Women’s Jobs and Family Adjustment*

126

Women’s Jobs and

Family Adjustment*CHAYA S. PIOTRKOWSKIPAUL CRITS-CHRISTOPH

Yale University

THIS study investigated the relationship between multiple characteristics ofwomen’s jobs and their family adjustment in a sample of 99 women in dual-earner families. Six work-related variables were considered simultaneously aspredictors of family adjustment: intrinsic job gratification, satisfaction withjob security, job-related mood, time spent at work, occupational prestige,and salary. The results were consistent with the hypothesis that women’s paidwork lives influence their family adjustment. Women’s experiences of theirwork were significantly related to two of three measures of family adjust-ment. Salary was negatively associated with satisfaction with family relationsfor women in low-status occupations only. Time spent at work and occupa-tional prestige showed no significant associations with reported adjustment.Of the three indicators of family adjustment, marital satisfaction appeared tobe relatively immune from work influences. We hypothesized that the maritalrelationship may be less sensitive to women’s paid work than are otheraspects of their family relations.

The increasing participation of married women in the laborforce has spurred research into the relationship of women~semployment to family adjustment in dual-earner families. iWhen women’s employment status is considered, currentresearch evidence indicates that, in general, husbands andwives in dual-earner and single-earner families do not differ

*The authors would like to thank Faye J. Crosby for her generosity in makingavailable data collected in a study of relative deprivation and working women. 7hisstudy of relative deprivation was supported by NIMH Grant 1R01-MH31595. Datawere collected in 1978. The present study was supported by NIH Biomedical ResearchSupport Grant 5-S07-RR07015 to Yale University.

JOURNAL OF FAMILY ISSUES, Vol. 2 No. 2, June 1981 126-147@ 1981 Sage Publicatians, Inc.

at PENNSYLVANIA STATE UNIV on February 20, 2016jfi.sagepub.comDownloaded from

http://jfi.sagepub.com/

127

significantly in level of family-and particularly-maritaladjustment (see, for example, Wright&dquo; 1978; Nye, 1974; Staineset al., 1978 ; Booth, 1979; Locksley, 1980). Despite the pro-liferation of such studies of women’s employment, we stillknow relatively little about the ways women’s paid work livesmay influence the quality of family relationships. Qualitativestudies that have investigated this relationship have limitedgeneralizability because of the small samples utilized and theiremphasis an professional and managerial families. Moreover,studies that compare family adjustment in dual-earner andsingle-earner families are not designed to explicate the pro-cesses that link women’s jobs to family adjustment.

It is our opinion that, at this time, an especially fruitfulapproach to the study of women’s work and family adjustmentin dual-earner families is to deemphasize questions aboutwhich family type fares better and to focus instead on theprocesses linking women’s jobs and family relations withingroups. Because men’s and women’s roles and experiences inthe family differ, such analyses need to be undertaken sepa-rately for husbands and wives. The development of anadequate theory for understanding the relationship betweenwork life and family adjustment in dual-earner families

ultimately rests on our ability to describe the interactionsbetween husbands’ and wives’ paid and unpaid work roles asthey directly and indirectly influence each family member’sadjustment. Such a conceptualization is necessarily complex.Consequently, it is useful to begin with simpler components. Inthe study reported here we consider the relationship ofwomen’s employment to women’s family adjustment in dual-earner families.

It is important that research on women and their employ-ment treat women’s work with sufficient complexity. Thepioneering qualitative studies of dual-career families have beenable to address the complexity of women’s work and familyroles by simultaneously considering a host of variables ininteraction (see, for example, Rapoport and Rapoport, 1971;Holmstrom, 1972). Quantitative research necessarily must



128

focus on fewer variables and relationships. However, toooften, quantitative studies have emphasized the fact of wo-men’s employment and attitudes toward and commitment tothe work role in general (see Safilios-Rothschild, 1970; Bailyn,1970: Nye, 1963; Ridley, 1973; Orden and Bradburn, 1969;Locksley, 1980). Unfortunately, such variables do not tell usmuch about women’s actual employment experiences and theconditions of their work (Hoffman, 1963), nor do they addressthe question of whether features of women’s particular jobsinfluence their family adjustment. A review of recent quantita-tive research on the relationship of women’s employment tofamily adjustment reveals that other job-specific variablesrarely are examined. Although some studies have consideredfactors such as women’s occupational prestige and job satisfac-tion (Ridley, 1973; Gover, 1963; Hoffman, 1963; Richardson,1979; Yarrow et al., Harrell and Ridley, 1975), to ourknowledge no studies have examined multiple characteristicsof women’s particular jobs simultaneously.

In the study reported here we utilized multiple regressionprocedures to investigate the relationship of specific jobexperiences and conditions of work to family adjustment for asample of women in dual-earner families. Family adjust-ment was defined as the degree to which a family member hassatisfying relationships with others in the family. More

specifically we focused on two general questions:

(1) Is the paid occupational life of women in dual-earner familiessignificantly related to their family adjustment?

(2) Which specific occupational factors are significantly asso-ciated with women’s family adjustment in dual-earner fami-lies, and what is the nature of these associations?

CONCEPTUAL BASIS OF STUDY

In selecting the work variables to be included in the analysis,we were guided by several considerations: Data were not



129

collected specifically for the purposes of this study; thus thevariables available to us were limited. Of the pool of potentialvariables, we included only those for which we could developreliable measures. Furthermore, we required that these vari-ables be meaningful for both men’s and women’s work in dual-earner families. With work life emerging as a significant liferole for women, it is fruitful to begin to conceptually integrateresearch on wives’ employment and the dual-earner familywith the literature on men’s occupations and family lives.These criteria led us to focus simultaneously on four generalfeatures of women’s paid work lives that may be related totheir family adjustment: job satisfaction, job-related psychol-ogical mood, time spent at work, and occupational rewards.

Men’s satisfaction with the intrinsic features of their workhas been viewed as a threat to their family participation and,therefore, to family adjustment (see Rapoport and Rapoport,1965; Young and Willmott, 1973). This notion is based on theassumption, noted by Marks (1977), that human psychological&dquo;energy&dquo; is scarce and finite. Consistent with Marks’s conceptof human psychological energy as expandable and as &dquo;created&dquo;by satisfying multiple roles, Piotrkowski (1979) alternativelyhas proposed that job gratification can result in positiveemotional &dquo;energy&dquo; available for family participation. She hasfurther proposed that the intervening link between degree ofjob gratification and family adjustment is the psychologicalmood state that results from what one does at work on a dailybasis.

Although limited, research on husbands&dquo; job satisfactionand their marital adjustment (Ridley, 1973; Kemper andReichler, 1976) and on mothers’ work satisfaction and mother-child relationships (Hoffman, 1963; Yarrow et al., 1962;Harrell and Ridley, 1975) supports the hypothesis that level ofgratification carries over into the family from the work place.Data concerning the relationship between wives’ job satisfac-tion and marital adjustment are less consistent (see Locksley,1980; Ridley, 1973; Staines et al., 1978). The possible influenceof job gratification on women’s family adjustment remains animportant research question.



130

Amount of time spent at work also has been hypothesized asa possible constraint on adequate family role performance forboth men (see Aldous, 1969; Young and Willmott, 1973) andwomen (see Sweet, 1973). However, this hypothesis generallywas not supported by a study of husbands’ work time (Clark etal., 1978). Although men’s work time may be less important forfamily relations than previously assumed, family adjustmentmay be particularly sensitive to women’s work time because oftheir central affective and household roles. Professional and

nonprofessional women alike complain of insufficient timewith family members (see Rapoport and Rapoport, 1971;Burke and Weir, 1976; National Commission on WorkingWomen, 1979; Heckman et al., 1977; Walshok, 1979). Sincefull-time employed women differ in the number of hours theyare employed, work time was included in our analysis as apossible predictor of women’s family adjustment.The effects of occupational rewards on family life have

received a great deal of attention in the literature on men’s

occupational and family lives. Although recent empirical evi-dence casts doubt on the continuing importance of husbands’occupational rewards (such as income or prestige) on marriage(Glenn and Weaver, 1978; Jorgensen, 1979), until recently ithas been agreed that men’s occupational rewards result inincreased marital solidarity and happiness (Scanzoni, 1970).On the other hand, instead of enhancing the marital relation-ship, women’s occupational achievements have been viewed asthreatening it. As Richardson (1979) has noted, Parsonianfunctionalist theory predicts that occupational status competi-tion between husbahds and wives poses a threat to affectionalrelations in the marriage. However, Richardson found noempirical support for the hypothesis that wives’ superioroccupational prestige was associated with decreased maritalhappiness. Safilios-Rothschild (1976) has suggested that awife’s higher occupational status can be tolerated as long as herincome is lower than her husband’s in those societies whereincome is the dominant status line. Including measures ofwomen’s occupational rewards a&dquo;owed us to evaluate theirrelevance for our sample.



131

There is a small body of literature suggesting that therelationships between occupational factors and family adjust-ment differ by occupational class. Conflicts between time spentat work and job gratification, on the one hand, and familyparticipation, on the other, have been assumed to be mostintense for men in managerial and professional occupations(Rapoport and Rapoport, 1965 ; Young and Willmott, 1973).In contrast, it has been assumed that lack of job gratificationand job involvement for men in working-class occupationsleads to a &dquo;natural&dquo; segmentation of work and family life (seeBlood and Wolfe, 1960; Kanter, 1977), to compensatory familyinvolvement (Dubin, 1956), or to the expenditure of unutilizedwork energy within the family realm (Young and Willmott,1973). Although these propositions have been advanced formen, they may be applied to employed women as well. Theexistence of such ~lass-conditional relations also was investi-

gated in this study.

METHOD OF STUDY

Participants for this study were drawn from a sample ofemployed white women from 25 through 40 years of age whoparticipated in a larger study of relative deprivation andworking women (Crosby, forthcoming). Respondents for thislarger study were obtained through stratified random samplingprocedures from the street lists of Newton, Massachusetts. Onthe basis of occupation, each potential respondent was as-signed an NORC status ranking (Davis, 1974). Participantswith NORC ratings of 40.0 or below were assigned to the low-status group. Participants with NORC ratings of 61.0 andabove were assigned to the high-status group. Included in thestudy reported here were the 40 low-status and 59 high-statusemployed married women in dual-earner families who worked35 hours per week or more. The final volunteer rate of over

84% for the study of relative deprivation increases ourconfidence that the results of the study are not biased due tosampling procedures.2



132

Research participants were interviewed in homes forapproximately one hour. Interview items tapping conceptuallyrelevant work and family variables were used to create themeasures for this study. Occupational rewards were measuredby the respondent’s NORC status ranking or low) and byreported salary.3 To provide an indicator of amount of timespent at work, respondents also were asked approximatelyhow many hours per week they worked. Multi-item self-reportscales measuring satisfaction with various aspects of work lifeand psychological mood state at work also were created. Scaleswith a Cronbach’s alpha coefficient of .65 or above4 were factoranalyzed to reduce the number of predictor variables. The datareduction procedures resulted in six predictor variables: twoself-report measures of job satisfaction (Intrinsic Job Gratifi-cation and Job Security), a measure of job-related mood(Positive Job Mood), measures of occupational reward (Statusand Salary), and a measure of work time (Time).5 Cronbach’salpha coefficients for the measures of job satisfaction and job-related mood ranged from .79 to .80.

Three indicators of women’s family adjustment were utilized.A scale measuring Marital Satisfaction was composed of eightitems asking about satisfaction with relations with spouse andwith marriage in general. A second indicator of familyadjustment, measuring satisfaction with Family Relations, wascomposed of two items tapping feelings about family life ingenera1.6 Finally, a Positive Home Mood score was computedas a measure of mood at home. This measure was included onthe assumption that moods experienced at home were oneindicator of how one felt about one’s relationships with otherfamily members. The procedure used in creating this scale wasparallel to that used in creating the Positive Job Mood scaledAlpha coefficients for these three scales ranged from .65 forFamily Relations to .88 for Marital Satisfaction.

SAMPLE CHARACTERISTICS

S a ir, n, iha rac- t ei 1 st- ics aie presented Table 1. As the d~t~. &dquo;c~&dquo; ’ ~ ¡ - { ;: ui1 a~~~~a~~5~



133

TABLE 1

Demogral)hic Characteristia

NOTE: The Ns may differ slightly because of missing data.

employed women. As expected, the high-status women weresignificantly more educated and earned significantly moremoney than the low-status women. High-status women alsotended to be older and to work significantly more hours perweek. These sample characteristics must be taken into accountwhen interpreting results.

DATA ANALYSES PROCEDURES

Status was used as a rough indicator of occupational class,and the data were analyzed separately for each status subgroupto determine whether relationships between women’s occupa-



134

tional lives and family adjustment differed by class. Threesimultaneous regression analyses were performed for eachstatus subgroup to determine the relation of the five remainingpredictor work variables to the dependent measures. In

addition, the presence or absence of children under the age of18 living at home was included as a control variable, as thisvariable has been associated with marital satisfaction (Nock,1979). The limited age range in the sample provided somecontrol for length of marriage. Within-group analyses resultedin controlling educational level, as educational level was highlycorrelated with Status (r = .75; p~.001). Other possible controlvariables were not included because of limitations in samplesize and potential problems of multicollinearity. 8To determine the possible effects of occupational prestige on

the dependent variables, the high- and low-status groups werecombined and regression analyses performed for each depen-dent variable. Because Status was moderately correlated withSalary (r = .53; p ~ . .00 1), a hierarchical regression analysis wasperformed in which Status was entered at Step 2 after thecontrol variable for presence or absence of children. At Step 3all remaining work variables were entered simultaneously.Entering Status before other work variables allowed for themost powerful test of the hypothesis that Status is a significant

predictor of family adjustment. Finally, cross-product interac-tion terms between Status and other work variables wereentered at Step 4 to determine if significant class interactionsexisted (see Cohen and Cohen, 1975). Because Positive JobMood was hypothesized as intervening between job satisfac-tion and family adjustment, a simultaneous regression analysisalso was performed for each status subgroup to determine theeffects of Job Security, Intrinsic Job Gratification, Time, andSalary on Positive Job Mood. Age was entered as a controlvariable because age has been associated with job satisfaction.

RESULTS AND DISCUSSION

Intercorrelations among the major variables are presentedin Table 2. Correlated predictor variables can create problems



135

TABLE 2

Zefo-OKkr CoereMaas ~nemg M~or VanaMes

*p<.05; ~~~ ~ .(11; ***p<.001NOTE: 1 = Salary; 2 = Time; 3 = Positive Job Mood; 4 = Intrinsic Job Gratifi-cation ; 5 = Job Security; 6 = Marital Satisfaction; 7 = Family Relations; 8 = Posi-tive Home Mood; 9 = Presence/Absence of Children (Absence = 1, Presence = 2).

of multicollinearity in the data analyses, thereby reducing ourability to discover significant relationships. Nevertheless, theseintercorrelations reflect the real conditions of work life. Asexpected, the three dependent variables were significantlycorrelated. Despite the substantial correlations between theFamily Relations scale and Marital Satisfaction, they wereretained as separate indicators. We assumed that the FamilyRelations scale tapped more global aspects of family adjust-ment than Marital Satisfaction.



136

TABLE 3

Multiple Regression Analysis Among Low-Status Women (N ::: 40)

*p ~ .05; **p ~ .01; ***p ~ .001NOTE: Absence of children = 1; presence of children = 2.

Results of the major data analyses are presented in Tables 3and 4. Limited sample size-especially for the low-statuswomen-means results must be interpreted cautiously. Thequestion of whether women’s occupational lives are signifi-cantly associated with their family adjustment can be answeredwith a cautious affirmative for this sample of highly educatedwomen. For each status subgroup, work variables accountedfor a significant proportion of the variance in two of the three



137

TABLE 4

Multiple Regression Anatysis Among High-Status Women (N = 59)

*p ~ .05; up ~ .01; ***p ~ .001NC3Ti;: Absence of children =1; presence of children = 2.

measures of family adjustment. The data analyses also ad-dressed our second research question of which particular workvariables were significant for family adjustment. When com-pared on the three dependent measures of family adjustment,the high- and low-status groups did not differ significantly.However, when we examined the relationship between specificwork variables and women’s family adjustment within eachgroup, interesting patterns emerged.



138

For the low-status group (see Table 3), level of Positive JobMood was positively related to women’s reports of MaritalSatisfaction, to their reports of satisfaction with FamilyRelations, and to reported level of Positive Home Mood.These results are consistent with the zero-order correlations

reported in Table 1. In the case of Marital Satisfaction, theoverall F ratio did not achieve significance and this findingmust be interpreted cautiously. Although Intrinsic Job Gratifi-cation did not show a direct association with family adjust-ment, it was a significant predictor of job mood (F= 7.36; d.f. =1,34; p~.05), with the overall F value for this regressionanalysis almost achieving significance at the .05 level.

These findings are consistent with the hypothesis that forwomen in low-status occupations, satisfaction with intrinsicaspects of work influences their family adjustment throughjob-related mood (Piotrkowski, 1979). Nevertheless, we can-not rule out the hypothesis that influences flow from home towork, an alternative that must be considered in future research.No support was found for the hypotheses that low-statuswomen compensate for lack of gratification at work with in-creased satisfaction with their family roles or that, for them,work and family are entirely segmented.

For the low-status group, a significant negative relationshipemerged between Salary and reported satisfaction with FamilyRelations, a result consistent with their zero-order correlation.This finding lends some support to Safilios-Rothschild’s (1976)hypothesis that a woman’s income may be especially salient forfamily adjustment. In contrast, no significant associationbetween Salary and Family Relations emerged for the high-status group (see Table 4). This interaction between Status andSalary achieved significance (F= 6.34; d.f. = 1,91; p ~.05~. Thedifference between the groups may be attributable to the

differing meanings that women’s salaries have for the familiesof high- and low-status women. In the families of high-statuswomen, 85.7% of families earned $31,000 or more annually(see Table 1). In the low-status group the majority of familiesearned $30,000 or less per year. It may be that in lower-incomefamilies, wives’ increased economic success threatens their



139

husbands’ perceptions of themselves as adequate breadwin-ners. In an affluent family a woman’s income may pose no suchthreat to her husband. A limitation of this interpretation isthat, theoretically~ a wife’s income should pose a threat to themarital relationship. Yet women’s incomes were not signifi-cantly related to our measure of marital adjustment. Possibly,women’s increased incomes in less affluent families may enablethem to report general-if not marital-dissatisfaction morefreely. The relationship of women’s salaries to their familyadjustment requires further exploration.

In the high-status group a somewhat different pattern ofsignificant associations emerged (see Table 4). Women’sreported Marital Satisfaction appeared immune to workinfluences. However, work variables were significantly relatedto the other two measures of family adjustment. As with thelow-status group, Positive Job Mood was positively associatedwith reported positive mood at home. Satisfaction with Job,Security and Intrinsic Job Gratification each were predictorsof high-status women’s reported satisfaction with FamilyRelations. As in the case of the low-status group, Intrinsic JobGratification also was positively associated with level ofPositive Job Mood (F = 4.96; d.f. = 1,53, p ~ .05), with theoverall F value just missing significance at the .05 level. Again,this finding suggests the indirect influence of job satisfactionon family adjustment.9

This pattern of results for the high-status group is consistentwith the hypothesis that job satisfaction with both intrinsic andextrinsic features of work life influences women’s familyadjustment. However, an additional finding not consistentwith this interpretation emerged as well: For the high-statuswomen, Positive Job Mood was negatively associated withsatisfaction with Family Relations. Alone, this finding wouldsuggest a conflict between work enjoyment and family adjust-ment in this group. However, the overall pattern of findingssuggests an alternative interpretation. For the professional ormanagerial woman, job satisfaction may contribute generallyto her family adjustment. However, in comparison with aparticularly exciting day at work, the family realm may appear



140

mundane and relatively unsatisfying. Moreover, her pleasurewith her work day may imply continued emotional andintellectual involvement with the occupational tasks of the day,causing conflict with the necessity of doing chores, attendingto squabbling children, and the like. The existence of dailywork-family conflict in the context of a generalized carry-over of satisfaction with one’s work into the family realm mayreconcile the apparently contradictory propositions advancedfor the relationship between work gratification and familyrelations among professional and managerial groups.The interaction between Positive Job Mood and Status

achieved significance for women’s reported satisfaction withtheir family relations (F = 8.36; d.f. = 1,91; p~ .01). For the groupof high-status women, Positive Job Mood was negativelyrelated to Family Relations; for the low-status women theassociation was positive. For a woman in a low-prestige jobthat may not be intrinsically interesting, feelings of pleasure atwork may not imply continuing involvement with the worktasks of the day. Rather, for her the pleasurable feelings atwork may contribute to positive family involvement.

Other apparent differences between the two groups alsodeserve comment. Job Security emerged as a significantpredictor of Family Relations for the high-status-, but not thelow-status women. The women in low-prestige occupations feltmore satisfied with their job security than the high-statuswomen (t = 1.77; d.f. = ~7; p ~ .~8~, a difference that approachedsignificance. The low-status women were well educated andmay have felt secure in the knowledge that they could remain intheir jobs or find other, similar ones. Since they were relativelysatisfied, their job security may not have been salient for theirfamily lives. On the other hand, the young, high-status womenof the sample were employed in jobs with career ladders, so thatjob security was associated with satisfaction with opportunitiesfor job advancement (r = .55; p ~ .001). Thus, for them,concerns about job security may have been particularly salient.This difference between the groups must be interpretedcautiously, as no significant class interaction effect for JobSecurity and Family Relations emerged in the overall analysis.



141

Sample size limitations also may have attenuated possibleassociations in the low-status group. Similarly, the apparentdifference between groups for Intrinsic Job Gratification as a

predictor of Family Relations was not reflected in the test forinteraction effects. Further research on larger samples isneeded to determine whether or not these apparent groupdifferences are reliable.

Although they must be interpreted cautiously, the nonsig-nificant findings are as noteworthy as the significant ones. Oc-cupational Status did not emerge as a significant predictor offamily adjustment in the regression analysis (see also Richard-son, 1979). Time at work also was not directly associated withwomen’s reported family adjustment (see also Clark et al.,1978). There are several possible explanations for this latterfinding. Despite women’s reports that they find the combina-tion of work and family roles burdensome, they also aresocialized to place their family needs before their own.

Consequently, when time is scarce, their own personal needsmay go unmet. Amount of time spent working also may be toogross an indicator of how work time influences familyadjustment. Time itself is a complex variable and includesdimensions such as timing of work hours, synchronization ofwork and family life-cycle tasks, and so on (Aldous, 1969).Working an eight-hour evening shift may have significantlydifferent effects on family adjustment than eight hours of workduring the day (see Mott et al., 1965). Moreover, the influenceof time on family adjustment may be indirect. The significantzero-order correlation between time spent at work and Intrin-sic Job Gratification in the high-status group (see Table 2)suggests that work time may influence women’s family adjust-ment indirectly, through its effect on their job satisfaction.Future research efforts need to be directed at examining thecomplexities of time as they may influence family adjustment.The lack of generally significant relationships between work

variables and women’s reports of their marital satisfaction isnoteworthy in light of the emphasis on marital adjustment inthe literature on dual-earner families. There are several

possible explanations of this finding. Our indicator of marital



142

adjustment may have been inadequately measuring thoseaspects of the marital relationship that are sensitive to women’sjobs. We also tested the possibility that rote moderatedthe relationship between job satisfaction and marital adjust-ment (see Ridlcy, 1973). Examining respondents’ scores on twoitems measuring commitment to the work role, we found thatover 85% of the sample reported that they definitely orprobably would continue working, even ii there were nofinancial need to do so. Thus, the lack of strong significantrelations between work satisfaction and marital adjustmentdid not appear to be due to the unimportance to these womenof their employment role. Some research evidence suggeststhat marital satisfaction is more sensitive to husbands’ than to

wives’job satisfaction and that both husbands’ and wives’ workroles must be considered (Bailyn, 1970; Ridle~ 1973). Togetherwith the results reported here&dquo; these studies suggest that themarital relationship is especially sensitive to husbands’ jobs,while family relations in general and relations with children aresensitive to wives’ jobs. Future research needs to consider thedifferential sensitivity of familial subsystems to husbands’ andwives’ employment roles.

~

CONCLUSION

The results of this study suggest that women’s jobs aresignificantly associated with their reported family adjustmentin dual-earner families. Despite the lack of significant differ-ences in family adjustment in the groups of high- and low-status women sampled, the data analyses revealed significantwork-family relationships within groups. Indicators of jobsatisfaction and job-related mood emerged as particularlyimportant predictors of reported family adjustment. Althoughpossible areas of work-family conflict were identified in bothgroups, the results generally support the hypothesis thatwomen’s experiences of their jobs &dquo;spill over&dquo; into the familyrealm. Such spillover was not uniform, for women’s maritaladjustment appeared to be relatively immune to the work



143

variables considered. Little support was found for the hypo-thesis of compensatory relations between work and family forthe low-status women studied.Not all work variables were measured in this study, and it is

possible that future research including measures of additionaljob characteristics and experiences will modify the pattern offindings. It should be noted, too, that the data are consistentwith influences that flow from the family to the work place.Social desirability factors and common method variance alsomay have inflated the associations between job and familyadjustment measures. Further research is needed to determinethe reliability and generality of the results reported here, theassociation of job characteristics not included in this studywith women’s family adjustment, and the causal direction ofobtained relationships.

Although we lacked specific data on the family adjustmentof the husbands of the women studied, the findings are

suggestive for further research in this area. The overridingconcern with the effects of wives’ employment on the maritalrelationship may obscure other aspects of husbands’ and wives’family adjustment that may be more sensitive to women’s jobs.Our finding suggests that women’s moods at home and theirsatisfaction with family relations were associated with their jobexperiences, and that husbands’ moods- and general familyadjustment would be associated indirectly with their wives’ jobs.The minimal relationship between job variables and marital ad-justment in our sample of work-committed women suggests thathusbands’ marital adjustment may be relatively independent oftheir wives’ work experiences. Instead, we would hypothesizethat husbands’ marital adjustment is associated with their ownjob satisfaction, which indirectly influences their wives.

Whereas the link between wives’ subjective work experi-ences and husbands’ family adjustment may be indirect,objective work variables may be directly linked to husbands’adjustment. For example, one interpretation of the negativeassociation between salary level and general family satisfactionin the sample of low-status women is that salary has an indirectimpact on wives through its direct effect on their husbands.



144

Thus, the nature of family roles may mediate the relationshipsbetween work roles and family adjustment (see Mortimer,1980). In order to develop an adequate theory of work-familyrelations in dual-earner families, systematic research into suchdirect and indirect pathways of influence is needed. The resultsof this study demonstrate the utility of within-group analysesto investigate the complex processes linking the specificfeatures of wives’ occupational roles to family adjustment indual-earner families.

NOTES

1. The term dual-earner is used instead of dual-worker because this latter termobscures the unpaid housework performed primarily by women.

2. See Faye J. Crosby, Relative Deprivation and Working Women (forthcoming)for a complete description of the sample, questionnaire, and results.

3. Respondents were asked, "Approximately how much do you earn yearly?" Thetwelve available response categories ranged from "nothing" to "over $50,000." Thesecategories were treated as interval data for purposes of analysis.

4. Alpha coefficients were computed from data on all employed married men andwomen in the larger study (N = 224).

5. The Intrinsic Job Gratification scale was composed of six scaled items such as,"Taking everything into account, how satisfied would say you are with your presentjob?" Responses ranged on a five-point scale from "very satisfied" to "verydissatisfied." The Job Security scale was composed of four scaled items with questionssuch as, "Within this last year, how often have you felt some sense of grievanceconcerning job security?" Responses ranged on a five-point scale from "always" to"never." The Positive Job Mood scale was derived from eleven summed items selectedfrom the Roseman Mood List (see Crosby, forthcoming). Participants responded tothe question, "Think for a second about the last two days at work (yesterday andtoday). Which of the following emotions did you feel at any time during the last twodays while at work?" by endorsing any of 38 adjectives describing mood states (such ashappy, sad, proud). A comparable negative job mood score was used as a predictor ofPositive Job Mood and a residual score for each respondent obtained, in order tocontrol for a possible generalized tendency to report feelings. The residual score wasused in the analysis.

6. The Marital Satisfaction scale included items such as, "Everything considered,how happy has your marriage been for you?" Responses ranges on a six-point scalefrom "nearly perfect" to "extremely unhappy." One item on the Family Relations scaleasked, "Now, taking everything into account how would you say you feel about yourfamily life?" Responses ranged on a five-point scale from "very satisfied" to "verydissatisfied."

7. See Note 5 above.8. Age was highly correlated with the presence or absence of children (r = .61; p

&prcue;.001 and r= .50; p&prcue;.001; low-and high-status groups, respectively), and household



145

income with salary (r = .52; p&prcue;.001 and r = .64; p&prcue;.001; low-and high-status groups,respectively).

9. The lack of significant associations between Job Security and reported jobmood suggests that the measure of mood does not reflect a general tendency to reportpositive or negative job attitudes.

REFERENCES

Aldous, J.1969 "Occupational characteristics and males’ role performance in the family."

J. of Marriage and the Family 31 (November): 707-712.Bailyn, L.

1970 "Career and family orientations of husbands and wives in relation to maritalhappiness." Human Relations 23 (April): 97-113.

Blood, R. O., Jr. and D. M. Wolfe1960 Husbands and Wives. New York: Macmillan.

Booth, A.1979 "Does wives’ employment cause stress for husbands?" Family Coordinator

28 (October): 445-449.Burke, R. J. and T. Weir

1976 "Relationship of wives’ employment status to husband, wife and pairsatisfaction and performance." J. of Marriage and the Family 38 (May):279-287.

Clark, R. A., F. I. Nye, and V. Gecas1978 "Work involvement and marital role performance." J. of Marriage and

the Family 40 (February): 9-22.Cohen, J. and P. Cohen

1975 Applied Multiple Regression/ Correlation Analysis for the BehavioralSciences. Hillsdale, NJ: Erlbaum.

Crosby, F. J.Forthcoming Relative Deprivation and Working Women. New York: Oxford

Univ. Press.

Davis, J. A.1974 Codebook for the Spring 1974 General Social Survey. Chicago: University

of Chicago.Dubin, R.

1956 "Industrial workers’ worlds." Social Problems 3 (January): 131-142.Glenn, N. D. and C. N. Weaver

1978 "A multivariate, multisurvey study of marital happiness." J. of Marriageand the Family 40 (May): 269-282.

Gover, D. O.1963 "Socio-economic differential in the relationship between marital adjust-

ment and wife’s employment status." Marriage and Family Living 25(November): 452-457.

Harrell, J. and C. Ridley1975 "Substitute child care, maternal employment and the quality of mother-

child interaction." J. of Marriage and the Family 37 (August): 556-564.



146

Heckman, N., R. Bryson, and R. Bryson1977 "Problems of professional couples: a content analysis." J. of Marriage and

the Family 39 (May): 323-330.Hoffman, L. W.

1963 "Mother’s enjoyment of work and effects on the child," pp. 95-105 in F. I.Nye and L. W. Hoffman (eds.) The Employed Mother in America. Chicago:Rand McNally.

Holmstrom, L.1972 The Two-Career Family. Cambridge, MA: Schenkman.

Jorgensen, S.1979 "Socioeconomic rewards and perceived marital quality. A re-examination."

J. of Marriage and the Family 41 (November): 825-835.Kanter, R. M.

1977 Work and Family in the United States: A Critical Review and Agenda forResearch and Policy. New York: Russell Sage.

Kemper, T. D. and M. L. Reichler1976 "Work integration, marital satisfaction, and conjugal power." Human

Relations 29 (October): 929-944.Locksley, A.

1980 "On the effects of wives’ employment on marital adjustment and com-panionship." J. of Marriage and the Family 42 (May): 337-346.

Marks, S. R.1977 "Multiple roles and role strain: some notes on human energy, time and

commitment." Amer. Soc. Rév. 42 (December): 921-936.Mortimer, J. T.

1980 "Occupation-family linkages as perceived by men in the early stages ofprofessional and managerial careers." Research in the Interweave of SocialRoles: Women and Men 1: 99-117.

Mott, P. E., F. C. Mann, Q. McLoughlin, and D. P. Warwick1965 Shift Work: The Social, Psychological and Physical Consequences. Ann

Arbor: Univ. of Michigan Press.National Commission on Working Women

1979 National Survey of Working Women: Perceptions, Problems and Pro-spects. Washington, DC: National Manpower Institute.

Nock, S. L.1979 "The family life cycle: empirical or conceptual tool?" J. of Marriage and

the Family 41 (February): 15-26.Nye F. I.

1974 "Husband-wife relationship," pp. 186-206 in L. Hoffman and F. Nye (eds.)Working Mothers. San Francisco: Jossey-Bass.

1963 "Marital interaction," pp. 263-281 in F. I. Nye and L. W. Hoffman (eds.)The Employed Mother in America. Chicago: Rand McNally.

Orden, S. R. and N. M. Bradburn1969 "Working wives and marriage happiness." Amer. J. of Sociology 74

(January): 391-447.Piotrkowski, C. S.

1979 Work and the Family System. New York: Macmillan.Rapoport, R. and R. Rapoport

1971 Dual-Career Families. New York: Viking.



147

1965 "Work and family in contemporary society." Amer. Soc. Rev. 30 (June):381-394.

Richardson, J. G. 1979 "Wife occupational superiority and marital troubles. An examination of

the hypothesis." J. of Marriage and the Family 41 (February): 63-72.Ridley, C. A.

1973 "Exploring the impact of work satisfaction and involvement on maritalinteraction when both partners are employed." J. of Marriage and theFamily 35 (May): 229-237.

Safilios-Rothschild, C.1976 "Dual linkages between the occupational and family systems: a macro-

sociological analysis." Signs 1, 3 (part 2): 51-60.1970 "The influence of the wife’s degree of work commitment upon some aspects

of family organization and dynamics." J. of Marriage and the Family 32(November); 681-691.

Scanzoni, J.1970 Opportunity and the Family. New York: Macmillan.

Staines, G. L., J. H. Pleck, L. J. Shephard, and P. O’Connor1978 "Wives’ employment status and marital adjustment." Psychology of

Women Q. 3 (Fall): 90-120.Sweet, J.

1973 Women in the Labor Force. New York: Seminar.

Walshok, M. L.1979 "Occupational values and family roles: women in blue-collar and service

occupations," pp. 63-83 in K. W. Feinstein (ed.) Working Women andFamilies. Vol. 4. Beverly Hills, CA: Sage.

Wright, J. D. 1978 "Are working women really more satisfied? Evidence from several national

surveys." J. of Marriage and the Family 40 (May): 301-314.Yarrow, M. R., P. Scott, L. deLeeuw, and C. Heinig

1962 "Child-rearing in families of working and non-working mothers." Soci-ometry 25 (June): 122-140.

Young, M. and P. Willmott1973 The Symmetrical Family. New York: Random House.

Chaya S. Piotrkowski (Ph.D., University of Michigan) is Assistant Professor ofPsychology at Yale University. She also is on the faculties of the Institute forSocial and Policy Studies and the Bush Center on Child Development and SocialPolicy at Yale. She recently published a book on work and family life in working-class families. Currently, she is conducting research on the relationship of parentaljob stress to family dynamics and children’s mental health.

Paul Crits-Christoph is a doctoral student in clinical psychology in the PsychologyDepartment at Yale University.



Women’s Jobs and Family Adjustment*

Documents

Transcript of Women’s Jobs and Family Adjustment*