Sex Differences in Absence from Work: A Reinterpretation

Sex differences in absence from work were investigated for parents and nonparents during a period of 11 months. Personnel records showed sex differences only among working parents, with mothers taking significantly more sick leave than fathers (p < .05). By using self-reports obtained under protection of anonymity it was found that in the parent group both sexes reported absences necessitated by child care as personal illness, but significantly more women than men were absent from work for such activities (p < .01). Results from this study shed light on the statement that “women are more likely than men to be absent from work because of illness” (U.S. Department of Labor Bureau Bulletin, 1977); child care, rather than personal illness, appears to be the major variable that mediates sex differences in absence from work.

The clarification of variables leading to absence from work is a crucial issue because, as O' Leary (1974) has suggested, one of the prevalent beliefs that might adversely affect women in the work force is the one that women require an inordinate amount of sick leave due to menstruation and pregnancy. Although research on sex differences in absence from work has yielded conflicting results (Hedges, 1973(Hedges, , 1977Hedges, Taylor & Hoyle, 1979;Loring & Wells, 1972) the U.S. Department of Labor (1977) reported that "women are considerably more likely than men to be absent from work because of illness" (p. 55).
Since it has been reported that single women have the same rate o f absence from work as single men and that greatest sex differences in Requests for reprints may be addressed to Paula Englander-Golden, Univcrsity of Oklahoma, Department of Human Kelatiorrs, 60 1 Elm, Norman, Oklahoma 7:10 19.
Psychology of Women Quarterly, Vol. 8(2) Winter 19x3 0 1983 Human Sciences Press 18.5 such absences are found for the age group 25-34 years (Hedges, 1973(Hedges, , 1977, the present study was designed to see whether sex differences in absence from work could result from a discrepancy between role demands faced by working mothers and fathers. Research has consistently documented such a discrepancy even in households in which the mother works a 40-hour week (Adams & Winston, 1980). Taking care of a sick child or taking a child to a doctor/dental appointment is not usually a cornpensable absence; therefore, a parent may report such an absence as personal illness rather than child care activities. Although both sexes may use personal illness as an excuse for such absence, this practice would inflate female sick leave absences more because of the greater amount of child care responsibilities women assume. Following this line of reasoning, we predicted that (1) sex differences in absence from work would be evident for parents but not for nonparents; and ( 2 ) among working parents, women would more frequently be absent from work because of child care activities.

METHOD
The study was conducted at a large corporation in the southwestern part of the United States. It was limited to persons who were required to keep daily and weekly time sheets (hourly and salaried nonexempt employees) and who were eligible for sick leave. Participation was voluntary arid anonymous. Employees were selected at random from personnel records. The following four categories were formed: 49 women and 47 men with children; 47 women and 47 men without children. Most of the women were clerks and secretaries and most of the men were clerks, accountants, machine operators, and security force. According to the corporation, the monthly income of the men was about $5 1.00 higher than that of the women.
Personnel records were used to obtain the four respondent categories, their job classification, and the number of paid personal sick leave hours these employees took during the last 1 1-month period.
A questionnaire asked respondents their sex, how many children were living at home, the children's ages, and whether the respondent had to be absent from work due to a child's illness, injury, or doctorldental appointment in the last 11month period. Those who answered "yes" were asked whether they took paid personal sick leave for these child care activities. In addition women were asked how many hours of paid sick leave they used during the 1 1-month period due to menstruation and whether they had been pregnant during this time frame.
The questionnaire was returned by 45 women with children living at home (average number of children was 1.62 and their average age was 8.16 years), 38 men with children living at home (average number of children was 1.68 and their average age was 8.52 years), 45 men without children, and 40 women without children. During some part of the 1 I-month time frame of the study five women with children and four women without children had been pregnant.

RESULTS
The records indicated that the women without children averaged 42.49 hours of paid sick leave compared to 35.21 hours averaged by the men without children. As predicted, this difference was not statistically significant (t (92) = 1.03, n.s.). However, the women with children averaged 50.48 hours compared to 34.3 hours averaged by the men with children. This difference was statistically significant (t (94) = 2.59, p It is important to notice that the above results were obtained even though the group of women with children included at least five women who had been pregnant during some part of the study's time frame and the group of women without children included at least four pregnant women (as determined on returned questionnaires). This suggests that pregnancy is not an important variable in the obtained sex differences in absence from work.
On the questionnaire 30 of the 45 women with children and six of the 38 men with children reported that a child's illness or injury or doctor/dental appointment necessitated their absence from work during the 11 months covered by the study. As predicted, this difference was also significant (x2 (1) = 19.71,p < .01). Both men and women tended to report these absences as personal illness, i.e., five out of the six men and 24 out of the 30 women (x2 (1) = .14; n.s.) did so.
Menstruation appeared to be insignificant as a cause of absence from work. The 40 women with children who had not been pregnant during any part of the time frame of the study reported 1.8 hours of sick leave related to menstruation during the 1 1-month period; the 36 women without children who had not been pregnant reported 2.97 hours during the same period.

DISCUSSION
The results obtained in this study shed light on the nature of sex differences in absence from work. Differences did not appear among nonparents. However, among parents, significantly more mothers than fathers were absent from work because of child care activities. Although both sexes tended to report such absences as personal illness, mothers' greater amount of child care responsibility yielded the spurious result that women have more absences due to illness.
This study indicates the need for a large-scale investigation of' child care among working parents. Should the present results be corroborated across the United States, policies pertaining to compensable absences might be changed to include caring for a sick child or taking a child to a doctor/dental appointment. Such a change in policy may still leave women with a larger share o f child care responsibilities; however, it may stop the perpetuation of myths pertaining to women's absence from work. Dissipating the myths may alleviate the stress and guilt often found among working mothers (Hoffman, 1974) as well as alleviate the penalties associated with absenteeism, i.e., reduced wages, reduced chances for promotion, and increased chances for dismissal (Allen, 1978).