Volume 1995 0717
Date: 17 July 1995
M.D.R. Evans and Jonathan Kelley
Institute of Advanced Studies, Australian National University
M. Borgers, J. Dronkers, and L. Rollenberg
University of Amsterdam
© Copyright 1995 by M.D.R. Evans, Jonathan Kelley, M.
Borgers, J. Dronkers, and L. Rollenberg.
All rights reserved.
Divorce rose dramatically throughout the Western world in the 1960s and 1970s. Some observers welcomed this change arguing that it would free people from unhappy marriages, thereby improving the sum of human happiness. In this optimistic view, divorcing parents and children would all benefit from the end of parental strife and dissension, and the mother would remarry a more compatible husband and together they would make a warm, secure family with firm guidance and control over the children. Others argued that easy divorce would eventually lead to the collapse of family life, to the detriment of all, especially the children. In this pessimistic view, raising children is a challenging enough task for two parents, but much harder for one parent, especially in the difficult teenage years.
The impact of divorce and remarriage on children's progress in school has, in particular, become a major policy issue, and a major focus of research, because success or failure in school has enduring consequences for the quality of life people get, the kind of jobs they work at, who they marry, and how much they earn. Optimists argue that parents getting a divorce would not harm their children's education, in part because the key impact of the family arguably lies in early socialization and in exposure to high culture, neither of which are much affected by divorce. But pessimists argue that the disorganization, loss of parential control, and financial restrictions associated with divorce will reduce the effectiveness of parents' socialization, so that children from divorced or separated families will not get the full benefit of their parents' resources and hence will have lower educational attainment than children from intact two-parent families. Moreover, the difficulties that a sole parent -- or a step-parent -- faces in raising children increase the risk that a child will do poorly in school. Evidence from the USA and several other countries is not unequivocal, nor unproblematic, but generally tends to support the pessimistic view, suggesting that divorce has a small but statistically significant cost to the child's education and job prospects (Biblarz and Raftery, 1993; Cherlin 1992: 75-90; Dronkers, 1994). Folkwisdom, too, emphasizes the up-hill battle of raising children in broken families: only minorities "agree" or "strongly agree" that "A single mother can bring up her children as well as a married couple" in Australia (27%), Britain (32%), and America (38%) in the International Social Survey Programme's "Family Values-Round 1" conducted in the late 1980s (Evans 1992).
In the past, divorce was rare and difficult in Austalia, as it was in most Western nations. But no fault divorce was established by the Family Law Act taking effect in 1976, which made divorce far easier to obtain. The result was a sudden but brief upsurge in divorce, presumably reflecting decades of pent- up demand. Divorce then settled down again but at a level roughly twice what it had been before the divorce reform. The lifetime risk of divorce was about 10% of marriages up to the 1960s, rising dramatically to about 40% for the marriages of the late 1970s (Carmichael and McDonald 1986). It then stabilized with perhaps a gradual increasing, reaching 43% in 1993 (Webster 1995). Current divorce levels are well within the usual range for Western nations, but still substantially below American rates.
Enough time has passed since the Australian divorce reforms that we can now begin to assess the impact of divorce on children's success in school using survey data. This paper does so.
We analyze the effect of divorce using data from the International Social Science Surveys of 1994 and 1989. These are both representative, nationwide samples of Australians in all states and territories, with 4,513 respondents from 1989 and 1,378 from 1994. The descriptive results of table 1 and the analysis of social differences between couples who stay together and those who divorce are pased on a pooled sample from these two surveys.
To evaluate the educational consequences of divorce we analyze the educational experiences of our 1989 respondents and their brothers and sisters. This gives information on fully 12,451 children, 834 of whom had their childhoods disrupted by divorce and 891 disrupted by the death of a parent.
In both these surveys, we asked people an extensive set of questions about their family situation when the were growing up (around age 14): about their mother's education, their father's education and occupation, the number of books in the home, and whether their parents had divorced or separated. We also asked about another kind of disruption more common in past generations: whether one of their parents had died. Measurement details are in an appendix.
Our data show that few children born early in the century had to cope with divorce: only 6 to 7 percent grew up in homes where there had been a divorce (table 1). This remained true until the 1960s. But then for children born in the late 1960s -- who were teenagers in the 1970s around the time of the divorce reform -- there was a dramatic change: 10 to 12 percent grew up with divorced parents. This pattern continues to the present, so that now about twice as many children must cope with divorce as in earlier generations. By contrast, in the United States about 20% of children had to cope with divorce, rising to a projected 40% for more recent cohorts (Bumpass, 1984).
Although children now are much more likely to face the difficulties of divorce, children in earlier generations faced another, in some ways similar risk: many parents died early from disease, accidents or war, thereby depriving children of a parent. Early in the century 10 to 12% of children suffered the death of a parent. This risk has declined steadily to 3% in modern times (except during the world wars).
These two trends pretty much off-set each other, with the risk of losing a parent through death declining about as fast as the risk of losing one through divorce rises. As a consequence, throughout this century around 15% of children have grown up in families that lost a parent either through divorce or through death. There was a similar pattern in the USA from about 1860 to the 1970s (Cherlin 1992: 25-26).
Our data suggest that divorce is as common among well-educated families as among poorly educated families, equally common among professional and managerial fathers at the top of the occupational hierarchy as among unskilled workers at the bottom, and equally common among intellectually oriented families living in homes full of books as it for families without a single book in the house (see appendix). The only difference is history: divorce was, as we have seen, rarer in the past than it is now.
Losing a parent through death is equally undiscriminating (see appendix). Death was more or less indifferent to the family's education, occupational status, or intellectual orientations. Again the only difference is history, with families in the past far more vulnerable than modern ones.
Background and Controls. To clearly establish the effect that a parent's death or divorce has on the child's education, we first take into account what the family was like to begin with: whether the parents were well or poorly educated, whether the father has a good job or a bad job, whether the family fostered an intellectual atmosphere by having lots of books around the house, how many other brothers and sisters were in the family, and whether they grew up long ago when few people got much schooling, or more recently when many people stay on in school or even go to university.
All of these are important:
Divorce. Our results suggest that divorce has an educational cost, but it is not an educational disaster: children whose parents divorce get on average, half a year less education than children from intact families. This loss is for children who are otherwise comparable -- children whose parents have the same education, whose fathers have the same occupation, who come from the same size families, and who grew up in the same historical period.
For comparison, this half-year educational loss from divorce is about the same size as:
Our estimate of a half-year disadvantage stemming from divorce in Australia is at the low end of the range found in the USA (Duncan, Featherman and Duncan, 1972; Featherman and Hauser, 1978). The figures for both countries may be somewhat too large because they do not take into account that some children's educational difficulties preceed their parents' divorce, rather than being caused by it. This may account for half (for boys) or less (for girls) of the apparent effect of divorce (Cherlin et al. 1991).
Losing a parent through death has the same effect as losing a parent through divorce: the child gets about about half a year less education. This loss is for children whose families are otherwise similar in education, father's occupation, family size, and living in the same historical period. Similar results have been found in the USA (Crain and Weisman 1972).
That the loss from a parent dying is the same as the loss occasioned by a parent divorcing suggests that the two may have a common cause: the loss of paternal encouragement, emotional support, and guidance (Hogan and Kitagawa 1985). Research on the USA tends to find that, net of economic factors, parents' divorce has a detrimental impact on their children's behavior (e.g. Furstenberg et al. 1987), and that these behavioral differences probably entirely account for the worse school marks of children of divorce (Mulkey, Crain, and Harrington 1992).
The loss occasioned by divorce may also reflect the loss of financial support. This
possibility has been repeatedly suggested for the USA but other analyses of the Australian
experience suggest that finances have only a little to do with a child's chance of
completing year 12 in school, and nothing to do with university attendance (Evans and
Kelley, forthcoming). Moreover, Australia's relatively strict enforcement of child support
payments and relatively generous welfare system shelter children financially.
A sibling-model approach to these issues is in:
N. Borgers, J. Dronkers, L. Rollenberg, M.D.R. Evans and Jonathan Kelley "Educational Resenblance Between Australian Siblings: Gender, Generations, Migration, Family Forms and Mothers' Work". Paper to be presented to the 1995 Annual Meetings of the American Sociological Association, Washington, D.C.
Copies from: Prof Jaap Dronkers/ COBO/ University of Amsterdam
Education. Father's education, mother's education, and sibling's education was asked in categories:
None (0 years)
Only a few years of primary (3.9 years)
Finished primary (6 years)
Left school about age 14 or 15 (8.9 years)
A little more than than (10.1 years)
Finished secondary (12 years)
Some further study beyond that (13.2 years; asked of parents only)
University or CAE graduate (15.7 years)
We recoded these into mean years of schooling using data on the actual mean years of schooling for respondents in the corresponding categories. This gives scores ranging from 0 (no education) to 15.7 (university completion), as shown above. Respondent's educaiton was ascertained by a series of questions on years of primary and secondary schooling and highest educational qualification (coded into the Australian Bureau of Statistics' 3 digit edcuational code), and recoded into usual years of schooling.
Fathers Occupational Status. Father's occupation was ascertained by a detailed series of questions on occupation, duties, and industry; coded into the 4- digit Australian Standard Classification of Occupations (ABS XXX); and thence recoded into Kelley's worldwide status scores which provide a reliable measure similar to Duncan's status scores for the USA (Kelley 1990). They range from 0 to 100:
Farm worker (status= 00)
Farm owner (status= 10)
Unskilled worker (status= 14)
Routine service (status= 18)
Semi-skilled worker (status= 24)
Routine sales (status= 32)
Higher service (status= 33)
Skilled workers (status= 37)
Routine clerical (status= 38)
Higher sales (status= 51)
Higher clerical (status= 60)
Technical (status= 70)
Admin, managerial (status= 75)
Higher professional (status= 100)
Father's occupationl status was not reported for only 7% of our respondents overall. But for families broken by divorce the figure was 20% and fully 46% for families broken by death. When it was not reported, we estimated it by OLS from father's education if that was available (father's education was missing for only 15% of families broken by divorce and 17% broken by death).
Books in the home. We measured the intellectual orientation of the family by the number of books in the home whern respondent was 14. For the analysis we use the natural log of the number of books.
Number of siblings was measured by direct questions on number of brothers and number of sisters. Preliminary analysis showed that the effects of havng brothers were the same as the effects of sisters, so we combined them for simplicity.
Year of birth was measued by a direct question.
Parents' divorce or death. Information on family type came from two questions:
" Were you living with both your own mother and father around
the time you were 14?"
Both own mother and father
Father and stepmother
Mother and stepfather
" (If you were not living with both parents) What happened?"
One or both parents died
Parents divorced or separated
Astone, Nan and Sara McLanahan. 1991. "Family Structure, Parental Practices, and High School Completion." American Sociological Review 56:309-320.
Carmichael, Gordon and Peter McDonald. 1986. "The Rise and Fall of Divorce in Australia". Presention, Annual Meetings of the Population Association of America, San Francisco.
Cherlin, Andrew J. 1992. Marriage, Divorce, Remarriage. (Revised Edition) Harvard University Press.
Cherlin, Andrew J. et al. 1991. "Longitudinal Studies of Effects of Divorce on Children in Great Britain and the United States" Science 252 (June): 1386- 1389.
Crain, R. and C. Weisman. 1972. Discrimination, Personaltiy, and Achievement. New York: Academic Press.
Biblarz, Timothy J. and Adrian E. Raftery. 1993. "The Effects of Family Disruption on Social Mobility". American Sociological Review 58: 97-109.
Bumpass, Larry L. 1984. "Children and Marital Disruption: A Replication and Update" Demography 21: 71-82.
Duncan, Otis Dudley, David Featherman and Beverly Duncan. 1972. Socioeconomic Background and Achievement. New York: Seminar Press.
Dronkers, Jaap. 1994. "The Changing Effects of Lone Families on the Educational Attainment of their Children in a European Welfare State". Sociology 28: 171-192.
Evans, M.D.R. 1992. Australian Family Values in International Perspective. Melbourne: Australian Institute of Family Studies. (300 Queen St., Melbourne, Victoria).
Evans, M.D.R. and Jonathan Kelley. (forthcoming) "Cultural resources and academic success: the beaux arts vs human capital" Revised version of a paper presented at the Annual Convention of the Hungarian Sociological Association (June, 1991). Forthcoming in Lifestyles and Social Stratification in Comparative Perspective edited by Harry B.G. Ganzeboom, Nan Dirk de Graaf, and M.D.R. Evans.
Featherman, David and Robert M. Hauser. 1978. Opportunity and Change. New York: Academic Press.
Furstenberg, Frank, S. Philip Morgan, and Paul Allison. 1987. "Paternal Participation and Children's Well-being after Marital Dissolution" American Sociological Review 52: 695-701.
Hogan, Dennis P. and Evelyn M. Kitagawa. 1985. "The Impact of Social Status, Family Structure, and Neighborhood on the Fertility of Black Adolescents." American Journal of Sociology 90:825-56.
Kelley, Jonathan. 1990. "The Failure of a Paradigm: Log-Linear Models of Social Mobility" Pp. 319-346 and "Kelley Replies to Muller". Pp. 349-57 in John Goldthorpe: Consensus and Controversy, edited by J. Clarke, S. Modgil and C. Modgil. London, England: Falmer Press.
McLanahan, Sara. 1985. "Family Structure and the Reproduction of Poverty." American Journal of Sociology 90: 873-901.
Mulkey, Lynn M., Robert L. Crain, and Alexander J.C. Harrington. 1992. "One- Parent Households and Achievement." Sociology of Education 65:48-65.
Webster, Andrew. 1995. "Recent Trends in Divorce: Duration-specific Divorce Rates, 1977-1993, Australia." Australian Bureau of Statistics, Demography Working Paper 95/1.
Effects of parents' divorce or death on their children's educational attainments. Australia, 1989. N= 12,271. -------------------------------------------------------- Variable B SE B Beta T -------------------------------------------------------- Parent divorced -.50 .09 -.04 -5.34 Parent died -.46 .09 -.04 -5.06 Mother's education .11 .01 .12 11.46 Father's education .11 .01 .12 10.84 Fathers occup. status 1.39 .10 .13 13.70 Books in home (ln) .33 .02 .19 20.32 Number of siblings -.10 .01 -.08 -9.28 Age -.04 .00 -.21 -24.32 (Constant) 8.88 .13 70.37 (R Square= .29) --------------------------------------------------------
The International Social Science Survey / Australia is a nation-wide
survey conducted by researchers at the Australian National University and the University
of Melbourne. Begun in 1984, it is Australia's leading academic survey and is devoted
entirely to academic research in the social sciences, is non-profit, and is not connected
with any business or political party. The survey's core sponsor is the Research School of
Social Sciences at ANU. The ISSS is based on large, representative national samples of all
states and territories, drawn from the electoral roll. The detailed and comprehensive
survey takes about two hours to complete. It is conducted by mail. The first mailing
includes a cover letter from the Australian National University and a postage-paid reply
envelope, followed by a further letter about two weeks later. Anyone who did not respond
within a month or so is then pursued by up to three more mailings over a six month period.
Comparison with the census shows the samples collected in this way to be representative of
the Australian population in age, sex, education, occupation, and other characteristics.
Dr. Jonathan Kelley is Director and principal investigator of the ISSS; Dr. Clive Bean
(Associate Director), Dr. M.D.R. Evans and Dr. Krzysztof Zagorski are co-principal
Visitors since 14 September 1998:
|WwA: Worldwide Attitudes||[ WwA Home | International Survey Center Home]|