The results of a study of 456 employees in an American service organization.
Because most people have multiple roles, such as parent and worker, interrole conflict is an important issue. (Interrole conflict is the conflicts between the expectations of one person’s roles. Identifying with one’s job — “job involvement” — can increase that conflict; while mentally walling off parts of one’s identity (“segmentation”) can reduce it. At least, that was our thought — so we tested to see if that was true.
Interrole conflict is usually defined as “incompatibility between the role expectations of different roles” (Frone & Rice, 1987, p. 46). For example, Greenhaus and Beutell (1985) defined interrole conflict as “a form of role conflict in which the sets of opposing pressures arise from participation in different roles.…when pressures arising in one role are incompatible with pressures arising in another role.” (p. 77)
Why should we care about interrole conflict? It’s been related to lower sales income (1987); work overload (Cooke & Rousseau, 1984); stress (Barling and MacEwen, 1991); increased cognitive diffulties (Barling & MacEwen, 1991); impaired martial functioning (Barling, 1986; Blood & Wolfe, 1960); and family conflict (Wiersma & van den Berg, 1991).
Interrole conflict was also related to lower satisfaction with the job (Cooke & Rousseau, 1984; Jackson & Schuler, 1985; Jones & Butler, 1980; Pleck, Staines, & Lang, 1980; Staines & O’Connor, 1980; Wiersma and van den Berg, 1991); with the family; (Kopelman, Greenhaus, and Connolly, 1983; Pleck et al., 1980; Staines & O’Connor, 1980); and with the role of working mother (Barling & MacEwen, 1991).
There are some interesting aspects of inter-role conflict:
- Peters and Cantrell (1993) found that, among working women, the most common reported cause was their parents’ disapproval of their life choices.
- Mednick (1987) found significant correlations of interrole conflict with family affect, family worries, family conflict, age and number of children, and job and work involvement. There were negative correlations between interrole conflict and family social support, sales, and job tenure (but not age).
- O’Driscoll, Ilgen, and Hildreth (1992) found that, while time invested in job-related activities was related to interference between job demands and off-the-job life, off-job time demands were not.
Job involvement, or “psychological identification with a job” (Kanungo, 1982, p.97), implies that a job-involved person sees her or his job “as an important part of his self-concept” (Lawler & Hall, 1970, p. 311), and that jobs “define one’s self-concept in a major way” (Kanungo, 1982, p. 82).
Job involvement (or lack of it) has been clearly linked to absenteeism (e.g., Blau, 1986; Farrell & Stamm, 1988; Shore, Newton, & Thornton, 1990; Scott & McClellan, 1990), and to turnover or intent to leave (e.g., Baba & Jamal, 1991; Huselid & Day, 1991; Ingram, Lee, & Lucas, 1991; Shore, Newton, & Thornton, 1990).
The most well-documented correlate of job involvement is job satisfaction (e.g., Baba & Jamal, 1991; Elloy, Everett, & Flynn, 1991; Gerpott, 1990; Mathieu & Farr, 1991; Paterson & O’Driscoll, 1990; Shore, Newton, & Thornton, 1990).
If one has high job involvement, the job becomes part of one’s identity and is more likely to cross situational boundaries — reducing segmentation of the job role. In short, job involvement should reduce segmentation and increase interrole conflict.
Job involvement may also directly affect interrole conflict. Job involvement may also be a direct factor if the question is not merely how much time there is, but who has priority over it: e.g., the job or the family. Thus, a further question is whether job involvement explains any variance in interrole conflict, after time, attention, and segmentation have been considered and partialed out.
One final question is which factor contributes the most unique variance to interrole conflict, which has direct relevance for those who wish to lower the amount of interrole conflict in their (or others’) lives. For example, if time contributes the most unique variance, people could reduce their conflict by cutting back on their hours or honing their time management skills. If segmentation contributes the most unique variance, people could try to separate their roles further.
The sample was the full staff of a service company with locations across the United States, one of which was recently acquired company. In the responding sample, the average age was 39 years, 70% were female, 68% were married, and 56% had no children.
One survey was sent to each full-time employee via interoffice mail on July 18, 1994, with a self-addressed, stamped envelope and a letter from a high-ranking contact.
Job involvement was measured with Kanungo’s (1982) JIQ. As for interrole conflict, there were no published methods that explored the full concept; Kopelman, Greenhaus, and Connolly (1983) focused on time and work demands, for example. The Peters and Cantrell (1993) method was too time-consuming. We ended up creating new measures of interrole conflict and, since there was none, segmentation.
New items were created based on definitions and any available sample questions. Three psychology graduate students, working separately, checked the face validity of each item; items and the instructions were modified or deleted accordingly, then returned for another round. After these students approved the items, three other students were asked to classify each item, and misclassified items were dropped. The scales were then presented to three different students for face validity checking.
Finally, the measures were then given to a pilot sample of working students, then to full-time employees at small companies. Items with an inter-item correlation of below .30 were revised or replaced. The measures were then given to the final sample, after which two items in the segmentation scale with an inter-item correlation of below .30 were removed.
The format of the interrole conflict items was based on Holahan and Gilbert’s (1979) measure; some of the items were identical or similar to Holahan and Gilbert’s. Time devoted to roles was measured using items similar to those used by O’Driscoll, Ilgen, and Hildreth (1992). Attention devoted to roles was measured by asking people whether they thought about their job, parent, spouse, and various “leisure activities” roles “rarely, sometimes, often, or continuously.” (Two other sets of items were judged as being less effective by the reviewers). The scales, with inter-item correlations for each item, are in the appendix.
The attention and interrole conflict scales were only partially filled out by people who did not have children or a spouse; in these cases, missing values were recoded as the minimum — the amount of attention they were most likely to have given to roles they did not have. Otherwise, the responding sample would have consisted only of married parents. The interrole conflict scale, which used Likert-type items, was calculated as a mean of completed items, for those who filled out most or all of them.
When the scales were factor analyzed, we found that one item from the JIQ was misclassified, and one item from the JIQ was spread across more than one factor (this was the case in past studies as well). Otherwise, all the items from the scales were correctly classified.
Reliabilities and descriptive statistics
|Interrole conflict (sum)||10||.90||259||30.5||11.7|
|Interrole conflict (mean)||>5||.85||457||3.0||1.2|
We checked for nonlinear relationships by using scatterplots and testing polynomial equations. None of these tests found a need for nonlinear analysis.
|Hours/week devoted to roles||M||s.d.||Interrole
|Family||35.8||16.5||– .01||– .12*|
|Other||17.3||12.9||– .08||– .18**|
|Attention devoted to roles (scale from 1 to 6)|
|Segmentation||22.0||7.5||– .28**||– .42**|
|* p ≤ .05 ** p ≤ .01|
Of the time and attention items, time and attention devoted to the job had the strongest relationship with conflict. Job involvement had a small but significant direct relationship with interrole conflict. The strength of the relationship between job involvement and the moderator variables, and between the moderators and interrole conflict, was stronger.
Job involvement contributed no significant, unique variance to conflict after controlling for the effect of either time or attention devoted to the job.
The unique contributions of each moderator to conflict were tested (see Table 3). The total variance contributed by each variable, including variance explained by other variables, is described as “total R2.”
Unique variance in conflict contributed by four variables
Unique R2 Total R2 Segmentation .04 .08 Time devoted to the job role .02 .04 Attention to the job role .01 .04 Job involvement .00 .01
The basic hypotheses were (1) job involvement directly affects interrole conflict; (2) job involvement increases time and attention devoted to the job, thereby causing conflict; and (3) job involvement prevents segmentation of roles, thereby increasing conflict. The latter two hypotheses were supported.
Time and attention as moderators
Job involvement was related to time and attention given to the job role, though the relationship with time was weak.
Prior research focused on time as the primary cause of interrole conflict (e.g., Greenhaus & Beutell, 1985). In this study, time devoted to the job role was related to conflict, but the relationship was slightly weaker than that of attention and conflict. The total amount of time devoted to all roles was not related to interrole conflict.
Frone and Rice (1987) wrote that the effects of time and attention demands may be higher for parents, as the parent role has high, inflexible time and attention demands. For those with children, that did not turn out to be the case —the correlation was weaker than for the general sample — but that may have been due to restriction of range.
Segmentation as a moderator
It was predicted that job involvement would prevent segmentation, because if a person identifies with their job, they must carry it with them (as part of their identity) at all times. The data supported this prediction. Because of this, there is also some support for the idea that job involvement is a matter of identity.
While segmentation may lower the conflict people feel between their roles, those who have high job involvement may not be able to use segmentation globally. If, however, people can leave incompatible job expectations at work (and retain compatible expectations), then interrole conflict might be reduced without reducing job involvement.
A direct relationship
If job involvement affects interrole conflict only by increasing time and attention given to certain roles, one would expect that job involvement would contribute no additional variance to conflict. A multiple regression showed that this was, indeed, the case. Job involvement contributed very little to the variance in interrole conflict even when no other variables were included.
Frone and Rice (1987) also wrote that role involvement may only increase conflict for those who have a role with inflexible demands (such as parent). Job involvement was not related to conflict among parents, but this may have been due to the considerably smaller sample size, or the varying ages of the children.
Predictors of conflict
The most substantial impact on conflict was made by segmentation. This implies that helping people to improve their segmenting skills may help to reduce their conflict.
Time and attention accounted for similar amounts of the variance in interrole conflict (4%) when examined separately. However, as part of an equation including all moderator variables and job involvement, time devoted to the job role accounted for 2% of the unique variance in conflict; attention to the job role accounted for only 1% of the variance. This may be due to the strong correlation of segmentation and attention to the job role (r = –.54, p ≤.001).
Of the two similar constructs, segmentation and attention, segmentation had a stronger unique impact on interrole conflict. Thus, it seems that when and where attention is given to the job role may be more important than simply how much attention is given to the job role.
Other possible moderators
All of the moderators, taken together, explained only 12% of the variance in conflict. Interrole conflict was not highly correlated with any of the variables in this study, including demographics.
The strongest determinants of conflict may be out of the range of this study. Other causes of conflict may include parental or social approval (Peters & Cantrell, 1993), family and social support, family conflict and affect (e.g. Mednick, 1987), and involvement in other than the job role (Frone & Rice, 1987).
Although this study used theoretical work by Frone and Rice (1987), their research was not focused on the same goals. Frone and Rice, following a paper by Greenhaus and Beutell (1985), sought to find out whether family involvement moderated the relationship between job involvement and work-family conflict. They also looked at whether within-domain (e.g., spouse and parent) and cross-domain (e.g., spouse and job) involvements had different relationships with conflict.
Conflict had a negative correlation with age. This may indicate that people learned to deal with conflict through, say, better time scheduling or delegation of roles. It may also indicate that younger people needed to participate in more roles, or to fill their roles more completely, because they could not delegate (due to their income, job level, or the nature of their roles).
Barling and MacEwen (1991) wrote that women may not place their family role aside while at work. However, there were no significant differences between men and women in either interrole conflict or role segmentation in this study; this was also true when the sample was limited to those with children.
The sample included only workers in a single organization, raising the possibility of sampling artifacts. However, the variety of subcultures in this company’s locations was shown by variety in the scale means. The inclusion of the recently acquired division may have helped to add variety, as well. Generalization beyond North Americans in office jobs may be premature.
Some items were subject to bias; for example, if a person has strong beliefs in family, they may inflate the time they devote to family. Using other measures of time and attention paid to various roles (such as diaries), as well as behavioral scale anchors, would lower the potential for bias and errors.
The role measures examined the conflict or segmentation of role pairs. Some role pairs may have different dynamics from others, so that it is safest to generalize from this study only to job-family, job-spouse, or job-self conflict. It is possible that there are different types of segmentation, such as job-parent, spouse-parent, etc., which have different characteristics and different relationships with types of involvement and interrole conflict.
One of the more important findings of this study is that job involvement may interfere with segmentation. This is important not only because of the issue of interrole conflict, but because it implies that job involvement is, as it is defined, identification with the job role. Generally, research and theory involving job involvement has not centered on identity, but on matters such as motivation (e.g. Kanungo, 1982), socialization (e.g. Lodahl & Kejner, 1965), and job characteristics (e.g. Hackman & Lawler, 1971). Treating job involvement as a matter of identity may help to clarify its relationship with other constructs. Some new questions arise; for example, if one relies on the job role for one’s identity, what happens to a person when their job role is taken away through layoffs or retirement?
The concept of segmentation has been largely neglected. Some questions for future research include whether segmentation is a trainable skill, and whether there are negative consequences.
The study of interrole conflict may benefit by following the lead of Peters and Cantrell (1993) in using qualitative measures to determine the causes of interrole conflict. The relationship of interrole conflict to role performance may also be explored further, with more powerful instruments.
This study was designed to investigate three possible moderators between job involvement and interrole conflict: segmentation and time and attention devoted to the job. While all three appear to be moderators, segmentation explained the most variance in interrole conflict. Job involvement and interrole conflict do not seem to be directly related.
Segmentation and the other coping mechanisms discussed here are fertile grounds for theory and practice. Many other ideas (such as delegation) are yet to be rediscovered and used.
While this study made a contribution to explaining causes of interrole conflict, the amount of variance in conflict explained by all the variables in this study was still relatively small. There are many opportunities for future research in this area.
- Baba, V.V., and Jamal, M. (1991). Type A behavior, prevalence, and consequences. Human Relations, 44, 1213-1228.
- Barling, J. (1986). Interrole conflict and marital functioning amongst employed fathers. Journal of Occupational Behaviour, 7, 1-8.
- Barling, J., and MacEwen, K.E. (1991). Maternal employment experiences, attention problems and behavioral performance: A mediational model. Journal of Organizational Behavior, 12, 495-505.
- Blau, G.J. (1985). A multiple study investigation of the dimensionality of job involvement. Journal of Vocational Behavior, 27, 19-36.
- Blau, G.J. (1987). Using a person-environmental fit model to predict job involvement and organizational commitment. Journal of Vocational Behavior, 30, 240-257.
- Blau, G.J. (1989). Testing the generalizability of a career commitment measure and its impact on employee behavior. Journal of Vocational Behavior, 35, 88-103.
- Blood, R., and Wolfe, D. (1960). Husbands and Wives. New York: Macmillan.
- Brooke, P.P., Russell, D.W., and Price, J.L. (1988). Discriminant validity of measures of job satisfaction, job involvement and organizational commitment. Journal of Applied Psychology, 73, 139-145.
- Cooke, R.A., and Rousseau, D.M. (1984). Stress and strain from family roles and work-role expectations. Journal of Applied Psychology, 69, 252-260.
- Crandall, W.R. (1992). The effect of work hours and scheduling on a restaurant manager’s interrole conflict, life satisfaction, job satisfaction, organizational commitment, and intentions to leave the organization. Dissertation Abstracts International, AAC 9311438.
- Elloy, D.F., Everett, J.E., and Flynn, W.R. (1991). An examination of the correlates of job involvement. Group and Organization Studies, 16, 160-177.
- Farrell, D., and Stamm, C.L. (1988). Meta-analysis of the correlates of employee absence. Human Relations, 41, 211-227.
- Frone, M.R., and Rice, R.W. (1987). Work-family conflict: The effect of job and family involvement. Journal of Occupational Behaviour, 8, 45-53.
- Gerpott, T.J. (1990). Intracompany job transfers: An exploratory two-sample study of the buffering effects of interpersonal support. Prevention in Human Services, 8, 113-137.
- Goode, W.J. (1960). A theory of role strain. American Sociological Review, 25, 483-496.
- Greenhaus, J.H., and Beutell, N.J. (1985). Sources of conflict between work and family roles. Academy of Management Review, 10, 76-88.
- Greenhaus, J.H., Parasuraman, S., Granrose, C.S., Rabinowitz, S., and Beutell, N.J. (1989). Sources of work-family conflict among two-career couples. Journal of Vocational Behavior, 34, 133-153.
- Hicks, W.D., and Klimoski, R.J. (1981). The impact of flextime on employee attitudes. Academy of Management Journal, 24, 333-341.
- Holahan, C.K., and Gilbert, L.A. (1979). Interrole conflict for working women: Careers versus jobs. Journal of Applied Psychology, 64, 86-90.
- Huselid, M.A., and Day, N.E. (1991). Organizational commitment, job involvement, and turnover: A substantive and methodological analysis. Journal of Applied Psychology, 76, 380-391.
- Ingram, T.N., Lee, K.S., and Lucas, G.H. (1991). Commitment and involvement: Assessing a sales force typology. Journal of the Academy of Marketing Science, 19, 187-197.
- Jackson, S.E., and Schuler, R.S. (1985). A meta-analysis and conceptual critique of research on role ambiguity and role conflict in work settings. Organizational behavior and human decision processes, 36, 16-78.
- Jones, A.P., and Butler, M.C. (1980). A role transition approach to the stresses of organizationally induced family role disruption. Journal of Marriage and the Family, 42, 367-376.
- Kanungo, R.N. (1982). Work alienation: An integrative approach. New York: Praeger Publishers.
- Kopelman, R.E., Greenhaus, J.H., and Connolly, T.F. (1983). A model of work, family, and interrole conflict: A construct validation study. Organizational behavior and human performance, 32, 198-215.
- Lawler, E.E. III, and Hall, D.T. (1970). Relationship of job characteristics to job involvement, satisfaction, and intrinsic motivation. Journal of Applied Psychology, 54, 305-312.
- Lodahl, T.M., and Kejner, M. (1965). The definition and measurement of job involvement. Journal of Applied Psychology, 49, 24-33.
- Mathieu, J.E., and Farr, J.L. (1991). Further evidence for the discriminant validity of measures of organizational commitment, job involvement, and job satisfaction. Journal of Applied Psychology, 76, 127-133.
- Merton, R.K. (1957). Social theory and social structure. London: The Free Press of Glencoe.
- O’Driscoll, M.P., Ilgen, D.R., and Hildreth, K. (1992). Time devoted to job and off-job activities, interrole conflict, and affective experiences. Journal of Applied Behavior, 77, 272-279.
- Paterson, J.M., and O’Driscoll, M.P. (1990). An empirical assessment of Kanungo’s (1982) concept and measure of job involvement. Applied Psychology—An International Review, 39, 293-306.
- Peters, D.K., and Cantrell, P.J. (1993). Gender roles and role conflict in feminist lesbian and heterosexual women. Sex Roles, 28, 379-392.
- Pleck, J.H., Staines, G.L., and Lang, L. (1980). Conflicts between work and family life, Monthly Labor Review, 103, 29-32.
- Sarbin, T.R., and Allen, V.L. (1968). Role theory. In G. Lindzey and E. Aronson (Eds.), Handbook of Social Psychology, Vol. 1., Addison Wesley, Reading, MA.
- Scott, K.D., and McClellan, E.L. (1990). Gender differences in absenteeism. Public Personnel Management, 19, 229-253.
- Shore, L.M., Newton, L.A., and Thornton, G.C. (1990). Job and organizational attitudes in relation to employee behavioral intentions. Journal of Occupational Behaviour, 11, 57-67.
- Staines, G.L., and O’Connor, P. (1980). Conflicts among work, leisure, and family roles. Monthly Labor Review, 103, 36-39.
- Thoits, P.A. (1987). Negotiating roles. In Crosby, F.J. (Ed.), Spouse, parent, worker: On gender and multiple roles, p. 11-22, New Haven: Yale University Press.
- Wiener, Y., and Gechman, A.S. (1977). Commitment: A behavioral approach to job involvement. Journal of Vocational Behavior, 10, 47-52.
- Wiersma, U.J., and van den Berg, P. (1991). Work-home role conflict, family climate, and domestic responsibilities among men and women in dual-earner families. Journal of Applied Social Psychology, 21, 1207-1217.
- Zatz, D.A. (1995). Job involvement and interrole conflict. (Doctoral dissertation, Columbia University, 1995).
This article copyright © 1996, David Zatz. Please do not reprint without written permission. For bibliographic references, you may wish to consult the original dissertation from which this was summarized, shown as reference #41 above.