours (OW) -.59 Hours Worked at Home (OW) .59 Employment Sector (OW) -.52. Job Tenure (OW) .47 Salary (OW) -.51 Age (OF) -.47 Marital Status (OF) -.33 An Employed Spouse (OF) .69 Spouses Work Hours (OF) .80 Number of Children (OF) .44 Hours i n Household Tasks .66 (OF) Hours Devoted to Dependent .74 Care (OF) Responsibility for -.82 Household Labour (OF) Help with Housework (OF) .61 Note. SW=Subjective work variable; SF=Subjective family variable; OW=Objective work variable; 0F=0bjective family variable. 

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.831.1-0087257/manifest

Comment

Related Items