The most incisive guide to issues facing the American family today . . . An invaluable resource for anyone wishing to stay on the cutting edge of research on family trends.

-W. Bradford Wilcox
Associate Professor of Sociology, University of Virginia 

A “Good” Time to Split?

For parents thinking about a divorce, one consideration is often their children’s ages.  Some research has indicated that children who experience a parental divorce later in life seem to fare better than children who experience such a trauma in the early years.  In a new study examining the results of relationship dissolution on Norwegian children’s academic performance, however, researchers discover that the relationship between children’s ages and parental relationship dissolution may not be quite so simple.

In their study, Wendy Sigle-Rushton et al. speculate that the best method to determine the effects of parental union dissolution on children’s academic performance is to compare the experiences of siblings.  Examining children who experience the same home environment, with the distinguishing factor of age at parental dissolution, would seem to be the best method to determine if the older siblings had an easier time of it than the younger ones did.  The researchers’ goal is to discover if parents would be wiser to stay in an unhappy relationship until their children are older.

Study data was taken from the Norwegian register data files and included both married and cohabiting heterosexual couples who had a child born in the period 1986-1990, another born in 1987-1992, and who had dissolved their relationship in the period 2002-2008.  The researchers explain that although researchers focused on data in the U.S. would exclude cohabiting couples in such a study, “those who study northern European countries have tended to combine, whenever possible, married and cohabiting couples,” and that hence it “made sense to follow this convention.”  The researchers are quick to point out, however, that most of the couples who experienced a first birth while still cohabiting married shortly thereafter, and that the “vast majority of the dissolutions” (over 80%) were to married, not cohabiting, couples.  Using this data, the researchers estimated linear regression models, with age at parental dissolution as predictor variable and final grades the outcome variable.

The researchers used three separate models to make their estimates.  In the first, “controls for birth order and year of birth were not included and represent the naïve assumption that all sources of bias are captured by family fixed effect.”  In the second model, the researchers controlled for birth order, as first-born children tend to fare better in school generally than higher-order children, and such results could skew data.  And in the third model, “additional controls for year of birth were introduced to control for a source of bias that should overstate the extent to which young children are able to adjust to the effects of the disruption.”

The researchers received very mixed results, depending on the model examined.  In the first model, the earlier the parental dissolution took place in children’s lives, “the lower their average grades at the end of compulsory education were.”  This model’s results, in other words, suggest that delaying a dissolution may in fact be good for children.  The second model, the one taking birth order into account, had dramatically different results.  In this model, “the age gradient reversed, and children who experienced their parents’ dissolution early in life did better at the end of compulsory schooling than children who experienced a parental union dissolution at an older age.”  The researchers speculate a “crisis” model, in which children who experience a dissolution earlier in life have more time to “recover” before the outcome measure (end of compulsory schooling).  The third model, which assumed some grade inflation over time, found “very little evidence . . . that age at dissolution matters, except perhaps when it took place in the last year of compulsory schooling.”  The researchers called these results further evidence for a “crisis” model.

The researchers close by cautioning that, although their results seemed to show evidence for a “crisis” model of dissolution, “[g]iven the powerful influence of each of a small set of control variables, however, this substantive finding should be interpreted . . . with some caution.  There may be additional confounding factors we have not included in our final model.”  In other words, divorce is messy, harmful, and its full effects on children’s academic achievement are not yet known.

(Wendy Sigle-Rushton et al., “Proceed With Caution?  Parents’ Union Dissolution and Children’s Educational Achievement,” Journal of Marriage and Family 76 [February 2014]: 161-74.)