Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Work-family role strain among employed mothers of preschoolers: the impact of workplace support Warren, Jennifer A. 1993-09-22

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
831-ubc_1993_spring_warren_jennifer.pdf [ 4.2MB ]
Metadata
JSON: 831-1.0086183.json
JSON-LD: 831-1.0086183-ld.json
RDF/XML (Pretty): 831-1.0086183-rdf.xml
RDF/JSON: 831-1.0086183-rdf.json
Turtle: 831-1.0086183-turtle.txt
N-Triples: 831-1.0086183-rdf-ntriples.txt
Original Record: 831-1.0086183-source.json
Full Text
831-1.0086183-fulltext.txt
Citation
831-1.0086183.ris

Full Text

WORK-FAMILY ROLE STRAIN AMONG EMPLOYED MOTHERS OF PRESCHOOLERS:THE IMPACT OF WORKPLACE SUPPORTbyJENNIFER ANN WARRENB.A., The University of British Columbia, 1990A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF ARTSinTHE FACULTY OF GRADUATE STUDIES(Department of Family Studies)We accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAMarch, 1993©Jennifer Ann Warren, 1993In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature) Department of  f---priA^STuDlE-5The University of British ColumbiaVancouver, CanadaDate^pti2L1-1 25 icici3DE-6 (2/88)iiAbstractThis study examined the impact of work demands andworkplace support on perceived work-family role strain ofemployed mothers with preschool age children in group daycare.Structural and psychological work demands were investigated.Components of workplace support included organizational culture,supervisor support, and family-oriented benefits offered by theworkplace. The relationship between available family-orientedbenefits and the use of family-oriented benefits was alsoassessed. In addition, moderating effects of supervisor supportwere investigated.The sample was recruited through licensed group daycarecenters in Vancouver. Questionnaires were left at 45 daycarecenters and participants were asked to return their completedquestionnaires to the School of Family and Nutritional Sciencesin the stamped, self addressed envelope provided. Eligibleparticipants were mothers who were employed outside the home ina position where they had an immediate supervisor, manager orboss. The respondents were 116 women who met the eligibilityrequirements and completed the questionnaire.The major findings of this study were that work-family rolestrain was associated with psychological work demands, workenvironment support, supervisor flexibility, and the desire touse family-oriented benefits. The results also showed that thenumber and percentage of family-oriented benefits used wererelated to the number of available family-oriented benefits. Iniiiaddition, respondents were more likely to use family-orientedbenefits under nonsupportive conditions. Although it wasexpected that supervisor support would moderate the relationshipbetween work demands and work-family role strain, and therelationship between use of benefits and work-family rolestrain, the results of this study did not provide support foreither prediction.ivTable of ContentsPageAbstract ^  iiTable of Contents ^  ivList of Tables  vAcknowledgements ^  viChapterI. Introduction. ^  1Purpose  4Conceptual Framework: Role Strain Theory ^ 5II. Review of the Literature ^  7Hypotheses ^  20III.Method ^  23IV. Results  38Response Rate ^  38Sample Characteristics ^  38Univariate Distributions  39Preliminary Analysis ^  45Hypothesis Testing  46Post Hoc Analysis ^  60V. Discussion^  65Limitations  82Conclusion ^  84Implications  86References ^  89Appendices  94List of TablesvPageTable 1. Frequencies of Use for Each of the 13Family-Oriented Benefits ^ 42Table 2. Correlations Between Sociodemographic Variablesand Work-Family Role Strain 47Table 3. Intercorrelations Between Work DemandVariables ^ 49Table 4. Intercorrelations Between Supervisor SupportVariables 54Table 5. Correlations Between Control Variables andWork-Family Role Strain ^ 62Table 6. Intercorrelations Between Family-OrientedBenefits and Supervisor Support Variables ^ 64viAcknowledgementsI would like to express my sincere appreciation to theSupervisor of my thesis committee, Dr. Phyllis J. Johnson, forher genuine commitment to my research and her enthusiasm aboutmy work. I would also like to thank the members of my thesiscommittee, Dr. Jim White and Dr. Nancy Langton, for theirguidance throughout the completion of this thesis.Finally, I would like to thank my aunt, Pam Harrison,without whose assistance I would never had made it through thegates.This research was funded by a UBC-Humanities and SocialSciences (HSS) Small Research Grant.1Chapter IIntroductionUntil relatively recently, employment and family werebelieved to be distinct domains operating independently of oneanother (Voydanoff, 1987). According to Kanter (1977), thisbelief was based on a traditional sex role ideology and theProtestant Work Ethic. Men worked outside the home performingthe breadwinner role and women performed household and childcaretasks within the home. When performing the work role, outsidethe home, individuals were expected to "act as though" they didnot have any other commitments or interests such as familyresponsibilities.Recent demographic changes affecting employment and familylife, however, have challenged this "myth of separate worlds"(Kanter, 1977) and have made it increasingly clear that the twodomains are interdependent. One of the most significantdemographic changes has been the increasing labour forceparticipation of women. According to Statistics Canada (1990),more than half (55.9%) of all Canadian women work outside thehome, and the majority (60.6%) of all women with children livingat home are in the labour force. While the participation ratesof women with children under 16 years has risen steadily, theincrease has been dramatic for mothers with preschool agechildren (Ontario Women's Directorate, 1991). Among women withhusband/partners present, the participation rates of those withonly preschool age children increased from 36.5% in 1976 to262.1% in 1986 (Statistics Canada, 1990). The participation ratesof women in lone-parent families with only preschool agechildren rose from 48.5% in 1976 to 59.2% in 1986 (StatisticsCanada, 1990).Although an increasing number of women are taking on the"non-traditional role" of provider, they continue to beprimarily responsible for household and childcare tasks. Severalstudies have found that married women spend a great deal moretime performing housework and taking care of children thanmarried men (Hochschild, 1989; Kome, 1982; Michelson, 1985).The recognition that women are combining paid employmentwith raising a family has in part led to a fervour of researchactivity directed at the impact of multiple role demands onwork-family outcomes. Research in this area has demonstratedthat a significant proportion of employed individuals areexperiencing some or a great deal of difficulty in managing workand family life (see review by Friedman, 1987, p.40-41).Employed parents report higher stress levels and greater work-family interference than non-parents (Hughes & Galinsky, 1988),and parents of two earner households and single-parent femaleearners report higher levels of work-family role strain thantwo-parent single earner families (Kelly & Voydanoff, 1985).When compared to employed fathers and employed mothers of olderchildren, employed mothers of preschool age children are morelikely to report spillover between work and family (Crouter,1984) and greater work-family role strain (Greenberger,3Goldberg, Hamill, O'Neil, & Payne, 1989; Kelly & Voydanoff,1985). Thus, while an increasing number of women with preschoolage children are in the labour force, this group also appears tobe the most susceptible to work-family difficulties.Research attention has been directed to ways of coping withdifficulties associated with combining employment and familyresponsibilities (see reviews by Hansen, 1991; Menaghan &Parcel, 1990). Research in this area typically focuses on howindividuals manipulate employment and family demands in anattempt to create patterns of relationships and activities thatare manageable (Elman & Gilbert, 1984; Moen & Dempster-McClain,1987; Piotrkowski, Rapoport, & Rapoport, 1987; Presser, 1987).Although individual level coping strategies may beeffective in helping the employed parent balance employment andfamily roles, such solutions have been considered "unsatisfying"as they fail to fully address the nature of the difficulties(Menaghan & Parcel, 1990, p.1089). It has been argued that theindividual approach is based on traditional values and attitudesabout employment and family that expect the employed parent todevelop idiosyncratic ways of coping with problems. Inconforming to the traditional work model, which has beenconsidered "plainly out of synchronization with the family livesof many workers" (Kamerman & Kahn, 1987, p.59), prevailing workvalues are perpetuated and family needs are ignored (Duffy,Mandell, & Pupo, 1989; Menaghan & Parcel, 1990).In an attempt to find a more satisfying solution to4difficulties associated with combining employment and familylife, increasing attention is being directed to the role of theworkplace (Menaghan & Parcel, 1990). Thus, instead of focusingon what individuals can do to cope with employment and familydifficulties, this course of study seeks to answer the question,"What can the workplace do to facilitate the integration ofemployment and family life?"PurposeThis study examines the relationship between the workenvironment and work-family role strain perceived by employedmothers with preschool age children. In this thesis, it ishypothesized that feelings of strain may be enhanced by thedemands of the job and reduced by a workplace that is supportiveof employees' family responsibilities.The objective of this study is to address five questions.First, what is the impact of work demands on work-family rolestrain? Second, what is the impact of workplace support on work-family role strain? Third, what is the relationship betweenavailable family-oriented benefits and use of family-orientedbenefits? Fourth, does the perception of supervisor supportmoderate the impact of work demands on work-family role strain?Finally, does the perception of supervisor support moderate theimpact of use of family-oriented benefits on work-family rolestrain?While it is recognized that the performance of householdand childcare tasks within the home is considered work,5references made to "work" throughout this thesis refer solely topaid employment performed outside the home. The term "family"refers to parenting as opposed to elder care responsibilities.Conceptual FrameworkAlthough no integrated theory of work-family relationshipsexists, theoretical grounding for research in this areagenerally reflects role strain and role expansion theory(Voydanoff, 1989). The proposed research is guided by Goode's(1960) role strain theory.According to Goode (1960), when engaging in rolerelationships, an individual is faced with a wide array ofdistracting and sometimes conflicting role obligations. Becausethe individual's total role obligations are overdemanding and itis impossible to fulfill all of one's role demands, some degreeof role strain or dissensus occurs. This "felt difficulty infulfilling role obligations" (p.483) is considered normal andunavoidable when social structures are viewed as made up ofroles. It is reasonable to assume that individuals performingthe multiple roles of worker and parent simultaneously willperceive some degree of inadequacy or difficulty in meeting workand family role obligations because the cumulative demands areoverdemanding.Goode's role strain theory identifies two sets oftechniques that can be used by the individual to reduce rolestrain: (1) those which determine whether or when the individualwill enter or leave a role relationship; and (2) those which6have to do with the actual role bargain which the individualmakes or carries out with another (1960, p. 486). Thesetechniques are individual level coping mechanisms that requirethe individual to manipulate his or her role structure.Although role strain theory focuses on coping mechanisms atthe individual level, current work-family literature hasaddressed ways of reducing work-family role strain that arebeyond the individual (Bowen, 1988; Hughes & Galinsky, 1988;Kamerman & Kahn, 1987). Thus, for this study, role strain theoryprovides the conceptual framework for examining work-family rolestrain and investigating ways of reducing role strain. Currentwork-family research provides the basis for expanding rolestrain theory to incorporate institutional level copingmechanisms, such as workplace support, as a means for reducingwork-family role strain.7Chapter IIReview of the LiteratureIn this section, research examining the impact of workdemands and workplace support on work-family role strain isreviewed. Empirical investigations of the interaction betweenuse of family-oriented benefits and supervisor support are alsodiscussed.Structural and Psychological Work Demands While early studies of work-family linkages focused on theimpact of employment status on family life, current empiricalinvestigations have examined structural and psychologicalaspects of the work role for relationships with work-familyoutcomes (Voydanoff, 1989).Structural work demands. Work-family research shows thatstructural work demands such as the amount of work time and thescheduling of the work week are related to difficultiesassociated with work-family coordination (Voydanoff, 1987).Individuals working long work weeks are more likely to reporthigher work-family role conflict (Pleck, Staines, & Lang, 1980)and role strain (Keith & Schafer, 1980; Voydanoff & Kelly, 1984;Voydanoff, 1988). The number of hours one's spouse works perweek also has implications for personal well-being. Keith andSchafer (1980) found that the number of hours the husbandworked, the greater the wife's work-family role strain. Wife'semployment hours, however, were not related to husband's rolestrain. Atypical work schedules may make it difficult to8coordinate work and family activities and thereby contribute tofeelings of strain. Pleck et al. (1980) showed that employedparents working afternoon, evening, and irregular shiftsexperienced greater schedule incompatibility between work andfamily life. Staines and Pleck (1983) found that working non-day, weekend and variable shifts was associated with higherlevels of work-family conflict.Psychological work demands. In addition to structuralcharacteristics of the work role, psychological aspects havebeen identified as predictors of work-family outcomes. Researchhas shown that psychological work demands such as heavyworkloads and pressure for output are related to work-familyrole conflict (Voydanoff, 1988) and work-family role strain(Katz & Piotrkowski, 1983).Workplace Support It has been suggested that a work environment that issupportive of employees' family responsibilities may help toimprove employed parents' ability to balance work and familylife and reduce associated strain (Bowen, 1988; Kamerman & Kahn,1987; McCroskey, 1982; Voydanoff, 1987). While research in thisarea is limited, two courses of study have been taken in theinvestigation of workplace support: (1) research examiningformal family-oriented benefits; and (2) research examiningsupervisor support.Family-oriented benefits. Family-oriented benefits refer tocompanies' formal benefits that are designed to help employees9coordinate work and family responsibilities (Raabe, 1990). Muchof the research on family-oriented benefits has focused on thenature of these benefits, the extent to which they are availableto employees (Paris, 1989; Raabe & Gessner, 1988), and thereceptiveness of workplaces to their implementation (Axel, 1985;McNeely & Fogarty, 1988). Research examining the relationshipbetween use of family-oriented benefits and work and familyrelated outcomes, however, has been limited (Voydanoff, 1989).Two benefits that have received the most extensive researchattention in this area are employer-sponsored child care andflextime.National surveys and empirical studies of individualcompanies have investigated the impact of employer-sponsoredchild care on work-related outcomes. Results of these studiesconsistently support the positive effects of this type of careon measures of productivity. Employers and human resourcemanagers generally report improvements in work outcomes such asemployee morale, employee work satisfaction, retention,recruitment and reductions in tardiness, absenteeism, andturnover (Burud, Aschbacher, & McCroskey, 1984; Magid, 1983;Perry, 1982). Youngblood and Chambers-Cook (1984) also foundthat the implementation of an "in-house" day care centre wasassociated with greater overall job satisfaction andorganizational commitment, and lower turnover intentions.While studies of employer sponsored child care havegenerally focused on work-related outcomes, one study was10identified in the literature that investigated the effects ofthis benefit on the integration of work and family life. Goff,Mount, and Jamison (1990) examined the impact of on-site childcare on work-family conflict for a sample of 253 parents whoworked in a large midwestern United States' electronics andcommunications firm. Contrary to what was expected, the resultsshowed that parents using on-site child care did not reportlower levels of work-family conflict than parents who were notusing this type of childcare arrangement.Although working irregular shifts has been identified as asource of work-family role strain (Pleck et al., 1980), work-family literature considers flextime a family-oriented benefitbecause of its suggested positive effects on work-familyoutcomes. Investigations of flextime have addressed non-workoutcomes such as work-family role interference and job-familystress (Bohen & Viveros-Long, 1981; Lee, 1983; Shinn, Wong,Simko, & Ortiz-Torres, 1989; Winett, Neale, & Williams, 1982).Shinn et al. (1989) found no relationship between access toflextime and perceived work/family interference for their sampleof 644 working parents (208 married fathers, 287 marriedmothers, 149 single mothers). Lee (1983) investigated therelationship between access to flextime and perceived stressassociated with family activities using a sample of 100 marriedemployees of a British research organization. Access to 2 hoursof flextime per day was associated with reduced stress relatedto childcare activities and child socialization activities.11Winett et al. (1982) examined the impact of flextime on 71working parents with children under the age of 13 years.Subjects were employed by two federal agencies in Washington,D.C. The use of flextime was associated with lower levels ofperceived work/family interference.Bohen and Viveros-Long (1981) examined the impact offlextime on perceived job-family stress for a sample of 393federal agency employees (200 men, 193 women). Comparisons weremade with a sample of 313 employees (172 men, 141 women) withsimilar background characteristics, working in a standard timefederal agency (no access to flextime). Employed women andemployed parents on flextime reported significantly lower levelsof job-family stress than those on standard time. When theparent group was divided by sex, however, mothers on flextimedid not report less stress than those on standard time. Contraryto the proposed hypothesis, flextime provided greater benefitsto childless women and single adults than to employed parents orsingle mothers.The studies cited above do not provide overwhelming supportfor either employer-sponsored child care or flextime as viablesolutions to work-family problems. However, it has been arguedthat the inconclusive results of these studies may be due tomethodological problems such as variations in the dependentmeasures and their definitions, inadequate statistical controlsbetween samples, and differences resulting from studying theeffects of access rather than use of benefits (Christensen &12Staines, 1990). It has also been suggested that the use of asingle benefit may not make a measurable difference in reducingjob-family stress for employed parents with a high level offamily-related obligations (Bohen & Viveros-Long, 1981).Instead of focusing on a particular family-orientedbenefit, such as employer-sponsored child care or flextime,Greenberger et al., (1989) sought to examine the relationshipbetween the number of family-oriented benefits used and work-family role strain. The sample consisted of 80 married men, 169married women, and 72 single women who were employed and had apreschool age child. Respondents were asked to identify, from alist of 20 family-oriented benefits, those benefits which wereoffered by their employer and which they had used. The number ofbenefits used by the respondent was summed and respondents werecategorized into one of three groups: those using no family-responsive benefits, those using a single family-responsivebenefit, and those using more than one family-responsivebenefit. The number of benefits used was a significantpredictor of work-family role strain for both married and singlewomen, but not for married men. While single women using morebenefits reported lower levels of work-family role strain, theusage of benefits was associated with higher levels of work-family role strain for married women. The authors suggest thatthis may be due to the type of benefit used. Married women weremore likely to report having used paid disability leave(maternity leave) than single women and therefore, were more13likely to have a young child at home.Due to their focus on the use of multiple benefits ratherthan on a specific family-oriented benefit, Greenberger et al.'s(1989) research helps to provide a broader understanding of therelationship between family-oriented benefits and strainassociated with combining work and family roles. Their measureof family-oriented benefits provides useful informationregarding the number as well as the type of benefit used. Amajor limitation of this study, however, is that data were onlycollected on family-oriented benefits that had been used by therespondent. Respondents were not asked to provide informationabout current use of family-oriented benefits or their spouse'suse of family-oriented benefits. Data could also be supplementedby asking respondents to indicate family-oriented benefits thatthey would consider using if they were available.Supervisor support. In addition to formal benefits designedto ease the difficulties associated with combining work andfamily roles, supervisor support has been identified as acentral component of workplace support.Numerous studies on job stress have investigated therelationship between supervisor support and employees' well-being (see reviews by House, 1981; Vaux, 1988). Job stressresearch has demonstrated main effects between supervisorsupport and various measures of mental and physical health, aswell as stress-buffering effects of supervisor support betweenjob conditions and well-being (Holahan & Moos, 1981; Kobasa &14Puccetti, 1983; LaRocco, House, & French, 1980; Repetti, 1987).Although supervisor support has not been the primary focusof work-family research (Hughes & Galinsky, 1988), the potentialeffects of such support on employees' perceived ability tocombine work and family roles have been recognized (Fernandez,1985; Galinsky & Stein, 1990; Greenglass, Pantony, & Burke,1989).In a study of 5,000 management and craft employees of fivelarge technically oriented companies, Fernandez (1985) found asignificant relationship between supportive supervisors andstress at work and at home. Respondents were asked the question,"To what extent does your supervisor support you and your childcare needs?" A greater percentage of employees reportedexperiencing stress at work and at home when they believed thattheir supervisor was not supportive about their child careneeds, than employees who believed that their supervisors werevery supportive about their child care needs.Galinsky and Stein's (1990) research on 71 Fortune 500corporations found that one of the greatest predictors of work-family problems was the supervisor relationship. Employees'perceived ability to balance work and family roles was linked tosupervisor's support of work-family obligations. A supportivesupervisor was associated with lower levels of stress, while anunsupportive supervisor was related to increased levels ofstress.Using a sample of 556 male and female Canadian teachers,15Greenglass et al. (1989) investigated the impact of supervisorsupport on six types of role conflict: professional vs. self,professional vs. spouse, professional vs. parents, spouse vs.parent, parent vs. self, and spouse vs. self. Supervisor supportwas associated with lower levels of total role conflict for bothmen and women but especially for women. While supervisor supportwas only significantly related to total role conflict for men,it had the greatest impact on role conflict between professionalvs. parental roles for women.When investigating the impact of supervisor support onemployees' perceived ability to combine work and family roles,it has been suggested that two dimensions of supervisor supportneed to be considered: sensitivity to employees' familyresponsibilities and flexibility when family needs arise (Hughes& Galinsky, 1988). Although a supervisor may act as a resourceto the employee by providing emotional support for work roleperformance, if she/he is insensitive and inflexible regardingemployee's work-family issues, difficulties associated withmeeting work and family demands may be exacerbated. A fewstudies have assessed these dimensions of supervisor support(Goff et al., 1990; Greenberger et al., 1989; Hughes & Galinsky,1988).Goff et al. (1990) investigated the effect of supervisorsupport on work-family conflict. The sample consisted of 253employed parents (161 male, 92 female) with children 5 years oldor younger. Supervisor support was measured by a 6-item scale16asking respondents to report their supervisor's willingness todiscuss family-related problems and flexibility when emergenciesarose. Work-family conflict pertained to both spillover from thefamily role to the work role and from the work role to thefamily role. Supervisor support was found to be significantlyrelated to the amount of work-family conflict experienced by theemployed parent. As hypothesized, employees who perceived theirsupervisors as sensitive to their family needs and flexible whenfamily emergencies arose, were more likely to report lowerlevels of work-family conflict.Hughes and Galinsky (1988) examined the impact ofsupervisor sensitivity on work-family interference in asubsample of 285 employed parents (83 women and 202 men) whowere married and had at least one child under 18. The samplewas composed of managers, scientists, and clerical/technicalworkers in a large pharmaceutical company. Supervisorsensitivity was measured by eight items from the University ofMichigan Quality of Employment Survey's measure of resourceadequacy (Quinn & Staines, 1979) and three items developed bythe Bank Street research team which were designed to tapsupervisor flexibility regarding family demands. Althoughsupervisor sensitivity was significantly related to stress(feeling overwhelmed and unable to control the important thingsin their lives) for both men and women, this type of support wasonly found to be significantly associated with work-familyinterference for fathers but not for mothers.17Greenberger et al. (1989) examined the relationship betweensupervisor support and work-family role strain for a sample ofemployed parents (80 married men, 169 married women, and 72single women) with a preschool age child. Supervisor support wasmeasured by a four item Overall Supervisor Support scaledeveloped by Caplan, Cobb, French, Harrison, and Pinneau (1975)and a Supervisor Flexibility scale designed by the authors toprovide information about the degree to which respondentsperceived their supervisors as allowing scheduling flexibilityand other latitude when family needs arose. Work-family rolestrain was measured by a 32-item scale developed by the authorsto tap the spillover of pressures from one role into another,conflict between roles, strain within roles, and generalizedrole overload. Neither overall supervisor support nor supervisorflexibility were found to be significant predictors of work-family role strain among married men, married women, or singlewomen.The three studies reviewed above do not provide extensivesupport for the positive effects of perceived supervisor supporton work-family coordination. The lack of significant findingsmay be due to low content validity of scales measuring perceivedsupervisor support and the failure to investigate possibleindirect as well as direct effects of supervisor support onwork-family outcomes.The validity of the scales used to measure supervisorsupport is questionable because they may not be adequately18assessing perceived supervisor support for combining work andfamily roles. In Hughes and Galinsky's (1988) study, supervisorsensitivity for combining work and family roles was measured byan 11-item scale. Of the 11 items, three items measuredperceived supervisor flexibility when family needs arose, andeight items assessed the degree to which respondents perceivedtheir supervisors as resources for performing the work role.Thus, although supervisor sensitivity did not make a differencein reducing work-family interference for employed mothers, itmay have been because the scale was primarily tapping perceivedprovision of resources by the supervisor for performing the workrole rather than sensitivity for combining work and familyroles. One of the two measures used by Greenberger et al. (1989)to assess supervisor support was a four item scale askingrespondents to indicate how much their supervisor "made theirwork life easier, was easy to talk with, could be relied on, andwas willing to listen to their personal problems" (p.765). Thisscale appears to tap emotional support from one's supervisor ina general sense rather than perceived support for combiningemployment with family responsibilities.Family-Oriented Benefits and Supervisor Support In the literature on family-oriented benefits, a link hasbeen identified between family-oriented benefits and supervisorsupport (Hughes & Galinsky, 1988; Kamerman & Kahn, 1987; Raabe& Gessner, 1988)Through in-depth interviews with 30 New Orleans employers,19Raabe and Gessner (1988) found that while employers were oftenmore accommodating regarding work-family issues than theirformal policies suggested, formal policies were sometimesundermined by supervisory practices.Based on their extensive research of corporate policies,Hughes and Galinsky (1988) conclude that although a company mayhave an innovative program, how or if employees make use of theprogram generally depends on the discretion of the supervisor.In their case studies of a variety of corporations,Kamerman and Kahn (1987) found that employees were often notaware of available family-responsive benefits. In addition, agap was often evident between formal and informal policies.Family-responsive programs were sometimes accompanied bypressure not to use available benefits, and inefficient anduncooperative administration.It is apparent that the extent to which individualsperceive their supervisor as supportive of combining employmentwith family responsibilities may influence (1) whether availablefamily-oriented benefits are used, and (2) whether benefits usedare effective in facilitating work-family coordination.Investigations of the use of family-oriented benefits to date,however, have failed to consider possible indirect effects ofperceived supervisor support. While Goff et al. (1990) andGreenberger et al. (1989) include both perceived supervisorsupport and family-oriented benefits in their investigations,neither study examines the relationship between the two20components of workplace support.HypothesesWork-family research has identified structural andpsychological aspects of the work role such as work hours,scheduling of work, and psychological work demands that arerelated to increased difficulty associated with coordinatingwork and family life (Katz & Piotrkowski, 1983; Keith & Schafer,1980; Pleck et al., 1980; Staines & Pleck, 1983; Voydanoff,1988; Voydanoff & Kelly, 1984). Based on this literature, it isexpected that:Hl: The greater the work demands, the greater the work-familyrole strain.While structural and psychological dimensions of the workrole may increase work-family role strain, components of thework environment may help to alleviate work-family difficulties.Although previous research examining the impact of workplacesupport on work-family role strain is limited, it has beensuggested that a work environment that is supportive ofemployees' family responsibilities may facilitate thecoordination of work-family roles (Bowen, 1988; Kamerman & Kahn,1987; McCroskey, 1982; Voydanoff, 1987). Based on this argument,the following hypotheses are advanced:H2: The more supportive the organizational culture, the lessthe work-family role strain.H3: The greater the number of family-oriented benefits used,the less the work-family role strain.21H4: The greater the perceived supervisor support for combiningemployment with family responsibilities, the less the work-family role strain.Due to the high family demands experienced by employedmothers of preschool age children, it is expected that benefitsthat are designed to assist employees in managing work andfamily responsibilities will be used if they are available. Thisrelationship has been assumed but has not been empiricallyinvestigated. It is expected that:HS: The more family-oriented benefits available, the morefamily-oriented benefits used.While it is expected that greater work demands areassociated with greater work-family role strain, thisrelationship may be moderated by perceived supervisor supportfor combining employment and family responsibilities. It ispredicted that:H6a: When work demands are high, individuals with high perceivedsupervisor support will have lower levels of work-familyrole strain than individuals with low perceived supervisorsupport.H6b: When work demands are low, individuals with low perceivedsupervisor support will have higher levels of work-familyrole strain than individuals with high perceived supervisorsupport.Perceived supervisor support may also interact with the useof family-oriented benefits to reduce work-family role strain.22Despite the use of available benefits, role strain may still behigh if the individual perceives low supervisor support (Raabe& Gessner, 1988). It is expected that:H7a: When use of family oriented benefits is high, individualswith high perceived supervisor support will have lowerlevels of work-family role strain than individuals with lowperceived supervisor support.H7b: When use of family oriented benefits is low, individualswith low perceived supervisor support will have higherlevels of work-family role strain than individuals withhigh perceived supervisor support.Perceiving one's supervisor as supportive of employeescombining work and family roles may also influence whetheravailable family-oriented benefits are used (Hughes & Galinsky,1988; Kamerman & Kahn, 1987). Regardless of the number offamily-oriented benefits available, individuals may not use themif they do not perceive their supervisors as supportive. Thus,it is expected that:H8: The relationship between available family-oriented benefitsand use of family-oriented benefits will be stronger whenperceived supervisor support is high than when perceivedsupervisor support is low.23Chapter IIIMethodRecruitment of Subjects The subjects were mothers who were employed outside thehome in a position where they had an immediate supervisor,manager or boss. All mothers had at least one preschool agechild. Subjects were recruited through Vancouver group daycarecenters licensed by the Provincial Child Care FacilitiesLicensing Board. A current list of 66 group daycare centers forchildren aged 3 weeks to 5 years was obtained from the WestCoast Child Care Resource Centre, Vancouver. Four of the centersprovided care for children under 3 years of age, 52 centresprovided care for children aged 3-5 years, and 10 centersprovided care for children in both age groups. Of the 66 centerslisted, nine centers were excluded from the target centersbecause of the special nature of the centers (5 centers forspecial needs children or integrated daycare, 1 center forchildren of teen mothers, 1 center providing only after schoolcare and 2 centers that were "on site" facilities). It wasbelieved that the characteristics of mothers with children insuch centers would not be representative of the largerpopulation of group daycare users.Daycare directors or head supervisors of the remaining 57group daycare centers were contacted in person or by telephoneand asked (1) if any of the mothers with children at theircenter met the eligibility requirements and (2) if they were24willing to allow the researcher to recruit subjects throughtheir center. Daycare directors/head supervisors from 45 centersagreed to participate and signed the permission form, requiredby the U.B.C. Ethics' Committee, for the recruitment of subjectsthrough their daycare center (see Appendix A). The 12 centresthat did not participate did so for the following reasons: (1)the English skills of the mothers were not strong enough tocomplete the questionnaire (n = 7), (2) the daycare center wasno longer running (n = 2), and (3) the director/head supervisorwas not interested or did not think that the mothers would beinterested in participating (n = 3).Data Collection ProcedureOver a one month period, questionnaires were distributed tothe 45 daycare centers that were included in the recruitmentprocess. From 2 to 15 questionnaires were initially left at eachcenter depending on the number requested by the director/headsupervisor. Recruitment notices, outlining the eligibilityrequirements for subjects, were posted in all 45 centers (seeAppendix A). Volunteer participants picked up questionnaires attheir daycare center and were asked to return their completedquestionnaire to the School of Family and Nutritional Sciencesin the stamped, self addressed envelope provided. A daycare codefor each center was marked on the return envelope in order tomonitor their return. A follow-up call was made to the daycaredirector/head supervisor 10 days after questionnaires wereinitially distributed to each center to see whether additional25questionnaires were required and to arrange a time for theresearcher to return to the center to pick up any questionnairesthat had not been taken. During the second and final visit tothe daycare center (approximately 2 weeks after the initialvisit), a reminder notice encouraging participants to returncompleted questionnaires was posted in each centre (see AppendixA).QuestionnaireParticipants were asked to complete an 11 pagequestionnaire (see Appendix B). The questionnaire includedquestions about psychological and structural demands of theirpaid work; perceptions about the supportiveness of theirworkplace and the supportiveness of their supervisor/manager orboss; difficulties in combining work and familyresponsibilities; and sociodemographic information. Questionswere also included to ensure that the participants had met theeligibility requirements (e.g., "Do you have an immediatesupervisor/manager or boss?", "Please list the ages of yourchild(ren) and the type of care arrangement(s) used.").The questionnaire was pretested by a selected group ofemployed mothers of preschoolers (n = 5). In addition tocompleting the questionnaire, these mothers were asked to (1)provide the time it took them to complete it; (2) identifyinstructions they thought were unclear or items they felt wereworded ambiguously; and (3) comment on any difficulties orconcerns that other mothers might experience. As a result of26concerns raised by the pretesting sample, the wording of oneitem of the work-family role strain scale (item 8) was modified."Does not" was removed from the item, "My time off from workdoes not match other family members' schedules well."Measures Work-Family Role StrainThe dependent variable was measured by Bohen and Viveros-Long's (1981) Job-Family Role Strain Scale. This 19-item Likerttype scale was designed to assess worries about adequatelyfulfilling the demands (felt obligations) of both work andfamily roles. Scale items pertain to internalized values andemotions, such as self-doubt, guilt, and pressure associatedwith felt obligations about work and family roles. Unlike otherscales designed to assess work and family arenas separately (howeach role by itself affects individual's well-being such as jobtension, family management), this scale focuses on the points atwhich individual's work and family roles connect or overlap andproduce pressure or tension for individuals. As reported byBohen and Viveros-Long (1981), Chronbach's alpha for this scalewas .72 for their sample of male and female parents. When usedwith a Canadian sample of employed single mothers withpreschoolers, Chronbach's alpha was .82 (see Campbell & Moen,1992).Using a 5-point Likert-type scale ('1' never, '5' always),respondents were asked to indicate the frequency with which theyexperienced the emotions expressed in the 19 statements. Typical27statements include, "I have a good balance between my job andfamily time." (reversed), and "I have more to do than I canhandle comfortably."Reversed items (3, 8, 11, 13, 14, 16, 17) were recoded (1= 5, 5 = 1) and individual items were summed and averaged toarrive at a value for the scale. High scores indicated highwork-family role strain. Internal consistency, as measured byChronbach's alpha, was .85 for this sample.Work DemandsTwo types of work demands were assessed: (1) structuralwork demands, and (2) psychological work demands.Structural work demands. Structural work demands wereassessed by asking respondents about the length of their workweek and the scheduling of their work time. The length of thework week was measured by the average number of hours worked perweek. Work scheduling was measured by a dummy variableindicating whether the respondent worked non-day shifts (coded0, 1), and a dummy variable indicating whether the respondentworked on weekend days (coded 0, 1).Psychological work demands. A 5-item scale was used toassess perceived workload and time pressure associated with theindividual's paid work. This scale was comprised of 4 items fromthe 1977 Quality of Employment Survey (Quinn & Staines, 1979)and 1 item regarding tight deadlines, developed for this study.The item format was a 5-point Likert-type scale ('1' stronglydisagree, '5' strongly agree), and respondents were asked to28indicate how much they agreed with 5 statements about theirworkload and time pressures to get their paid work done. Itemswere summed and averaged. High scores reflected high levels ofpsychological work demands. The reliability of the scale,established by Chronbach's alpha, was .76 for this sample ofemployed mothers.Organizational CultureFor the purposes of this study, organizational culturerefers to the philosophy or set of expectations/beliefs held bythe business organization regarding combining work and familyroles. Two measures were used to assess organizational culture:(1) available family-oriented benefits, and (2) work environmentsupport.Available family-oriented benefits. Family-orientedbenefits refers to companies' formal policies and practices thathave the potential to assist employees with the coordination ofwork and family responsibilities. Respondents were asked tocheck from a list of 13 family-oriented benefits (see AppendixC), those benefits that they knew were available to them intheir current employment position. The list included benefitspertaining to (a) alternate work arrangements (6 items); (b)leave related policies (5 items); and (c) miscellaneous issues(2 items). Respondents marked a check next to "yes", "no", or"don't know" for each of the 13 benefits. Affirmative responsesfor each benefit were counted to achieve the total number ofbenefits respondents knew were available to them at the29workplace.The 13 benefits included in this study were selected froma list, identified by the Conference Board of Canada (Paris,1989), of family-related benefits offered by Canadianworkplaces. Benefits were selected based on the followingcriteria: (1) relevance to combining paid employment withparenting responsibilities; and (2) applicability to a sample ofgroup daycare users. Examples of excluded benefits include eldercare and disabled relative related benefits, and child carerelated benefits such as resource and referral services.The benefits included in this study have been cited in theliterature on workplace support (Axel, 1985; Hughes & Galinsky,1988; Kamerman & Kahn, 1987) and have been included in otherempirical investigations (CARNET, unpublished questionnaire;Greenberger et al., 1989; Raabe & Gessner, 1988).Work environment support. A single item was used to assesshow supportive respondents perceived their work environment tobe of employees with work-family difficulties. This item wasdeveloped for this study. Using a 5-point response scale ('1'not at all, '5' very supportive), respondents were asked toanswer the question, "How supportive is your work environment ofemployees when they have difficulties coordinating work andfamily responsibilities?"Use of Family-Oriented BenefitsRespondents were asked to check from the list of family-oriented benefits mentioned previously, those benefits that (a)30they were currently using or had used in the past year, and (b)they would consider using if they were available. The number ofbenefits that they were currently using or had used in the pastyear was summed to create a total "use" score. The number ofbenefits that they would use if available was summed to createa "future use" score.Respondents were also asked to identify, from the list ofbenefits provided, those family-oriented benefits that werecurrently being used or had been used by their spouse in thepast year. The number of benefits used by the respondent'sspouse was summed to provide the total number of benefits usedby the respondent's spouse/partner.Perceived Supervisor SupportPerceived supervisor support was operationalized by threescales measuring (1) general supervisor support, (2) supervisorflexibility, and (3) supervisor sensitivity.General supervisor support. Four items developed by Caplanet al. (1975) were used to measure the instrumental andemotional support an individual perceives she receives from hersupervisor. When used by Greenberger et al. (1989), in theirinvestigation of workplace support and parental well-being,Chronbach's alpha ranged from .79 to .85 for their sample ofmarried men, married women and single women.In order to establish consistency in response options, thewording of the 4 items was modified so that each item was in theform of a statement rather than a question. Respondents were31asked to indicate, using a 5-point Likert-type rating scale ('1'strongly disagree, '5' strongly agree), how much they felt theirimmediate supervisor, manager or boss made their work lifeeasier, was easy to talk to, could be relied on, and was willingto listen to their personal problems. Items were summed andaveraged with a high value reflecting high levels of generalsupervisor support. Chronbach's alpha for this sample was .83.Although this general measure of supervisor support was notdesigned to tap supervisor support for combining work and familyresponsibilities, it was included in this study in order tocompare the influence of general supervisor support withsupervisor sensitivity and supervisor flexibility.Supervisor flexibility. A 9-item Supervisor FlexibilityScale, developed by Greenberger et al. (1989), was used tomeasure the degree to which respondents perceived theirimmediate supervisor, manager or boss as allowing schedulingflexibility and other latitude when family needs arise.Chronbach's alpha was reported by the authors to be .88 formarried mothers of preschool age children and .90 for singlemothers of preschool age children.In order to increase the number of response options, theoriginal 3-point Likert-type response scale ('1' seldom ornever, '2' sometimes, '3' usually or always) was expanded tocreate a 5-point Likert-type rating scale ('1' never, '2'seldom, '3' sometimes, '4' usually, '5' always). Respondentswere asked to indicate the extent to which each of nine32supervisory practices applied to their own work situation. Anitem typical of the scale is, "My supervisor/manager lets mecome in late or leave early to accommodate my family needs."Scores for each item were summed and averaged such that thehigher the score, the greater the perceived supervisorflexibility. Chronbach's alpha for this sample was .84.Supervisor sensitivity. Due to the unavailability of anexisting scale, an 8-item Supervisor Sensitivity Scale wasdeveloped to assess respondents' perceptions of theirsupervisor, manager or boss as aware and understanding ofemployees work-family responsibilities. Items included in thescale were based on definitions and characteristics of sensitivesupervisors/managers identified in the research literature onworkplace support (Bowen, 1988; Fernandez, 1985; Galinsky &Stein, 1990).The validity of this scale was evaluated by 3 professorsand 2 graduate students in the School of Family and NutritionalSciences. Evaluators were asked to answer the followingquestions (1) Are the items stated in a clear and concisemanner?; (2) How relevant is each item to perceived supervisorsensitivity for employees' family responsibilities and workdifficulties?; (3) Are there other aspects of perceivedsupervisor sensitivity that need to be considered?; and (4) Arethe supervisor sensitivity and supervisor flexibility scalestapping different dimensions of supervisor support? The numberof items and the wording of the items was not altered, however,33the original 4-item Likert-type response scale was changed to a5-item response scale in order to include a middle ground orneutral response category.Respondents were asked to indicate the extent to which theyagreed with each of the 8 statements about their immediatesupervisor/manager or boss using a 5-point Likert-type ratingscale ('1' strongly disagree, '5' strongly agree). Reverseditems (4, 6) were recoded (1 = 5, 5 = 1). Individual items weresummed and averaged such that the higher the score, the greaterthe perceived supervisor sensitivity. Chronbach's alpha for thissample was .88.Control VariablesOccupational role commitment. Occupational role commitmentwas included as a control because women who are committed totheir paid work role may take on more challenging tasks at workand may devote more time and energy to their work. As a result,they may perceive their work as more psychologically demandingand may also perceive greater difficulties in meeting all oftheir work and family obligations.Occupational role commitment was evaluated using a 5-itemscale developed by Amatea, Cross, Clark, and Bobby (1986).Commitment was defined by the authors as "the extent to whichthe person demonstrates a willingness to commit personalresources to assure success in the role or to develop the role"(p.832). Amatea et al. (1986), reported Chronbach's alpha to be.83 for their married couple sample.34Respondents indicated their agreement with the 5-itemsusing a 5-point Likert-type rating scale. Reversed items (item1) were recoded (1 = 5, 5 = 1) and individual items were summedand averaged to provide a total score for the scale. Thereliability of the scale for this sample was .81 when one itemwas eliminated. The item "I want to work but I do not want tohave a demanding job/career" was eliminated due to its very lowcorrelations (r < .15) with other scale items and the resultingdepressive effect it had on the alpha coefficient for the scale.Family role commitment. Due to the emotional involvementand time devoted to family responsibilities, women who arecommitted to their family role may have difficulty fulfillingtheir work demands. Family role commitment may differentiallyaffect psychological work demands and perceived work-family rolestrain. Thus, family role commitment was included as a controlfor this study.Three items from the 1977 Quality of Employment Survey(Quinn & Staines, 1979) were used to measure commitment to thefamily role. Using a 5-point Likert-type response scale ('1'strongly disagree, '5' strongly agree), respondents indicatedtheir agreement with three statements about how important theirfamily is to them. Individual items were summed and an averagescore for the scale was computed. Chronbach's alpha was .77.Outside help. The use of outside help was measured by adummy variable indicating the presence or absence of outsidehelp with household chores on a regular basis (coded 0,1). This35variable was included as a control because women who hireindividuals to help with household chores may experience fewerfamily demands and thus may perceive lower levels of work-familyrole strain.Occupation length. The length of time one is employed in anoccupation position may influence the supervisor-superviseerelationship as well as access to and eligibility for family-oriented benefits. As a result, occupation length was controlledfor in subsequent analyses.This variable was measured by the number of monthsrespondents had been employed in their current employmentposition with their present employer.Difficulty finding alternate child care. Although previousresearch has not investigated difficulty finding alternate childcare and work-family outcomes, this variable was included as acontrol because finding alternate care arrangements has beenidentified as problematic for parents of young children(Galinsky, 1986; Galinsky & Stein, 1990). Worrying aboutalternate care arrangements may influence perceived work-familyrole strain.A single item was used to measure difficulty findingalternate child care. Using a 3-point Likert-type response scale('1' not difficult, '2' somewhat difficult, '3' very difficult),respondents were asked to answer the question, "How difficultwould it be for you to find alternate child care arrangements36when your regular arrangements break down?"Satisfaction with child care. A single item was used tomeasure the respondent's satisfaction with her current childcare arrangements. Respondents were asked to answer thequestion, "How satisfied are you with your current child carearrangements?" A 4-point Likert-type response scale ('1' verydissatisfied, '4' very satisfied) was used.This variable was included as a control because women whoare not satisfied with their child care arrangement may feelguilty about leaving their child at the daycare center and mayworry about them while they are at work. These feelings mayinfluence work-family role strain and may also override supportoffered by the workplace.Sociodemographic InformationRespondents were also asked to provide information aboutpersonal, spouse/partner and family characteristics. Personalcharacteristics included type of occupation position, age,marital status, education, total personal income, and ethnicity.An occupational prestige score was constructed for eachrespondent by coding the type of occupation position accordingto the 1980 four-digit Canadian Classification and Dictionary ofOccupations and reconciling these codes to the 1981Socioeconomic Index for Occupations in Canada (Blishen, Carroll,& Moore, 1987). Spouse/partner characteristics includedquestions about spouse/partner's employment status, type ofoccupation position, work hours and work schedule. Family37characteristics included number of children, ages of children,and total household income.38Chapter IVResultsResponse Rate Two hundred and forty questionnaires were taken by employedmothers from the 45 daycare centers involved in the recruitmentof subjects. Over a period of 2 months, 143 questionnaires werereturned by mail to the School of Family and NutritionalSciences (response rate = 60%). Of the 143 questionnairesreturned, 20 were not eligible (e.g., did not have an immediatesupervisor, did not have a preschool age child, were not workingoutside the home) and 7 were eligible but were excluded due toconsiderable missing data on the dependent and independentvariables. The sample used for analysis consisted of 116mothers.Characteristics of the Sample A summary of characteristics of the sample is presentedhere. For a detailed demographic profile see Appendix D, Table1. The ages of the participants ranged from 23 to 46 years withmost mothers in their 30's (72%). Forty-seven percent of thesample was married. The majority of women (67%) had only onechild. Most mothers did not identify with an ethnic group otherthan Canadian. The majority of women had personal incomes of$40,000 or less (80%) and reported household incomes of $60,000or less (62%). Almost all women (97%) had completed high school,15% had completed a university undergraduate degree and 15% hadcompleted a university post graduate degree. Eighty percent of39the sample worked in one of four occupation classifications (1)managerial, administrative and related (n = 35), (2) clericaland related (n = 35), (3) medicine and health (n = 13), and (4)teaching and related occupations (n = 10).The majority of women (96%) who were married or living incommon law had a spouse or partner who was employed. Thirty-three percent of the employed spouses/partners worked inmanagerial, administrative and related occupations, 14% workedin service occupations, and the remaining spouses/partners wereemployed in a range of occupations. The majority (85%) ofemployed spouses/partners worked full-time (30 or more hours perweek), with almost half working atypical shifts.Univariate Distributions Before testing the proposed hypotheses, univariatedistributions for the dependent, independent and controlvariables were examined.Work-Family Role StrainThe average score on this 19-item Likert type scale was2.94 (SD = .50), where the possible range was 1 to 5. There wasvariation in the distribution of scores around the mean. Thedistribution was not significantly skewed and approximated anormal distribution.Work DemandsWork hours. The average length of the work week was 35.90hours (SD = 12.42). The distribution was significantly skewedtoward higher values. This distribution may be due to the40preselected characteristics of the sample. Because of thelimited number of part-time spaces available, the majority ofwomen in this sample would have their preschoolers in full-timedaycare. As a result, these women are more likely to be workingfull-time.Work schedule. The majority of mothers did not work anirregular work schedule. Ninety percent of the sample reportedthat they did not work evening or night shifts. Five percentreported that they did work evening or night shifts and fivepercent failed to answer this question. Eighty-two percentreported that they did not work weekend days, fifteen percentworked weekend days, and three percent failed to respond to thisitem.Psychological work demands. The average score onpsychological work demands was 3.55 (SD = .75), where thepossible range was 1 to 5. The distribution of scores was notsignificantly skewed and approximated a normal distribution.Organizational CultureAvailable family-oriented benefits. The average number ofavailable benefits was 4.12 (SD = 2.26), where the possiblerange was 0 to 13 available benefits. The distribution of scoreson this summed index was significantly skewed to the right ofthe median, 4.00, with clustering around lower values. Since fewfamily-oriented benefits are available in Canadian workplaces(see Paris, 1989), it was expected that the scores for thisindex would be distributed in such a manner.41Work environment support. The average score for workenvironment support was 3.66 (SD = 1.06), where the range was 1to 5. This one item Likert-type scale was significantly skewedtoward higher values, however, variation was evident in thedistribution of scores. For example, 5 of the cases scored avalue of one, 9 cases scored a value of two, and 34 cases scoreda value of three. While skewness was significant, kurtosis wasnot significant.Use of Family-Oriented BenefitsRespondent's use of family-oriented benefits. The averagenumber of benefits that respondents had used in the past year orwere currently using was 1.52 (SD = 1.78), where the possiblerange was 0 to 13 benefits. The distribution of scores for thissummed index was significantly skewed to the right of themedian, 1.00, with clustering around the values of 0 and 1. Dueto the limited number of available benefits in Canadianworkplaces, combined with eligibility requirements withincompanies, it was expected that the scores would be distributedin such a manner. See Table 1 for frequencies of use for each ofthe 13 family-oriented benefits.Spouse's use of family-oriented benefits. Eighteenrespondents had a spouse who had used family-oriented benefitsin the past year or was currently using one of the 13 family-oriented benefits listed. The average number of benefitsrespondents said their spouses had or were currently using was2.44 (SD = 1.10), where the minimum was 1 and the maximum was42Table 1Frequencies of Use for Each of the 13 Family-Oriented BenefitsPercent UsingFamily-Oriented Benefit^ Each BenefitAlternate Work ArrangementsFlextime^ 31.0Part-time (prorated benefits)^ 11.2Part-time (no benefits)^ 9.5Compressed work week 8.6Job sharing^ 5.2Work at home 2.6LeavesLeave in lieu of overtime^ 23.3Short term leave (personal/family)^21.6Sick child days with pay^ 15.5Personal days with pay 12.1Extended leave (personal/family)^5.2MiscellaneousEmployee assistance programs^ 5.2Workshops/seminars (work-family) 0.94313. The scores fell between 1 and 4 benefits and thedistribution was not significantly skewed.Future use of family-oriented benefits. The average numberof benefits that respondents said they would use if they wereavailable was 4.80 (SD = 3.29), where the possible range was 0to 13 benefits. The distribution of scores was not significantlyskewed but kurtosis was significant. The distribution wasflatter than that of a normal distribution with almost equalnumbers of cases for each value. For example, 13 respondentswould not use any of the family-oriented benefits if they wereavailable, 9 respondents would use 1 benefit, 15 respondentsreported they would use 2 benefits, 9 respondents said theywould use 3 benefits, and 11 respondents said they would use 4benefits.The three most frequently reported benefits thatrespondents said they would use if they were available were sickchild days with pay (61.2%), personal days with pay (53.4%), andworkshops and seminars on balancing work and familyresponsibilities (49.1%).Supervisor SupportGeneral supervisor support. The average score for generalsupervisor support was 3.47 (SD = .82), where the possible rangewas 1 to 5. The distribution was not significantly skewed andapproximated a normal distribution.Supervisor sensitivity. The average score for supervisorsensitivity was 3.43 (SD = .71), where the possible range was 144to 5. Skewness was significant with clustering of scores to theright of the median, 3.5, and most of the extreme values to theleft of the median. Although the distribution was slightlyskewed, there was variation in the distribution of scores.Kurtosis was not significant.Supervisor flexibility. The average score for supervisorflexibility was 3.22 (SD = .85), where the range was 1 to 5. Thedistribution of scores was not significantly skewed andapproximated a normal distribution.Control VariablesOccupational role commitment. The average score foroccupational role commitment was 2.97 (SD = .79), where therange was 1 to 5. Skewness was not significant and thedistribution of scores approximated a normal distribution.Family role commitment. The average score for family rolecommitment was 4.27 (SD = .73), where the range was 1 to 5. Thedistribution of scores was significantly skewed with more scoresclustering around higher values. For example, 31 cases scored a5 on this scale. Kurtosis was also significant. Because thesample consisted of mothers with at least one preschool agechild, there was no reason to assume that scores would benormally distributed on this Likert-type 3-item scale.Outside help. Ninety-eight cases (84.5%) did not hireanyone from outside their household to help with home chores ona regular basis. Eighteen cases (15.5%) did hire outside help.Occupation length. The average number of months mothers had45been in their current employment position was 55.52 months (SD= 58.89). The distribution of scores was significantly skewedwith most cases clustering around lower values. For example, 76cases (66%) had been in their current employment position for 48months (4 years) or less. This distribution of scores may be dueto the fact that all women had at least one preschool age childand may have taken time off work after the child was born.Satisfaction with child care arrangements. The averagescore for child care satisfaction was 3.36 (SD = .86), where thepossible range was 1 to 4. Skewness was significant withclustering of scores around the median, 4.0, and most of theextreme values to the left of the median. Eighty-nine percent ofthe sample was satisfied or very satisfied with their presentchild care arrangements.Difficulty finding alternate child care. The average scorefor this 1-item Likert type scale was 2.36 (SD = .65), where therange was 1 to 3. The distribution was significantly skewedtoward higher values. For example, 45% of the sample thought itwould be somewhat difficult and 45% of the sample thought itwould be very difficult to find alternate child care shouldtheir regular child care arrangements break down. With onlythree response categories, it was not expected that thedistribution of scores would follow the normal curve.Preliminary Analyses Because a significant proportion of the sample was notmarried or living in common law (40%), a t-test was run to check46for differences on role strain between mothers who were marriedor living in common law and mothers who were not married orliving in a common law relationship. No significant differenceswere found between the two groups, t (96) = -.09, R = .93. As agroup, the employed mothers in this sample experienced moderatelevels of work-family role strain. As reported earlier, the meanscore on the 5-point role strain scale was 2.94, where the rangewas 1 to 5.Pearson's correlations were also run between thesociodemographic variables and the dependent variable to seewhether any of these variables needed to be included asadditional control variables in testing the hypotheses (seeTable 2). Only number of children was significantly related towork-family role strain. Because the strength of thisrelationship was weak, r = .197, n = 99, R = .03, and mostmothers had only one child (67%), number of children was notcontrolled in subsequent analyses.Hypothesis TestingMultiple regression was used to test Hypotheses 1 through4. R square was initially examined for the sets of variables assessing work demands and components of workplace support.Depending on whether or not R square was significant for eachset of variables, one of two procedures was followed. If Rsquare was significant, the Beta coefficients for each variablewithin the set were examined for significance. The significanceof R square change was also examined when the set of variables47Table 2Correlations Between Sociodemographic Variables and Work-FamilyRole Strainna Role StrainSociodemographic VariablesMother's ageNumber of children9999-.041.197*Age of child (1 child) 67 -.120Age of child (2 children)Youngest 26 -.044Oldest 30 -.106Personal income 98 .009Household income 94 .084Occupational prestige 99 .048Education (years) 98 -.006Education (level) 99 .008Spouse/partner employed 99 .011Spouse/partner work hours 57 .203Spouse/partner evening/night shift 60 .142Spouse/partner weekendshift 59 .090a Number varies due to missing data and/or inapplicability.* p <.05.48was entered into the regression equation after the six controlvariables (occupational role commitment, family role commitment,outside help, occupation length, child care satisfaction, anddifficulty finding alternate child care).If R square was not significant, separate regressionanalyses were run for each variable in the set, with thedependent variable. R square and Beta for each variable werethen examined for significance. Multiple regression, rather thanbivariate correlation analysis, was used so that the six controlvariables could be taken into account. For those variables thatwere significantly related to the dependent variable, thesignificance of R square change was examined when the variablewas entered into the regression equation after the six controlvariables.Correlation analysis was used to test Hypotheses 5 and 8.Interaction effects, proposed in Hypotheses 6 and 7, were testedwith analysis of variance.Hypothesis 1: The greater the work demands, the greater thework-family role strain.Intercorrelations between the four work demand variablesindicated that except for the two shift variables, the workdemand variables were not significantly related to one another(see Table 3). The correlation between evening/night shifts andweekend shifts was too low to combine these variables and createa unidimensional measure. The four work demand variables weretherefore treated as separate independent measures.Table 3Intercorrelations Between Work Demand Variables2 3 4Work Demands1. Work hours2. Evening/night shifts3. Weekend shifts4. Psychological work demands.01 .05.43*-.02-.05-.05*R < .001.4950Average weekly work hours, evening/night shifts, weekendshifts and psychological work demands were entered as a blockinto the regression equation with the dependent variable, work-family role strain. The set of work demand variables was notrelated to work-family role strain, F (4, 86) = 2.06, R2 = .09,R = .09.Multiple regression was then used to examine therelationship between each work demand variable and work-familyrole strain. Thus, four separate regression equations, one foreach work demand variable, were examined. Of the four workdemand variables, only the psychological work demand variablemade a significant contribution to the variance in work-familyrole strain, F (6, 87) = 7.68, R2 = .075, p = .007, and had asignificant coefficient, Beta = .274, p = .007. The resultsshowed that the greater the psychological work demands, thegreater the work-family role strain.When the psychological work demand variable was enteredinto the multiple regression equation after the six controlvariables, it no longer explained any variance in work-familyrole strain (R2 change = .033, p = .06).Although the results do not provide support for thehypothesized relationship between total work demands and work-family role strain, they do provide some support for therelationship between psychological work demands and role strain.Hypothesis 2: The more supportive the organizationalculture, the less the work-family role strain.51Available family-oriented benefits and work environmentsupport were significantly correlated, r = .25, n = 116, p =.003. Because the strength of the relationship was weak,however, available benefits and work environment support weretreated as independent measures of organizational support.Available family-oriented benefits and work environmentsupport were entered as a block into the multiple regressionequation with the dependent variable, work-family role strain.This set of variables assessing organizational culture accountedfor a significant proportion of variance in work-family rolestrain, F (2, 96) = 3.23, R2 = .063, p = .044. Of theorganizational culture variables, only work environment supporthad a significant coefficient, Beta = -.228, p = .03. Thus, thegreater the work environment support, the less the work-familyrole strain.As indicated by R square change, the set of organizationalculture variables continued to make a significant contributionto work-family role strain when entered into the multipleregression equation after the six control variables (R2 change= .07, p = .02).The results provide support for the hypothesizedrelationship between organizational culture and work-family rolestrain.Hypotheses 3: The greater the number of family-orientedbenefits used, the less the work-family role strain.The correlation between respondent's use of benefits and52spouse's use of benefits was not significant, r = -.09, n =18, p = .36. Thus, the two variables were considered independentmeasures of total use of benefits.Both the respondent's use of family-oriented benefits andspouse's use of family-oriented benefits were entered as a blockinto the regression equation with the dependent variable, work-family role strain, to assess the contribution of total use offamily-oriented benefits. This set of variables did not accountfor any variance in work-family role strain, F (2, 13) = 1.64,R2 = .201, p = .23.Because this set of variables assessing use of benefits wasnot significantly related to work-family role strain,respondent's use and spouse's use of benefits were examinedseparately with work-family role strain. Thus, two multipleregression equations were examined. Neither respondent's use offamily-oriented benefits nor spouse's use of family-orientedbenefits were related to work-family role strain.T-tests were also run to investigate possible differenceson work-family role strain between (1) women who were not usingany family-oriented benefits and women who were using one ormore family-oriented benefits; and (2) women who were using onefamily-oriented benefit with women who were using more than onefamily-oriented benefit. No significant differences on work-family role strain were found between women who were not usingany benefits and women using one or more benefits, t (97) = .60,p = .55; or between women using one benefit and women using more53than one benefit, t (57) = -.93, p = .36.The results do not provide support for the hypothesis thatthe more family-oriented benefits used, the less the work-familyrole strain.Hypothesis 4: The greater the perceived supervisor supportfor combining employment with family responsibilities, theless the work-family role strain.Supervisor sensitivity, supervisor flexibility and thegeneral measure of supervisor support were entered as a blockinto the multiple regression equation with the dependentvariable in order to assess the contribution of this set ofsupervisor support variables to the explanation of variance inwork-family role strain. Although moderate to highintercorrelations were found between the three measures ofperceived supervisor support (see Table 4), tolerance levelswere above .01 when the three variables were entered as a blockin the multiple regression equation.As a set, general supervisor support, supervisorsensitivity and supervisor flexibility did not account for anyvariance in employed mother's work-family role strain, F (3, 76)= 2.24, R2 = .081, p = .09.Because the set of total supervisor support variables wasnot related to work-family role strain, each supervisor supportvariable was examined separately using multiple regressionanalysis. Thus, three separate multiple regression equationswere examined. Of the three supervisor support variables, only54Table 4Intercorrelations Between Supervisor Support Variables2^3Supervisor Support1. General supervisor support^.87*^.45*2. Supervisor sensitivity .52*3. Supervisor flexibility*p < .001.55supervisor flexibility made a significant contribution to theexplained variance in work-family role strain, F (1, 78) = 6.87,R2 = .08, p = .01, and had a significant Beta coefficient, Beta= -.285, p = .01. The results showed that the greater thesupervisor flexibility, the less the work-family role strain.As indicated by R square change, supervisor flexibilitycontinued to make a significant contribution to work-family rolestrain when entered into the multiple regression equation afterthe six control variables (R2 change = .067, p = .013).While these results do not provide support for thehypothesized relationship between overall supervisor support andwork-family role strain, the findings show that supervisorflexibility, one dimension of supervisor support, is a relevantpredictor of work-family role strain.Hypothesis 5: The more family-oriented benefits available,the more family-oriented benefits used.The number of available benefits was significantly relatedto the number of family-oriented benefits used by therespondent, r = .40, n = 116, p = .001. The greater the numberof available benefits, the greater the number of benefits used.The scatterplot for available benefits and use of benefitsshowed that clustering occurred around lower values for bothvariables. Most women had fewer than six available benefits andwere using fewer than three of the available benefits. Thescatterplot also indicated that as the number of availablebenefits increased, the number of benefits used typically did56not surpass the use of four benefits.The correlation between the number of benefits availableand the percentage of benefits used supports the patternsidentified by the scatterplot. As the number of availablebenefits increased, the percentage of benefits used decreased,r = -.316, n = 71, p = .004. Respondents were less likely touse all of the benefits available to them when the number ofavailable benefits was large.The results provide partial support for the hypothesizedrelationship between available benefits and use of benefits.Hypothesis 6a: When work demands are high, individuals withhigh perceived supervisor support will have lower levels ofwork-family role strain than individuals with low perceivedsupervisor support.Hypothesis 6b: When work demands are low, individuals withlow perceived supervisor support will have higher levels ofwork-family role strain than individuals with highperceived supervisor support.Prior to testing these hypotheses with analysis ofvariance, a work demand index was created and a total supervisorsupport scale was constructed.Due to the lack of significant intercorrelations betweenlength of the work week, evening/night shifts, weekend shiftsand psychological work demands (see Table 3), an index, ratherthan a scale, was created for total work demands. Thestandardized scores for each of the four work demand variables57were computed, summed and averaged to create the work demandindex.General supervisor support, supervisor sensitivity andsupervisor flexibility were summed and averaged to create atotal supervisor support scale. The moderate to highintercorrelations between these three measures of supervisorsupport supported the construction of the total supervisorsupport scale (see Table 4). The reliability of this scale, asmeasured by Chronbach's alpha, was .82. Two-way interactionsbetween the work demand index and the total supervisor supportscale were then examined.The two-way interaction for the work demand index and thetotal supervisor support scale was not significant, F (3, 71) =.61, p = .44. These two factors did not jointly affect work-family role strain.Because total supervisor support may moderate therelationship between different aspects of work demands and work-family role strain, analysis of variance was used to examinepossible interaction effects between the total supervisorsupport scale and each of the four work demand variables makingup the work demand index.Due to the distribution of responses for work hours and thescheduling of work hours, cell sizes were extremely unequal forthese variables. As a result, interaction effects could not beexamined. Cell sizes for psychological work demands and total58supervisor support were relatively equal and analysis ofvariance was conducted. The two-way interaction forpsychological work demands and total supervisor support was notsignificant, F (3, 75) = .97, p = .33. Thus, psychological workdemands and total supervisor support did not jointly affectwork-family role strain.Because the three supervisor support variables comprisingthe supervisor support scale were designed to measure differentdimensions of supervisor support, analysis of variance was alsoused to investigate interaction effects of psychological workdemands and each of the three supervisor support measures onwork-family role strain. No significant interaction effects werefound for psychological work demands and either generalsupervisor support, F (3, 93) = 2.14, p = .15; supervisorsensitivity, F (3, 93) = .607, p = .44; or supervisorflexibility F (3, 75) = .019, p = .89.The results of these analyses of variance do not providesupport for the prediction that supervisor support, ordimensions of supervisor support, moderate the relationshipbetween work demands and work-family role strain.Hypothesis 7a: When the use of family-oriented benefits ishigh, individuals with high perceived supervisor supportwill have lower levels of work-family role strain thanindividuals with low perceived supervisor support.59Hypothesis 7b: When the use of family-oriented benefits islow, individuals with low perceived supervisor support willhave higher levels of work-family role strain thanindividuals with high perceived supervisor support.Prior to testing hypotheses 7a and 7b with analysis ofvariance, a use of benefits index was constructed by summing andaveraging respondent's use of benefits and spouse's use ofbenefits. Due to the limited number of respondents with spouseswho were using one or more family-oriented benefits (n = 16),the sample size was too small to use the constructed index. Asa result, only respondent's use of family-oriented benefits wasincluded in the analysis. The total supervisor support scale,described in the results section for hypothesis 6a and 6b, wasalso used to test these hypotheses.Two-way interactions for the respondent's use of benefitsand total supervisor support on work-family role strain wereexamined. The two-way interaction was not significant, F (3, 76)= .05, p = .82.Because the three supervisor support variables comprisingthe supervisor support scale were designed to measure differentdimensions of supervisor support, analysis of variance was alsoused to investigate interaction effects between respondent's useof benefits and each of the three supervisor support measures onwork-family role strain. No significant interaction effects werefound for respondent's use of benefits and either generalsupervisor support, F (3, 95) = .02, p = .88; supervisor60sensitivity, F (3, 95) = .03, p = .87; or supervisorflexibility, F (3, 76) = 3.55, p = .06.The results of these analyses of variance do not providesupport for the prediction that supervisor support, ordimensions of supervisor support, moderate the relationshipbetween respondent's use of benefits and work-family rolestrain.Hypothesis 8: The relationship between available family-oriented benefits and use of family-oriented benefits willbe stronger when perceived supervisor support is high thanwhen perceived supervisor support is low.Using a median split, cases were divided into two groups:(1) low supervisor support, and (2) high supervisor support.Correlation coefficients between the number of availablebenefits and the number of benefits used were then computed forthe two groups.The relationship between available benefits and use ofbenefits was significant for both supervisor support groups.Contrary to what was predicted, the strength of the relationshipwas greater for women with low supervisor support, r = .43, n= 44, p = .002, than for women with high supervisor support, r= .30, n = 43, p = .028.Post Hoc Analysis While testing the proposed hypotheses, relationships notspecified in these hypotheses became evident. Three of the sixcontrol variables and future use of family-oriented benefits61were associated with work-family role strain. Also, the use offamily-oriented benefits was significantly related to dimensionsof supervisor support.The results of the one-tailed correlation analysis showedthat of the six control variables, occupational role commitment,family role commitment and difficulty in finding alternate childcare arrangements were significantly related to work-family rolestrain (see Table 5). The strength of the relationships was weakbut all were statistically significant. The more committed theemployed mother is to her work and family roles and the greaterher difficulty in finding alternate child care when regular careis unavailable, the greater the work-family role strain.Future use of family-oriented benefits made a significantcontribution to the variance in work-family role strain, F (1,97) = 10.88, R2 = .10, p = .001, and had a significantcoefficient, Beta = .317, p = .001, when it was the onlyvariable in the multiple regression equation with role strain.The results showed that the greater the number of family-oriented benefits that would be used by the respondent if theywere available, the greater the work-family role strain. Asindicated by R square change, future use of benefits continuedto make a significant contribution to role strain when enteredafter the six control variables, R2 change = .039, p = .04.A median split was done for future use of benefits and a t-test was run to check for significant differences on role strainbetween mothers with low future use and mothers with high future62Table 5Correlations Between Control Variables and Work-Family RoleStrainControl Variables na Role StrainOccupational role commitment 98 .17*Family role commitment 99 .23**Occupation length 98 -.07Use of outside help 99 -.06Satisfaction with child care 99 -.07Difficulty finding alternate care 98 .26**a Number varies due to missing data.*p < .05. **p < .01.63use of family-oriented benefits. Significant differences werefound between the two groups, t (97) = -2.99, p = .004. Motherswith high future use reported greater work-family role strain (M= 3.13) than mothers with low future use (M = 2.83).The results of the one-tailed correlation analysis showedthat available family-oriented benefits was significantlyrelated to general supervisor support and supervisor sensitivity(see Table 6). In addition, the results indicated that futureuse of benefits was associated with general supervisor support,supervisor sensitivity, and supervisor flexibility, however,current use of family-oriented benefits was only significantlyrelated to supervisor flexibility (see Table 6).64Table 6Intercorrelations Between Family-Oriented Benefits and Supervisor Support Variables Supervisor SupportFamily-Oriented BenefitsGeneral Sensitivity FlexibilityAvailable .27** .33*** .12Current use .08 .15 .30**Future use -.20* -.20* -.22***p < .05. **p < .01. ***R <.001.65Chapter VDiscussionThis study investigated the impact of work demands andworkplace support on work-family role strain of employed motherswith preschool age children. Availability and use of family-oriented benefits was also assessed. In addition, moderatingeffects of supervisor support were examined.Role strain theory provided the conceptual framework forexamining work-family role strain, however, current work-familyliterature provided the basis for exploring institutional levelcoping mechanisms for reducing strain associated with combiningwork and family roles. The results of this study, therefore, arediscussed in terms of their contribution to existing work-familyliterature, particularly the literature on workplace support forcombining work and family roles.Work-Family Role StrainPrevious research shows that when compared to employedfathers and employed mothers of older children, employed mothersof preschool age children are more likely to report spilloverbetween work and family (Crouter, 1984) and greater work-familyrole strain (Greenberger et al., 1989; Kelly & Voydanoff, 1985).Although none of these studies investigated the relationshipbetween type of child care used and work-family outcomes, it wasassumed that employed mothers using group daycare mightexperience especially high levels of work-family role strain dueto the lack of flexibility in the operating hours of full-time66daycare centers and the need to make alternate child carearrangements when the child is ill. The mothers in this sample,however, experienced a moderate level of work-family rolestrain. Due to the lack of current research investigating therelationship between child care arrangements and work-familyoutcomes, it is not known how mothers using other carearrangements might score on work-family role strain.Work Demands Contrary to what was predicted in Hypothesis 1, employedmothers who perceive their paid work as both psychologically andstructurally demanding do not experience greater strain betweentheir work and family roles. When structural and psychologicalwork demand variables were examined separately, however, theresults indicate that work-family role strain is not influencedby either the number of hours worked per week or the schedulingof these hours, but is related to the psychological demands ofthe job.Although the number of paid work hours and work-family rolestrain are not related for this sample, this finding is notconsistent with previous research that reports higher work-family role strain with a longer work week (Campbell & Moen,1992; Keith & Schafer, 1980; Voydanoff, 1988; Voydanoff & Kelly,1984). In this study, the lack of a significant relationshipbetween work hours and work-family role strain may be due tolimited variation in the number of hours the mothers wereemployed per week. The majority of mothers (75%) worked between6730 and 40 hours with only 5% working more than 40 hours, and 18%working fewer than 30 hours per week.While working non-day and irregular shifts has beenassociated with higher levels of work-family conflict (Staines& Pleck, 1983) and greater schedule incompatibility between workand family life (Pleck et al., 1980), the results of this studyindicate that working evening/night shifts or weekend days isnot related to work-family role strain. This finding may be dueto the fact that only 5% of the mothers worked evening or nightshifts and 15% worked weekend days.Because of the rigid operating hours of group daycarecenters, it was believed that mothers with preschoolers in groupdaycare might be more likely to work full-time hours and typicalshifts. The viability of this assumption was questionable due tothe lack of research on work characteristics of group daycareusers. Because previous research has found work hours and thescheduling of work hours to be related to work-family outcomes(Campbell & Moen, 1992; Keith & Schafer, 1980; Pleck et al.,1980; Staines & Pleck, 1983; Voydanoff, 1988; Voydanoff & Kelly,1984), excluding these variables from this study, withoutempirical support, would not have been justified.Psychological work demands was significantly related towork-family role strain. The more psychologically demanding thepaid work, the greater the perceived work-family role strain.Mothers who feel they have too much work to do for their job andnot enough time to get all their work completed are more likely68to perceive that they cannot adequately fulfill their work andfamily demands. This finding is supported by work-familyresearch which has also found that psychological work demandssuch as heavy workloads and pressure for output are positivelyrelated to work-family role conflict (Voydanoff, 1988) and work-family role strain (Katz & Piotrkowski, 1983). Whilepsychological work demands may create job stresses thatspillover into family life (Voydanoff, 1987), this findingprovides further support for the belief that the psychologicaldemands of one's paid employment influence employed mothers'perceived ability to adequately fulfill the demands of boththeir work and family roles.Although psychological work demands was significantlyrelated to work-family role strain, it is important to addressthe fact that this relationship was no longer significant whencommitment to work and family roles, satisfaction with childcare, difficulty finding alternate childcare, and the length oftime in current employment were controlled. This findingsuggests that other factors need to be considered wheninvestigating the relationship between psychological workdemands and work-family role strain.Workplace Support While it was hypothesized that work demands would increasethe level of strain between work and family roles, it was alsopredicted that workplace support would help to reduce work-family role strain for employed mothers. Three aspects of69workplace support were assessed: organizational culture, use offamily-oriented benefits, and supervisor support.Organizational culture. The results of this study providesupport for the hypothesized relationship between organizationalculture and work-family role strain. The more supportive theorganizational culture of employees with familyresponsibilities, the less the strain between work and familyroles. This finding supports the view that having a "family-friendly" organizational culture or philosophy is an integralpart of how business organizations can help employees balancework and family responsibilities (Bowen, 1988; Galinsky & Stein,1990; McCroskey, 1982).Within the set of organizational culture variables, onlywork environment support emerged as a relevant predictor ofwork-family role strain. The more supportive mothers perceivetheir work environment to be of employees with work-familydifficulties, the less difficult they feel it is to adequatelyfulfill their work and family demands. This finding providessupport for the view that a work environment that is supportiveof employees' work-family difficulties may help to improveemployed parents' ability to balance work and family life andreduce associated strain (Bowen, 1988; Galinsky & Stein, 1990;McCroskey, 1982).The availability of family-oriented benefits was notsignificantly related to work-family role strain. This may bedue to the fact that while almost all women in the sample (97%)70had one or more benefits available to them, simply having themavailable may not have been enough to lower their level of work-family role strain.Although available benefits is not significantly related towork-family role strain, the number of available family-orientedbenefits is positively related to work environment support,general supervisor support, and supervisor sensitivity. Motherswith access to family-oriented benefits are more likely toperceive their work environment as supportive, and theirimmediate supervisors as instrumentally and emotionallysupportive as well as sensitive to their work-family needs.While current work-family research has not examined therelationship between available benefits and such measures ofsupport, this finding suggests that perceptions of workenvironment support and supervisor support may be influenced bytangible benefits offered by the workplace.Use of family-oriented benefits. It was predicted inhypothesis 3 that the more family-oriented benefits used, theless the work-family role strain. When respondent's use andspouse's use of family-oriented benefits were entered as a groupin the multiple regression equation, total use of benefits wasnot related to work-family role strain. When respondent's use offamily-oriented benefits and spouse's use of family-orientedbenefits were examined in separate regression analyses, neithervariable was related to work-family role strain. The findingsalso showed that no significant differences in role strain were71evident between (1) mothers with no use of benefits and motherswith use of benefits; and (2) mothers with some use of benefitsand mothers with high use.Greenberger et al. (1989) found that while single women whouse more family-oriented benefits report reduced role strain,married women report increased strain. It is surprising thatalthough 71 mothers (61%) in this sample report that they haveused in the past year or are currently using one or more family-oriented benefits, the use of family-oriented benefits is notassociated with work-family role strain. One possibleexplanation for this finding is that the most frequently usedfamily-oriented benefits were leave-related benefits such assick child days, personal days, short term leave for familyreasons, and leave in lieu of overtime. While such benefitsprovide temporary solutions to specific work-family problems,they are not long term solutions and therefore, may not beuseful in reducing work-family role strain. In addition, leave-related benefits are more likely to have been used sometime inthe past year rather than at the time of assessing currentlevels of work-family role strain.The lack of a significant relationship between spouse's useof family-oriented benefits and work-family role strain may bebecause only a few women had spouses who were using family-oriented benefits. Of the 55 women who had spouses, only 18 hada spouse who had used in the past year or was currently usingone of the 13 family-oriented benefits.72Although predictions were not made about the relationshipbetween work-family role strain and the desire to use family-oriented benefits if they were available, the regression resultsindicate that the greater the number of benefits that would beused by the respondent if they were available, the higher thework-family role strain. Significant differences were also foundbetween women who would use five (the median) or more family-oriented benefits and women who would use less than fivebenefits. Those in the high future use group reported higherwork-family role strain than those women in the low future usegroup.The findings for future benefit use may be an indication ofthe salience of family-oriented benefits to employed mothers andan actual desire to have a greater number of benefits availableto them. Another possible explanation for these findings is thatthis variable may be another way of assessing work-family rolestrain. While 11% of the sample would not use any of the family-oriented benefits if they were available, 47% would use one tofive benefits and 41% would use six or more. While it would bepossible to use leave-related benefits, employee assistanceprograms and workshops/seminars "simultaneously", the use ofseveral different types of scheduling benefits at one time isless feasible. The desire to use numerous benefits seems to bemore an indictor or measure of the degree of difficultyperceived in combining work and family life than a predictor ofwork-family role strain.73Supervisor support. Although research is limited,supervisor support has been identified as a central component ofworkplace support and influential in work-family outcomes(Fernandez, 1985; Galinsky & Stein, 1990; Greenglass et al.,1989). Hypothesis 4 predicted that the greater the supervisorsupport, the less the work-family role strain.The set of supervisor support variables (supervisorflexibility, general supervisor support, and supervisorsensitivity) was not useful in predicting work-family rolestrain. When the three supervisor support variables wereexamined separately, rather than as a set, only one dimension ofsupervisor support (supervisor flexibility) was significantlyrelated to work-family role strain.A significant negative relationship was found betweensupervisor flexibility and work-family role strain. Employedmothers who perceive their supervisors as flexible may feel theycan more adequately meet their work and family demands becausethey are able to alter their work demands in order to meet theirfamily demands (e.g., able to come in late or leave work early),or they are able to let family demands overlap with work demands(e.g., allowed to receive phone calls from home at work, canbring the child to work). Such perceived flexibility is likelyto reduce feelings of anxiety when coordinating work and familyresponsibilities.This finding is not consistent with Greenberger et al.'s(1989) research even though the same measure of supervisor74flexibility was used. This may be due to the type of child carearrangement used. Participants for Greenberger et al.'s (1989)study were recruited through preschools, however the type ofcare arrangement used when the child was not in preschool wasnot stated. The subjects for the current study were all usinglicensed group daycare. Due to the limited number of part-timespaces available, children were likely to be in daycare for thefull day. Supervisor flexibility may be more instrumental towork-family role strain for mothers using this type of carearrangement due to the lack of flexibility at the child carecenter.Consistent with Greenberger et al.'s (1989) findings,general supervisor support was not significantly related towork-family role strain. The lack of a significant finding forgeneral supervisor support is not surprising since this measurewas designed to assess emotional and instrumental supervisorsupport in general rather than supervisor responsiveness orsensitivity toward employees with family responsibilities.Although it has been suggested that fulfilling work andfamily demands will be perceived as less difficult for employeeswho perceive their supervisors as sensitive to their familyresponsibilities, the results do not support this view. Thisfinding suggests that regardless of the perceived level ofsupervisor sensitivity, if the supervisor is not able to providesome assistance when work-family difficulties arise, work-familyrole strain may not be reduced.75The results indicate that supervisor flexibility but notsupervisor sensitivity is related to work-family role strain.These findings suggest that supervisor practices may play a moreinstrumental role than supervisor attitudes in alleviatingemployees' perceived strain between work and family roles.Available Benefits and Use of Benefits It was predicted in Hypothesis 5 that the more family-oriented benefits available, the more family-oriented benefitsused. Partial support was provided for this hypothesis. Whilethe number of benefits used, increases with the number ofbenefits available, the plot of this relationship revealsseveral patterns. First, the relationship between availabilityand use of family-oriented benefits is strongest when five orfewer benefits are available and when fewer than three benefitsare used. Second, there appears to be a limit on the number ofbenefits that are used. As the number of available benefitsincreases, the number of benefits used generally does notsurpass the use of four benefits.The relationship between available benefits and thepercentage of benefits used indicates a similar pattern. Thepercentage of benefits used decreases as the number of availablebenefits increases. Thus, individuals are less likely to use allof the benefits available to them when a range of benefits isavailable. These findings suggest that from the availablebenefits, employed mothers choose benefits that are useful tothem and that meet their specific needs.76It has been suggested that the use of available benefitsmay be influenced by the supportiveness of supervisors ormanagers. For example, use of family-oriented benefits may beundermined by supervisory practices if employees feel that theirsupervisors or managers are uncooperative or pressure them notto use available benefits (Hughes & Galinsky, 1988; Kamerman &Kahn, 1987; Raabe & Gessner, 1988). Based on this literature, itwas predicted in Hypothesis 8 that the strength of therelationship between available benefits and use of benefitswould be greater for women who perceive high overall supervisorsupport than for women who perceive low overall supervisorsupport.Contrary to what was expected, the results of the selectivecorrelation analysis show that women who perceive low overallsupervisor support are more likely to use available benefitsthan women who perceive high overall supervisor support. Thisfinding suggests that under nonsupportive conditions, women aremore likely to use available family-oriented benefits. Thus,family-oriented benefits may be relied on to alleviate work-family difficulties when awareness or understanding ofemployees' work-family responsibilities is low and supervisorflexibility for family emergencies is minimal.The lack of support for the proposed hypothesis may be dueto the type of supervisor support that was assessed. Themeasures of support comprising the overall supervisor supportvariable were designed to assess emotional and instrumental77support as well as flexibility and sensitivity for employeescombining work and family roles. Measures of supervisor supportthat are more specific to the use of family-oriented benefitsmay be more applicable when examining the relationship betweensupervisory practices and the use of available family-orientedbenefits.Moderating Effects of Supervisor SupportIn addition to investigating direct effects of supervisorsupport on work-family role strain, this study also sought toexamine interaction effects between supervisor support and twowork environment variables in predicting work-family rolestrain.It was predicted in Hypotheses 6a and 6b that the impact ofwork demands on work-family role strain would be greater formothers who perceive low supervisor support than for thosemothers who perceive high supervisor support. The resultsindicate that total supervisor support and the work demand indexdo not jointly affect work-family role strain.Although structural work demands could not be examined forinteraction effects with supervisor support, due to the lack ofvariation in these variables, the joint effects of psychologicalwork demands and total supervisor support, as well aspsychological work demands and dimensions of supervisor supporton work-family role strain were investigated. None of theinteractions however, were significant.The lack of significant interaction effects for total78supervisor support and psychological work demands, and for eachdimension of supervisor support and psychological work demandsindicates that regardless of supervisor support, employedmothers who perceive greater psychological work demands alsoperceive greater work-family role strain. Despite the absence ofinteraction effects for these measures of supervisor support andpsychological work demands, other dimensions or types ofsupervisor support not included in this study may moderate therelationship between psychological work demands and work-familyrole strain. Work related supervisor support for example, may bemore influential than general supervisor support or support forcombining work and family roles.It has been suggested that the use of family-orientedbenefits may be counteracted by negative attitudes andunsupportive supervisors or managers (Hughes & Galinsky, 1988;Kamerman & Kahn, 1987). Based on this literature, it waspredicted in hypotheses 7a and 7b that the impact of use offamily-oriented benefits on work-family role strain would begreater for mothers who perceive high supervisor support thanfor mothers who perceive low supervisor support.The results of the analyses of variance used to test thesehypotheses, do not reveal significant interaction effects foreither total supervisor support or each dimension of supervisorsupport and respondent's use of family-oriented benefits. Due tothe lack of a significant relationship between use of family-oriented benefits and work-family role strain it was expected79that the interaction effects would not be significant.As mentioned earlier, the type of benefit used mayinfluence the level of work-family role strain. Hence,supervisor support may not moderate the relationship betweenthose benefits used by this sample and work-family role strain,but may influence the relationship between the use of otherfamily responsive benefits such as alternate work arrangements,employee assistance programs, workshops/seminars and rolestrain. Due to the limited number of respondents using thesebenefits in this study, such relationships could not beinvestigated. Future research is needed which examines themoderating effects of supervisor support for a larger group ofindividuals who are using these benefits.Supervisor Support and Family-Oriented Benefits Although interaction effects are not evident for dimensionsof supervisor support and current use of benefits in predictingwork-family role strain, significant relationships betweendimensions of supervisor support and current and future use ofbenefits are evident.Greater usage of family-oriented benefits is associatedwith the perception of more flexible supervisory practices whenfamily emergencies arise. There are several possibleexplanations for this finding. First, employees who perceivetheir supervisors as flexible may be more willing to use family-oriented benefits because they may not feel they will bepenalized for doing so. Second, employees who are using family-80oriented benefits may already perceive their supervisors asflexible because their supervisors have enabled them to use suchbenefits. Thus, these employees may be more likely to perceivethat their supervisors will be accommodating and flexible inother work-family situations. Finally, the nature of the job mayinfluence whether supervisors can be flexible and whetherfamily-oriented benefits can be used. Employees may not perceivetheir supervisors as flexible when family emergencies arise andmay not be able to use family-oriented benefits because the typeof job will not accommodate such flexibility regardless of howsensitive the supervisor is perceived. This suggests that use offamily-oriented benefits and supervisor flexibility need to beinvestigated in conjunction with the type of employmentposition.Employed mothers who perceive their work environment assupportive and their supervisors as generally supportive,sensitive and flexible are less likely to report higher numbersof family-oriented benefits that they would use if they wereavailable. This finding provides support for McCroskey's (1982)suggestion that employers who cannot afford to implement family-oriented benefits can provide their employees with an intangiblebenefit by creating an atmosphere that recognizes employees'family responsibilities.Control Variables and Work-Family Role StrainAlthough investigating the relationship between controlvariables and work-family role strain was not an objective of81this study, three of the six control variables emerged asrelevant predictors of work-family role strain.The results show that greater occupational role commitmentis related to higher work-family role strain. This finding isconsistent with Piotrkowski et al.'s (1987) view that highinvolvement in the work role can interfere with familyinvolvement, and may increase the potential for competitionbetween work and family for the individual's emotionalinvolvement. Ladewig (1990) also suggests that despite thegeneral acceptance and respectability of female labour forceparticipation, societal expectations prevail that a woman'sprimary role is homemaker and childrearer. Thus, women who arehighly committed to their paid work role may experience internalconflict as well as negative social sanctions that may not beexperienced to such an extent by those who are less committed totheir paid work role.The results of this study also indicate that the greaterthe family role commitment the higher the perceived work-familyrole strain. Women who are highly committed to their family rolemay experience greater work-family role strain because bycommitting so much time and energy to their family role they mayhave to struggle to meet the demanding claims of their workrole. The family role may interfere or overlap with the workrole to such an extent that they cannot adequately meet theirwork demands. Employed mothers who are highly committed to theirfamily role may also feel guilty about leaving their young child82in someone else's care which contributes to their level of rolestrain.Greater perceived difficulty in finding alternate childcare is also related to higher levels of work-family rolestrain. Employed mothers' ability to work outside the home isdependent upon care for their children while they are working.If the child cannot go to daycare (e.g., due to illness), it isessential that the mother have alternate care unless she is ableto stay home from work. Finding alternate child care may be adaily worry for mothers who are completely dependent on theirregular child care arrangement and have no other form of care torely on if necessary.While previous research has addressed sociodemographiccharacteristics such as age, income, and occupational prestigeas control variables (cf. Greenberger et al., 1989), the resultsof this study suggest that future research on work-familyoutcomes must also consider other possible control variables(e.g., occupational role commitment, family role commitment, anddifficulty finding alternate child care) that may influence thefindings.LimitationsWhen discussing the findings of this study, severallimitations due to sampling, research design, and measurementneed to be addressed. One limitation due to sampling is self-selection bias. Because respondents volunteered to participatein this study, the characteristics of these participants may83differ from women who either did not take or returnquestionnaires. Participants may have had a high degree ofinterest in parenting issues, workplace policies and practices,and/or ways of managing work-family difficulties. Anothersampling limitation is that the findings of this study cannot begeneralized to mothers of preschoolers in other childcarearrangements, mothers with older children, or mothers who do nothave an immediate supervisor, manager or boss at work. Inaddition, limiting the sample to mothers perpetuates the viewthat parenting is a woman's issue and difficulties associatedwith combining work and family roles are women's problems.A limitation of the research design is that it is a cross-sectional study. Thus, it is questionable whether differences inworkplace support predict differences in work-family role strainor whether some women choose to work in a supportive workenvironment and go into employment that is inherently moreflexible because of their family demands. Longitudinal researchis necessary to adequately address this issue.A major issue for research investigating the impact of theuse of family-oriented benefits on well-being is the time atwhich the assessment of use is made. Greenberger et al. (1989)state that the point of reference taken in their study limitedtheir findings because only family responsive benefits thatrespondents had used were assessed. For the current research,respondents were asked to identify family-oriented benefits thatthey were currently using or had used within the past year. This84point of reference was chosen in an attempt to obtaininformation about the use of leave-related benefits,workshops/seminars on work-family issues, and employeeassistance programs that may not have been used at the time ofthe assessment but rather sometime in the near past. While dataon the use of alternate work arrangements can be collected atthe time of the assessment, collecting current data on the useof other benefits is problematic. Future research needs toestablish a more accurate way of assessing the use of suchbenefits in order to provide a better understanding of theirimpact on work-family outcomes.ConclusionThe limitations discussed above should not obscure thecontributions of this research. Partial support was provided forthe proposed relationship between work demands and strainbetween work and family roles. Structural work demands were notrelated to work-family role strain for this sample, however,psychological demands of the job were predictive of work-familyrole strain, in the absence of control variables. While previouswork-family research has generally focused on structural demandsof paid employment, these results provide support for assessingsubjective, as well as objective, dimensions of work demandswhen investigating work-family outcomes.Partial support was also found for hypothesizedrelationships between work-family role strain and components ofworkplace support. Although the findings cannot be generalized85to other populations, the results showed that work environmentsupport and supervisor flexibility were predictive of work-family role strain perceived by employed mothers withpreschoolers in group daycare. Greater perceived supervisorflexibility and higher levels of perceived work environmentsupport were related to lower levels of strain between work andfamily roles.In addition, partial support was found for the hypothesizedrelationship between available benefits and the use of family-oriented benefits provided by the employer. While the number ofbenefits used increased with the number of benefits available,the percentage of benefits used decreased. The women in thissample were less likely to use all of the benefits available tothem when the number of available benefits was high. Thus,employees appear to be selective in their use of availablebenefits. Contrary to what was expected, women with low overallsupervisor support were more likely to use available benefitsthan women with high overall supervisor support. This findingsuggests that the use of benefits is greater under nonsupportivework conditions.Perceptions of supervisor support did not moderate theimpact of work demands on work-family role strain for thissample. Future research is needed using samples that havegreater variation in structural work demands, such as work hoursand the scheduling of these hours.Finally, perceptions of supervisor support did not moderate86the impact of the use of benefits on work-family role strain.This finding may be an artifact of the type of benefit used, andthe point of reference taken in assessing the use of benefits.ImplicationsThe findings of this study have implications for thefamily, work-family research and theory, and businessorganizations. Implications for the family are that having asupportive, flexible workplace may help to facilitate parents'ability to balance work and family roles. While employed parentsmay not be able to choose employment positions in workenvironments that are supportive of employees with familyresponsibilities, parents can play an influential role inmodifying the overall culture or philosophy of their workorganization.Much of the work-family research on coping with work-familydifficulties has focused on how individuals and familiesmanipulate employment and family demands in an attempt tocoordinate work and family roles. This study addressed the needto investigate coping mechanisms at the institutional level. Thefindings of this research support the view that the workplacecan play an important role in alleviating work-familydifficulties for employed parents. Due to the limited amount ofresearch on this issue, further research is needed whichcontinues to investigate dimensions of workplace support usingdiverse samples. Such research may help to remove the soleresponsibility for solving work-family problems from the87employed parent, and in particular the employed mother.The findings of this study also have implications fortheory development. The results of this study, corroborated withemerging research on work-family outcomes, suggest that anintegrated theory of work-family relationships is needed whichincorporates individual, family and institutional levels ofanalysis.This research also raises^issues^for businessorganizations. Although previous research has not examined therelationship between availability and use of family-orientedbenefits, the results of this study indicate that benefitsoffered by the workplace are used. The women in this sample didappear to limit their use of available benefits to four or fewerbenefits and were less likely to use all of the benefitsavailable to them when a range of benefits was offered. The mostfrequently used family-oriented benefits were flextime, leave inlieu of overtime, and short term leave for personal or familyreasons. Of the benefits that were not available to this sample,a large proportion of respondents reported that they would usesick child days with pay, personal days with pay, and workshopsor seminars on work-family issues if they were offered by theirworkplace. These findings suggest that by conducting a thoroughneeds assessment, workplaces may be able to limit the number ofbenefits available and still meet the salient needs of theiremployees.A second implication for business organizations is that88working in an environment that is supportive of employees' work-family difficulties and having a supervisor who is flexible whenfamily emergencies arise was found to be related to lower levelsof work-family role strain for this sample of employed mothers.This finding suggests that workplaces that cannot afford tooffer formal family-oriented benefits can assist their employeesby establishing informal types of workplace support. While thisstudy does not directly assess work-related outcomes, reducedstrain between work and family roles may have beneficialoutcomes for employee productivity and morale.89ReferencesAmatea, E. S., Cross, E. G., Clark, J. E., & Bobby, C. L.(1986). Assessing the work and family role expectations ofcareer-oriented men and women. Journal of Marriage and theFamily, 48, 831-838.Axel, H. (1985). Corporations and families. New York: TheConference Board.Blishen, B. R., Carroll, W. K., & Moore, C. (1987). The 1981socioeconomic index for occupations in Canada. CanadianReview of Sociology and Anthropology, 24(4), 465-488.Bohen, H. H., & Viveros-Long, A. (1981). Balancing jobs andfamily life: Do flexible work schedules help? Philadelphia:Temple University Press.Bowen, G. L. (1988). Corporate supports for the family lives ofemployees: A conceptual model for program planning andevaluation. Family Relations, 37, 183-188.Burud, S. L., Aschbacher, P. R., McCroskey, J. (1984). Employersupported childcare: Investing in human resources. Boston,MA: Auburn House.Campbell, M. L., & Moen, P. (1992). Job-family role strain amongemployed single mothers of preschoolers. Family Relations,41, 205-211.Caplan, R. D., Cobb, S., French, T. R., Harrison, R. V., &Pinneau, S. R. (1975). Job demands and worker health (DHEWPublication No. NIOSH 75-160). Washington, DC: U.S.Government Printing Office.CARNET, The Canadian Aging Research Network. (1992). UnpublishedQuestionnaire, Research being conducted at the Universityof Guelph, Gerontology Research Centre.Christensen, K. E., & Staines, G. L. (1990). Flextime: A viablesolution to work/family conflict. Journal of Family Issues,11(4), 455-476.Crouter, A. C. (1984). Spillover from family to work: Theneglected side of work-family interference. HumanRelations, 37, 11-28.Duffy, A., Mandell, N., & Pupo, N. (1989). Few choices: Women, work and family. Toronto: Coach House Press.90Elman, M. R., & Gilbert, L. A. (1984). Coping strategies forrole conflict in married professional women with children.Family Relations, 33, 317-327.Fernandez, J. P. (1985). Child care and corporate productivity: Resolving family/work conflict. New York: Lexington Books.Friedman, D. (1987). Family-supportive policies: The corporatedecision-making process. New York: The Conference Board.Galinsky, E. (1986). Family life and corporate policies. In M.W. Yogman & T. B. Brazelton (Eds.), In support of families(pp. 109-272). Cambridge, MA: Harvard University Press.Galinsky, E., & Stein, P. J. (1990). The impact of humanresource policies on employees. Journal of Family Issues,11, 368-383.Goff, S. J., Mount, M. K., & Jamison, R. L. (1990). Employersupported child care, work/family conflict, andabsenteeism: A field study. Personnel Psychology, 43, 793-809.Goode, W. J. (1960). A theory of role strain. AmericanSociological Review, 25(4), 483-496.Greenberger, E., Goldberg, W. A., Hamill, S., O'Neil, R., &Payne, C. K. (1989). Contributions of a supportive workenvironment to parents' well-being and orientation to work.American Journal of Community Psychology, 17(6), 755-783.Greenglass, E. R., Pantony, K. L., & Burke, R. J. (1989). Agender-role perspective on role conflict, work stress andsocial support. In E. B. Goldsmith (Ed.), Work and family: Theory, research, and applications (pp. 317-328). NewburyPark, CA: Sage.Hansen, G. L. (1991). Balancing work and family: A literatureand resource review. Family Relations, 40, 348-353.Hochschild, A. (1989). The second shift: Working parents and therevolution at home. New York: Viking.Holahan, C. J., & Moos, R. H. (1981). Social support andpsychological distress: A longitudinal analysis. Journal ofAbnormal Psychology, 90(4), 365-370.House, J. S. (1981). Work stress and social support. Reading,MA: Addison-Wesley.91Hughes, D., & Galinsky, E. (1988). Balancing work and familylives: Research and corporate applications. In A. E.Gottfried & A. W. Gottfried (Eds.), Maternal employment andchildren's development (pp. 233-268). New York: PlenumPress.Kamerman, S. B., & Kahn, A. J. (1987). The responsive workplace.New York: Columbia University Press.Kanter, R. B. (1977). Social science frontiers: Work and familyin the United States. New York: Russell Sage Foundation.Katz, M. H., & Piotrkowski, C. S. (1983). Correlates of familyrole strain among employed black women. Family Relations,32(3), 331-339.Keith, P. M., & Schafer, R. B. (1980). Role strain anddepression in two-job families. Family Relations, 29, 483-488.Kelly, R. F., & Voydanoff, P. (1985). Work/family role strainamong employed parents. Family Relations, 34, 367-374.Kobasa, S. O., & Puccetti, M. C. (1983). Personality and socialresources in stress resistance. Journal of Personality andSocial Psychology, 45(4), 839-850.Kome, P. (1982). Somebody has to do it. Toronto: McClelland &Stewart.Ladewig, B. H. (1990). The occupational commitment of men andwomen: Personal and family variables. Home EconomicsResearch Journal, 18(4), 336-344.LaRocco, J. M., House, J. S., & French, J. R. Jr. (1980). Socialsupport, occupational stress, and health. Journal of Healthand Social Behavior, 21(3), 202-218.Lee, R. A. (1983). Flextime and conjugal roles. Journal ofOccupational Behavior, 4, 297-315.McCroskey, J. (1982). Work and families: What is the employer'sresponsibility? Personnel Journal, 61, 30-38.McNeely, R. L., & Fogarty, B. A. (1988). Balancing parenthoodand employment: Factors affecting company receptiveness tofamily-related innovations in the workplace. FamilyRelations, 37, 189-195.Magid, R. Y. (1983). Child care initiatives for working parents: Why employers get involved. New York: American ManagementAssociation.92Menaghan, E. G., & Parcel, T. L. (1990). Parental employment andfamily life: Research in the 1980's. Journal of Marriageand the Family, 52, 1079-1098.Michelson, W. (1985). From sun to sun: Daily obligations andcommunity structures in the lives of employed women andtheir families. New Jersey: Rowman & Allenheld.Moen, P., & Dempster-McClain, D. I. (1987). Employed parents:Role strain, work time, and preference for working less.Journal of Marriage and the Family, 49, 579-590.Ontario Women's Directorate. (1991). Work and family: Thecrucial balance. Ottawa: Ministry of Community and SocialServices.Paris, H. (1989). The corporate response to workers withfamily responsibilities. Ottawa: The Conference Board ofCanada.Perry, K. S. (1982). Employers and child care: Establishingservices through the workplace. Washington, DC: Women'sBureau, U.S. Department of Labor.Piotrkowski, C. S., Rapoport, R. N., & Rapoport, R. (1987).Families and work. In M. B. Sussman & S. K. Steinmetz(Eds.), Handbook of Marriage and the family (pp. 251-284).New York: Plenum Press.Pleck, J. H., Staines, G. L., & Lang, L. (1980). Conflictsbetween work and family. Monthly Labor Review, 103(3), 29-31.Presser, H. B. (1987). Work shifts of full-time dual-careercouples: Patterns and contrasts by sex of spouse.Demography, 24, 99-112.Quinn, R. P., & Staines, G. L. (1979). The 1977 Duality ofEmployment Survey. Ann Arbor, MI: University of Michigan.Raabe, P. H. (1990). The organizational effects of workplacefamily policies. Journal of Family Issues, 11(4), 477-491.Raabe, P. H., & Gessner, J. C. (1988). Employer familysupportive policies: Diverse variations on the theme.Family Relations, 37, 196-202.Repetti, R. L. (1987). Individual and common components of thesocial environment at work and psychological well-being.Journal of Personality and Social Psychology, 52(4), 710-720.93Shinn, M., Wong, N. W., Simko, P. A., & Ortiz-Torres, B. (1989).Promoting the well-being of working parents: Coping, socialsupport, and flexible job schedules. American Journal ofCommunity Psychology, 17, 31-55.Staines, G. L., & Pleck, J. H. (1983). The impact of workschedule on the family. Ann Arbor, MI: The University ofMichigan.Statistics Canada. (1990). Focus on Canada: Women and the labourforce. Ottawa: Minister of Supply and Services, Canada.Vaux, A.^(1988). Social support: Theory, research andintervention. New York: Praeger.Voydanoff, P. (1987). Work and family life. Beverly Hills, CA:Sage.Voydanoff, P. (1988). Work role characteristics, familystructure demands, and work/family conflict. Journal ofMarriage and the Family, 50, 749-761.Voydanoff, P. (1989). Work and family: A review and expandedconceptualization. In E. B. Goldsmith (Ed.), Work andfamily: Theory, research, and applications (pp. 1-22).Newbury Park, CA: Sage.Voydanoff, P., & Kelly, R. F. (1984). Determinants of workrelated family problems among employed parents. Journal ofMarriage and the Family, 46(4), 881-892.Williams, G. (1990). A manager's guide to work options: Anoverview. Victoria, BC: Work Well Publications.Winett, R. A., Neale, M. S., & Williams, K. R. (1982). Theeffects of flexible work schedules on urban families withyoung children: Quasi-experimental, ecological studies.American Journal of Community Psychology, 10, 49-64.Youngblood, S. A., & Chambers-Cook, K. (1984). Child careassistance can improve employee attitudes and behavior.Personnel Administrator, 29(2), 45-46 & 93-95.Appendix ADaycare Correspondence9495Daycare Director/Supervisor Information LetterTHE UNIVERSITY OF BRITISH COLUMBIASchool of Family andNutritional Sciences2205 East MallVancouver, B.0 Canada V6T 1W5Division of Family SciencesDear Director/Supervisor,I am a graduate student in Family Studies at UBC and for my Master's thesis I amresearching combining work and family life. I am particularly interested inexamining how the work environment influences employed mothers' ability tointegrate paid employment with raising a family.In order to carry out my research I need volunteers! I am looking for employedmothers who have an immediate supervisor or manager at work. Mothers who agreeto participate will be asked to fill out an 11 page questionnaire at theirconvenience. The questionnaire will require approximately 30 minutes of their time.If you approve of this project, I would like to recruit volunteers through your daycare center. This would entail providing parents with a recruitment letter; postingan announcement; and distributing questionnaires to willing participants throughyour center. Participants are asked to mail the completed questionnaire to UBC.The study will be organized and conducted so as to avoid, as much as possible, anyinconvenience to you or your staff. Any specific procedures, identified by yourdaycare center, for the recruitment of participants and the distribution ofquestionnaires will be followed. If you have any questions regarding the studyplease contact us. We will return your call as soon as possible.Thank you for your consideration of this project. Please sign the attached form toconfirm whether or not you approve of the present study and are willing to let usrecruit subjects and distribute questionnaires through your center.Sincerely,Jennifer Warren^Phyllis J. Johnson, PhDM.A. Candidate Associate Professor(604) 822-2502 (604) 822-430096Permission Form to Recruit Volunteer ParticipantsThrough the Daycare CenterDateI, ^  Director/Supervisor of do approve of thepresent study and am willing to let the investigators recruitvolunteers and distribute questionnaires through this daycarecenter.I, ^  Director/Supervisor of do not approve of thepresent study and am not willing to let the investigatorsrecruit volunteers and distribute questionnaires through thisdaycare center.97Certificate of U.B.C. Ethics Approval Given to theDaycare Director/Supervisor The University of British Columbia^B92-220Office of Research ServicesBEHAVIOURAL SCIENCES SCREENING COMMITTEE FOR RESEARCHAND OTHER STUDIES INVOLVING HUMAN SUBJECTSCERTIFICATE^of APPROVALINVESTIGATOR: Johnson, P.J.UBC DEPT:^Family & Nutr SciINSTITUTION:^UBC CampusTITLE:^Work-family role strain among employedmothers of preschoolers: the impact ofworkplace supportNUMBER:^B92-220CO-INVEST:^Warren, J.APPROVED:^AUG 2 5 1992The protocol describing the above-named project has beenreviewed by the Committee and the experimental procedures werefound to be acceptable on ethical grounds for researchinvolving human subjects.Of^Dr. R.D. :prat ey^/Director, Research S rvicesand Acting ChairmanTHIS CERTIFICATE OF APPROVAL IS VALID FOR THREE YEARSFROM THE ABOVE APPROVAL DATE PROVIDED THERE IS NOCHANGE IN THE EXPERIMENTAL PROCEDURES98Recruitment Notice Posted in All Participating Daycare Centers ARE YOU COMBINING PAIDEMPLOYMENTWITH RAISING A FAMILY?VOLUNTEERS NEEDED TO PARTICIPATE IN A STUDYABOUT WORK AND FAMILY LIFE!!ELIGIBLE PARTICIPANTS WILL:-- BE MOTHERS OF AT LEAST ONE PRESCHOOL AGECHILD-- BE EMPLOYED OUTSIDE THE HOME-- HAVE AN IMMEDIATE SUPERVISOR, MANAGER, ORBOSS AT WORK-- HAVE 15-30 MINUTES TO COMPLETE AQUESTIONNAIRETO PARTICIPATE IN THIS STUDYPLEASE PICK UP A QUESTIONNAIREFROM YOUR DAYCARE CENTER'SDIRECTOR OR SUPERVISOR.FOR MORE INFORMATION PLEASE CONTACTJENNIFER WARREN. MA CANDIDATE. 822-2502PHYLLIS J. JOHNSON, PHD. 822-4300SCHOOL OF FAMILY AND NUTRITIONAL SCIENCESTHE UNIVERSITY OF BRITISH COLUMBIAReminder Notice Posted in All Participating Daycare CentersREMINDERRE: WORK AND FAMILY LIFE STUDYIF YOU HAVE NOT YETRETURNED YOURCOMPLETEDQUESTIONNAIRE WEWOULD APPRECIATERECEIVING IT AS SOON ASPOSSIBLE!THANKS TO EVERYONE WHOPARTICIPATED IN THIS STUDY!Jennifer WarrenMA CandidateSchool of Family & Nutritional SciencesThe University of British ColumbiaAppendix BParent Contact Letter and Questionnaire100101THE UNIVERSITY OF BRITISH COLUMBIASchool of Family andNutritional SciencesDivision of Family Sciences2205 East MallVancouver, B.C. Canada V6T 1W5Dear Participant,Thank you for volunteering to participate in the Work and Family Life Project. Thepurpose of this questionnaire is to try to better understand employed mothers'experiences in combining paid work and family roles. The questionnaire is dividedinto five parts. Within each part, instructions accompany each set of questions. Itshould take you approximately 30 minutes to complete. Your completion of thisquestionnaire will be taken as your consent to participate in this project.All of your responses will be confidential. Individual responses will never bereported. Questionnaires will be assigned a code for data entry and the data willbe used in statistical form only. To insure confidentiality, please do not write yourname on any part of this questionnaire. Your answers to all of the questions wouldbe greatly appreciated, but you are free to refuse to answer any part(s) of thisquestionnaire just as you are free to withdraw your participation at any point.Should you decide not to participate or not to complete all of the questions, pleasebe assured that access to the services of your daycare center will not bejeopardized.As soon as you have completed the questionnaire, please mail it in the stampedand self addressed envelope provided!Thank you very much for taking the time to fill out this questionnaire. Yourresponses will help to provide a better understanding of the role of the workplacein facilitating work-family coordination. If you have any questions or would likefUrther information, please contact us. One of us will return your phone call assoon as possible.Sincerely,Jennifer Warren^Phyllis J. Johnson, PhDMA Candidate Associate Professor(604) 822-2502 (604) 822-4300102PART IPLEASE ANSWER THE FOLLOWING QUESTIONS ABOUT YOUR OCCUPATION AND WORKSETTING BY FILLING IN THE REQUESTED INFORMATION OR BY PLACING A CHECK( V) NEXT TO THE BEST RESPONSE.1. What kind of work do you do? (Please provide your complete job title; e.g.receptionist, bookkeeper, salesperson)2. For how long have you had your current employment position with your presentemployer?^ years   months3. How many hours do you work (for pay) in an average week? ^ hours4. In a typical week, do you work after hours (without additional pay) for your jobeither at your workplace or at home?^ no^ yes, at my workplace only^ yes, at my home only^ yes, both at home and at work5. About how long does it usually take you to get from:your home to your workplaceyour workplace to your home minutesminutes6. Does your work schedule usually include evenings/nights?^ yes   no7. Does your work schedule usually include weekend days?^ yes   no8. Please list one workplace change that would make it easier for you to balancework and family life (be as specific as possible).9 Below is a list of benefits that are sometimes offered by workplaces. Pleasemark a check (^) next to those benefits that(1) are available to you in your current employment position.(2) you are currently using or have used in the past year.(3) you would consider using if they were available to you.103Benefit is available Currently usingor used in pastyearWould useif availableDon'tknow No Yesa. flexible hours/flextime^ (^)^(^)^(^)b. job sharing (^)^(^)^(^)c. compressed work week^( )^( )^( )d. part—time with nobenefits^ (^)^(^)^(^)e. part—time withprorated benefits^( )^( )^( )f. work at homearrangements^( )^( )^( )g• short term leave forpersonal/family reasons(3 days-90 days)h. extended leave forpersonal/family reasons(greater than 90 days)^( )^( )^( )i. personal days with pay^( )^( )^( )j. sick child days with pay ( )^( )^(^)k. leave in lieu of overtime ( )^( )^( )1. employee assistanceprograms (EAP)^( )^( )^( )m. seminars/workshops forbalancing work andfamily10410. Are there any benefits not listed in question 9 that you would like to havemade available to you (please specify)?11 a. Has your spouse used (in the past year), or is your spouse currently usingany of the 13 benefits listed in question 9?no   not applicableb. If yes, please circle the letter(s) below that correspond to each benefit yourspouse used or is currently using.a^b^c^de^f^g^h112. Please indicate how much you agree or disagree with the following statementsabout your job by circling the number that best expresses your opinion.a. On my job, there is always agreat deal of work to be done.b. I have too much work to doeverything well on my job.stronglydisagree disagree1^21^21^21^21^21^21^2neitherdisagreenor agree3333333stronglyagree^agree4^54^54^54^54^54^54^5c. I never seem to have enough timeto get everything done on my job.d. My job requires me to work veryfast.e. On my job, I usually have tightdeadlines to meet.f. I want to work, but I do not wantto have a demanding job/career.g^I expect to make as many sacrificesas are necessary in order toadvance in my job/career.h. I value being involved in a job/careerand expect to devote the time andeffort needed to develop^it. 1 2 3 4 5yes105stronglydisagreei. I^expect to devote a significantamount of time to building myjob/career and developing the skillsnecessary to advance in myjob/career.^ 1j. I expect to devote whatever timeand energy it takes to move up inmy job/career field.^ 1k. Even when I'm busy doing otherthings, I often think about my job.^1disagree222neitherdisagreenor agree333agree444stronglyagree5551. My main satisfaction in life comesfrom my job.m. The most important things thathappen to me involve my job.112233445513. How supportive is your work environment of employees when they have difficultiescoordinating work and family responsibilities? Please circle the number that bestexpresses your opinion.not at all^not very^somewhat^moderately^verysupportive^supportive^supportive^supportive^supportive1^2^3 4^5PART IIPLEASE ANSWER THE FOLLOWING QUESTIONS ABOUT YOUR IMMEDIATE SUPERVISOR,MANAGER OR BOSS AT WORK.14. a. Do you have an immediate supervisor, manager or boss--someone who is directlyover you at work?^ yes   nob. If yes, is this person ^ male or ^ female?10615. Please circle the number that best describes how much you agree with thefollowing statements about your supervisor/manager.stronglydisagreea. My supervisor/manager goes out ofhis/her way to do things to makedisagreeneitherdisagreenor agree agreestronglyagreemy work life easier for me.b. It is very easy for me to talk tomy supervisor/manager.c. My supervisor/manager can be reliedon when things get tough at work.d. My supervisor/manager is willing tolisten to my personal problems.e^My supervisor/manager understandsthat I have to meet familyresponsibilities as well^as thoserelated to my job.f. My supervisor/manager is awareof the family demands being placedon me.g. My supervisor/manager tries to findways of helping me meet my familyresponsibilities.h. My supervisor/manager does notunderstand that it may be difficultfor me to coordinate work andfamily^responsibilities.i. I can talk to my supervisor/managerabout family—related problems that1111111111122222222222333333333334444444444455555555555are making it difficult for me tocombine work and family roles.j. My supervisor/manager expects me tokeep my work and home life separate.k. My supervisor/manager isknowledgeable about companypolicies that apply to familyissues.1^I can talk to my supervisor/managerabout work—related problems that1 2 3 4 5are making it difficult for me tocombine work and family roles.10716. If the following situations were to happen, how do you think yoursupervisor/manager would behave? Please circle the number that best describeshow she/he might behave.Never Seldom Sometimes Usually Alwaysa. If I ask for extra vacation time(unpaid) so I can spend more timewith my family, my supervisor/managergives it to me.b. My supervisor/manager is flexible inscheduling so as to accommodate myfamily needs (e.g. take child to thedoctor.c. If I receive phone calls (at work) fromfamily members, my supervisor/manageris understanding.d. My supervisor/manager lets me take workhome if I need to, instead of asking meto work late at the office.e. My supervisor/manager lets me bring mychild to work in an emergency.f. My supervisor/manager lets me come inlate or leave early to accommodate myfamily needs.g• My supervisor/manager will let me takean occasional day off without pay.h. My supervisor/manager lets me come inat a non—scheduled time (e.g., on theweekend) to make up work I missedbecause of family commitments.i. My supervisor/manager lets me work fromhome if I can't come in on a given daybecause of family matters.1 2 3 4 51 2 3 4 51 2 3 4 51 2 3 4 51 2 3 4 51 2 3 4 51 2 3 4 51 2 3 4 51 2 3 4 5PART IIIPLEASE ANSWER THE FOLLOWING QUESTIONS ABOUT COMBINING WORK AND FAMILYRESPONSIBILITIES.17. How much do your job and family life interfere with each other? Please circle thenumber that best expresses your opinion.not at all^just a little^a moderate amount^quite a bit^a great deal110818. Please indicate, by circling the relevant number next to each statement, your levelof agreement with the following statements.stronglydisagreea. The most important things thatdisagreeneitherdisagreenor agree agreestronglyagreehappen to me involve my family. 1 2 3 4 5b. Even when Pm busy doing otherthings, I often think about myfamily. 1 2 3 4 5c. My main satisfaction in lifecomes from my family. 1 2 3 4 519. Please indicate, by circling the relevant number next to each statement, how oftenyou feel each of the following-.a. My job keeps me away from mynever rarelysome ofthe timemost ofthe time alwaysfamily too much.b. I worry about how my childrenare while I am working.c. I have a good balance betweenmy job and my family time.d. I wish I had more time to dothings for my family.e. I feel physically drained when11112222333344445555I get home from work.f.^I feel emotionally drained when1 2 3 4 5I get home from work.g. I feel I have to rush to geteverything done each day.h. My time off from work matchesother family members' scheduleswell.i. I worry that othe^people at workthink my family interferes withmy job.j. I^feel^I don't have enough timefor myself.k.^I^feel^more respected than^Iwould^if I^didn't^have^a job.1111112222223333334444445555551091.^I worry whether I should workless and spend more time withnever rarelysome of^most ofthe time^the time^alwaysmy children.m. I am a better parent because Iam not with my children all day.n. I find enough time for mychildren.o. I feel I have more to do than Ican handle comfortably.p. I have as much patience with mychildren^as I^would like.q. I am comfortable with thearrangements for my childrenwhile I^am working.r. Making arrangements for mychildren^while I work involvesa^lot^of effort.s.^I^worry^that other people^feel^Ishould spend more time with mychildren.1111111122222222333333334444444455555555PART IV20. Would you be interested in attending a workshop or seminar on any of thefollowing topics? Please check (N/ ) yes or no for each topic.Stress management   yes   noTime management   yes   noBalancing work and family   yes   noChild care yes   noIf you checked yes for any of the above topics, please answer questions 21 and 22.If you did not check yes for any of the above topics, please proceed to question 23.21. When would be the most ideal time for you to attend a workshop or seminar onany of the topics listed in question 20? Please check only one category.lunch houreveningweekendother (please specify) ^11022. a. Workshops and seminars may be offered at a variety or locations. Which of thefollowing locations would be the most ideal location for you to attend aworkshop or seminar on any of the above topics. Please check only onecategory.workplacecommunity centreeducational institution (e.g. school, college, university)churchconference centreother (please specify) ^b. Why would this be the most ideal location for you?PART VPLEASE ANSWER THE FOLLOWING QUESTIONS BY FILLING IN THE REQUESTEDINFORMATION OR BY PLACING A CHECK (v/) NEXT TO THE BEST RESPONSE.23. What is your age?   years24. What is your present marital status? Please check only one category.^ married   divorced^ common-law (for at least 6 months) ^ never married^ separated   widowed25. If you are married or living in common-law, please answer the followingquestions about your spouse/partner.a. Is your spouse/partner employed? ^ yes   no (go to question 26)b. What kind of work does your spouse/partner do? (Please give full job titlee.g., retail clerk, lawyer, mechanic, teacher.)c. How many hours does your spouse/partner work on this job in an averageweek?hoursd. Does your spouse/partner's regular work schedule include evenings/nights?yes^noe. Does your spouse/partner's regular work schedule include weekend days?^ yes   no2 6 How many children are currently living with you?27. Please provide the following information about your child(ren).Type of childcare arrangement(e.g. group day care, after school carechild^age ( yrs)^nanny, babysitter, relative, none)123456728. How satisfied are you with your current childcare arrangements'very dissatisfieddissatisfiedsatisfiedvery satisfied29. How difficult would it be for you to find alternative care should your regulararrangements breakdown?not difficultsomewhat difficultvery difficult30. How many years of education do you have?   years(e.g. 12 years of public school + 1 year of trade school + 1 year of college=14 years)If you have completed any part of your education on a part-time basis, pleasecalculate the full-time equivalent.31. What is the highest level of education that you have completed? Please check onlyone category.^ some elementary or public school^ completed elementary or public school^ some high school^ completed high school^ some vocational or technical college^ completed vocational or technical college^ some training in a special diploma program (e.g., nursing, teaching)^ completed special diploma program^ some university^ completed undergraduate university degree^ some university post-graduate level^ completed post-graduate university degree111Less than $10,000^ $50,001-$60.000$10,001-$20,000 $60,001-$70,000$20,001-$30,000 $70,001-$80,000$30,001-$40,000^ $80,001-$90,001$40,001-$50,000 Over $90,00111232. What was your total personal income from wages and salaries in 1991, before taxeswere deducted?33. What was your total household unit's income from all sources in 1991, before taxeswere deducted?^ Less than $10,000   $70,001-$80,000^ $10,001-$20,000   $80,001-$90,000^ $20,001-$30,000   $90,001-$100,000^ $30,001-$40,000   $100,001-$110,000^ $40,001-$50,000   $110,001-$120,000^ $50,001-$60,000   Over $120,000$60,001-$70,00034. a. Do you identify with any ethnic group, other than Canadian?yes^ nob. If yes, with which ethnic group do you identify? Please specify (e.g. English,Italian, Chinese, Polish etc.).35. a. Do you hire anyone from outside your household to help with home chores ona regular basis?yes^nob. If yes,What do they do?How often (e.g. every day, once a week)?THANK YOU VERY MUCH FOR TAKING THE TIME TO COMPLETE THIS QUESTIONNAIRE!PLEASE MAIL THE QUESTIONNAIRE, AS SOON AS POSSIBLE, IN THE ENVELOPE PROVIDED.Appendix CFamily-Oriented Benefits 113114FlextimeA standard number of hours worked per week with flexible startand finish times.Job sharingTwo or more people share the work hours, responsibilities,salary, and benefits of one job.Compressed work weekA standard number of hours worked per week but in less than fivedays.Part-time (no benefits) Working fewer than 30 hours per week with no benefits.Part-time (pro-rated benefits) Working fewer than 30 hours per week with benefits coverage paidon a pro-rated basis.Work at homeWork responsibilities completed at home rather than at theworkplace.Personal days with payPaid days off work that can be taken for personal reasons.Sick child days with payPaid days off work that can be taken to care for a sick child.Short-term leave for personal/family reasons Leave from work for up to 90 days for personal or familyreasons.Extended leave for personal/family reasons Leave from work for longer than 90 days for personal or familyreasons.Leave in lieu of overtimeTime off work in lieu of overtime pay.115Employee assistance programs (EAP) Counselling services available to help employees deal with workand family related problems.Seminars/workshops on balancing work and family lifeSeminars or workshops designed to assist employees incoordinating work and family responsibilities.Sources: Paris, 1989; Ontario Women's Directorate, 1991;Williams, 1990.Appendix DSociodemographic Characteristics of the Sample116117Table 1Sociodemographic Characteristics of the Sample of Employed Mothers with Preschoolers (n = 116)Characteristic^ PercentMother's Age20-29^ 19.130-39 62.040-49 18.9Marital StatusMarried^ 47.4Common Law 12.1Separated 11.2Divorced 8.6Never Married^ 18.1Widowed^ 0.9Number of ChildrenOne 67.2Two^ 25.9Three or more^ 6.9Ethnic BackgroundNone^ 70.7Asian 9.5European 7.8United Kingdom^ 4.3Other^ 7.7OccupationManagerial, administrative & related^30.2Clerical & related^ 30.2Medicine & health 11.2Teaching & related 8.6Other^ 19.1Personal Income< $10,000$10,001-20,000$20,001-30,000$30,001-40,000$40,001-50,000> $50,0014.318.132.825.013.87.7118Table 1 Continued.Characteristic^ PercentHousehold Income< $10,000$10,001-20,000$20,001-30,000$30,001-40,000$40,001-50,000$50,001-60,000$60,001-70,000$70,001-80,000$80,001-90,000> $90,001EducationSome high schoolCompleted high schoolSome vocationalCompleted vocationalSome special diplomaCompleted special diplomaSome universityCompleted undergraduateSome university postgradCompleted university postgrad3.412.919.07.86.912.19.510.86.97.03.413.88.610.34.311.215.514.73.414.7Spouse/partner employment statusEmployed^ 95.6Not employed 4.4Spouse/partner work hoursLess than 30 hours^ 9.530 hours or more 90.5Spouse/partner shift workEvening/night shifts^ 52.0Weekend shifts^ 42.0Spouse/partner occupationManagerial, administrative & related^33.3Sales^ 13.6Service 7.6Transport equipment operating^7.6Other 37.9

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.831.1-0086183/manifest

Comment

Related Items