EDITORIAL Working Mothers and Their Children ALBERTA E. SIEGEL, PH.D.
Twenty-five years ago, working in collaboration with two graduate students, Lois Meek Stolz and I studied kindergarten children of working and nonworking mothers (Siegel et aI., 1959). Separately, she reviewed the research studies then extant on the children of working mothers (Stolz, 1960). Subsequently, in collaboration with one of my graduate students, I reviewed the research literature on working mothers (Siegel and Haas, 1963). Maternal employment had been increasing for two decades, as indeed it has continued to increase in the intervening years. The research available at that time amounted to a substantial body of evidence that maternal employment per se is not a very powerful influence on the way children develop. In the past two decades the research literature has grown dramatically, and the conclusion from it is about the same (Hoffman, 1979; Huston, 1983). Some studies show that children of working mothers, especially daughters, are more favorably disposed than others to outside employment for women and to educational achievement. And the children of working mothers are sometimes observed to differ from others in their sex-role attitudes and stereotypes. But in most ways the children of working mothers are not importantly different from their peers. This conclusion is least well documented in the case of infants and very young children, for it is only in quite recent years that their mothers have been in the labor force in very large numbers. One must be careful to speak about the influence of maternal employment per se, because in our world maternal employment is intertwined with other influences, and some of those are quite powerful. For example, mothers who are separated or divorced are typically in the work force, proportionally more commonly than mothers having husbands with jobs who live with them and the children and who support them. There is growing evidence from research that separation and divorce are powerful influences on children's lives. The task for research is to separate the influence
of maternal employment from the influence of separation and divorce. The same can be said about poverty. An unselected group of working mothers in America will include proportionally more poor women than will an unselected group of mothers who do not work outside the home. And poverty has powerful influences on children, because of its links with quality of schooling, housing, nutrition, medical care, and the like. The task for research has been to separate the influence of maternal employment from that of poverty. What may we conclude from the report in this issue of Brodkin et al. (1984) that medical students whose mothers had been employed when the students were young children were lonelier and more distressed in their initial year of medical school than other medical students? Not as much as we would be able to conclude if we knew why these mothers were working, 20 or 30 years ago, when their children were so young. Was'it because of divorce? Was it because ofpoverty? The data on these central questions are not given. We are left wondering whether a more accurate conclusion from their data might be that young adults whose early years were blighted by parental divorce and/or poverty are more vulnerable to loneliness when in a challenging new situation. I can readily understand why many teachers and many physicians believe that mothers should not work outside the home when their children are young. These busy professionals often see children in trouble. Casual inquiry frequently elicits that these children's mothers are in the labor force. The teacher and the doctor can easily leap to the conclusion that it is because the mothers work that the children are in trouble, An inquiry into why the mothers work would be more productive. It could lead the busy clinician to apprehend that parental separation and divorce and parental poverty are linked to the child's troubles, and to appreciate that the mother is working in an effort to cope with those destructive influences on her life and the lives of her children. The task of untangling intertwined forces is not unique to the research worker whose interest is maternal employment. The same task challenges many other research workers in the biomedical and biobehavioral sciences. For example, it faces the research worker seeking to determine how smoking influences
Dr. Siegel is a Visiting Professor of Psychology at the Yale Child Study Center, Yale University. She will return to her position as Professor of Psychology in the Department of Psychiatry and Behavioral Sciences, Stanford University, in September 1984. 0002-7138/84/2304-0486 $02.00/0 © 1984 by the American Academy of Child Psychiatry. 486
WORKING MOTHERS AND THEIR CHILDREN
health, or the one seeking to identify links between diet and disease. Does long-term smoking of cigarettes make a person susceptible to cancer and heart disease? Or do the same factors which induce some people to smoke also make those people vulnerable to disease? The formal structure of this research problem is the same as the formal structure of the problem of maternal employment. There are statistical approaches to it, and there are also ways to design research which bring us closer to an answer than other designs will bring us. These statistical techniques and design approaches have informed research on maternal employment just as they have informed research on smoking. The outcome has been the opposite. When the influence of intertwined forces is controlled or partialled out in studies of smoking and disease, then smoking emerges more starkly as a pathogen. When the influence of intertwined forces is controlled or partialled out in studies of maternal employment, then maternal employment per se emerges as a weak influence. When dealing with such a complex research problem, in addition to relying on studies which are well designed and whose data are analyzed in sophisticated ways, we search for a confluence of evidence from diverse sources. We require a body of research literature, not a single study. We rely on evidence from different disciplines-sociology, psychology, psychiatry, education, social work, anthropology, economics. And we rely on information generated at diverse sources-different universities, different clinics, different nations. Linked to the need for clear research designs and sophisticated statistical analyses is the need for clear thinking. Muddled concepts generate muddled research which yields muddled results. As the computer technicians say, "Garbage in, garbage out." The concept of "maternal deprivation" has proven to be too broad and too inclusive. Progress in empirical research was accelerated when that overinclusive concept was analyzed into useful components by Yarrow (1961). Perhaps most unfortunate is the comingling of maternal death in the same category with maternal employment. No sensible person can doubt that the death of a child's principal caretaker is an event with extraordinarily profound consequences for that child. That is true in any part of the world. These consequences can be mitigated or exacerbated or rechanneled by the way the death is handled in the family and community and by the sequelae in the child's life. But a parent's death is a catastrophe. It is simply preposterous to link that catastrophic event conceptually with the everyday fact that mothers throughout the world have work responsibilities away from their homes.
Throughout the world mothers take care of their children. Throughout the world they share their childcaring responsibilities with other adults in the home and nearby, including sometimes fathers but more importantly grandmothers and other female relatives. They also share child-caring responsibilities with older children. In the third world, the modal caretaker for an infant is a prepubertal sister. She is assisted in her caretaking responsibilities by her siblings and cousins, and she is typically responsible to her mother and grandmother for doing her job well. But a child is the infant's caretaker during most of the infant's waking hours (Kermoian, 1982). With more and more children entering school in third world nations, the child caretaker now may be as young as four or five. In China, the typical caretaker is the grandmother, to whom a young child's care is assigned while the mother works away from home. In the world at large, maternal employment outside the home is the norm, not an aberration (Werner, 1979). Observations of infants and young children in orphanages had a profoundly useful effect on our appreciation of the importance of infancy and early childhood. We must pay honor to the pediatricians, psychologists, and psychiatrists who recognized the importance of the "natural experiment" represented by the orphanages-Skeels, Spitz, Ribble, Bowlby, Provence, Goldfarb, Dennis, Rheingold, Hunt, and others. Through their efforts in gaining access to the orphanages, making careful observations in them, and following up on children who had left them, much was learned. In my judgment, the term "maternal deprivation" has proven useful as a shorthand description of the experience of an infant or child in such a benighted setting. Maternal loss is not the same as life in an orphanage, and subsuming both under the rubric "maternal deprivation" may be mixing apples and oranges. But grouping those two with maternal employment is mixing apples and oranges with string beans! We must find another term to describe the experience of an American child living in a family with two parents, both of whom leave home every weekday to go to their jobs, returning at the end of the day to be with their children. In summary. the general trend of social research in the past 40 years on the topic of working mothers has suggested that maternal employment per se is not very influential in the lives of children. Further, even before we had that research evidence it probably was not a good idea to link maternal death and orphanage life with maternal employment, grouping them all under the general heading of maternal deprivation. That concept is a helpful one, but such links stretch the concept beyond its usefulness.
ALBERTA E. SIEGEL
Larger-scale and more sophisticated research on maternal employment and its effects on children is not necessary. What is needed are more promising lines of investigation. More research is needed on the range and diversity of caretaking arrangements that seem to benefit children. We need to know more about good mothering, whether by an employed mother or an at-home mother, whether by a foster mother, an adoptive mother, a biological mother, or a father. How can children be assured devoted care in a society in which high geographical mobility has isolated many nuclear families from the other adults who might collaborate with the parents in child care? Is long-term stability in caretaking arrangements as central as we believe it to be? What are the best methods for training, selecting, and recruiting the adults who care for children in group settings? There are many reasons why maternal employment is on the rise. Ours is a society in which females routinely go to school and in which adult females have the vote. Increasingly, females are benefiting from higher education. Females possess work skills needed in the economy. We need more and better teachers, nurses, social workers, physical therapists, speech therapists, child care workers for retarded and disturbed children, dental technicians, etc. Those are all professions in which females are the majority and likely to remain so. Females believe they need to work in order to support themselves and their families. This might be arguable in some of the instances in which a wife's income is the second income in the home, but in others it is not, nor is it arguable when the woman
is single, widowed, or divorced. Some women find themselves bored and frustrated in the home and wish to pursue careers or follow other interests. Females have a dramatically longer life expectation than they did a century ago. And with birth control and smaller families, the span of years when a female has young children at home has been shortened dramatically. All of these are powerful social forces to make full-time homemaking less likely for females. The creative energies and constructive impulses of those in the mental health professions are needed in the effort to cope with these social changes in ways beneficial to children.
References BRODKIN, A. M., SHRIER, D., ANGEL, R., ALGER, E., LAYMAN, W. A. & BUXTON, M. (1984), Retrospective reports of mothers' work patterns and psychological distress in first-year medical students. This Journal, 23:479-485. HOFFMAN, L. W. (1979), Maternal employment. Amer. Psyclwl., 34:859-865. HUSTON, A. (1983), Sex-typing. In: Handbook of Child Psyclwlogy, Vol. 4, Chap. 5, ed. P. H. Mussen. New York: Wiley, pp. 387-468. KERMOIAN, R. (1982), Infant attachment to mother and child caretaker in an East African community. Unpublished Ph.D. dissertation, Stanford University, Stanford, California. SIEGEL, A. E. & HAAS, M. B. (1963), The working mother: a review of research. Child Develpm., 34:513-542. SIEGEL, A. E., STOLZ, L. M., HITCHCOCK, E. A. & ADAMSON, J. (1959), Dependence and independence in the children of working mothers. Child Develpm., 30:533-546. STOLZ, L. M. (1960), Effects of maternal employment on children: Evidence from research. Child Develpm., 31:749-782. WERNER, E. E. (1979), Cross-cultural Child Development. Monterey, Calif.: Brooks/Cole. YARROW, L. J. (1961), Maternal deprivation: toward an empirical and conceptual re-evaluation. Psyclwl. Buli., 58:459-490.