Our job as social scientists is to help disentangle causation from correlation. This can be a very difficult, and sometimes impossible, job. Sometimes it is a matter of the chicken and the egg. For instance, we know that divorced people are more likely to drink, use drugs, have lower income, and be less happy. But did divorce cause these bad things or did these bad things in fact cause the divorce? The latter phenomena — what social scientists call reverse causation — is easy to understand. For example, most people would understand if someone dealing with a spouse with a chronic drug or alcohol problem was more likely to seek a divorce or that depression could spill over into marital difficulties. In fact, research has implicated some of the negative outcomes observed among the divorced in the divorce itself. For example, when a man is laid-off the couple is more likely to divorce in subsequent years. Other research has shown that the happiest people are the most likely to get married, while those divorcing are less happy than the average married person. Moreover, people who divorce are actually happier a few years after the divorce compared with their happiness in the years prior to the divorce.
Disentangling causation from correlation is not only about accounting for reverse causation. The second difficulty is the possibility of a third factor that is influencing both the likelihood of divorce and other outcomes about which society cares. The most obvious example here is to talk about children and divorce. It is clear that children from divorced households fare worse along a range of outcomes than those from intact households. However, this does not imply that these children would be better off if the divorce had never occurred. Children of divorced families may have worse outcomes because the type of parents and households that end up divorced are different from those that do not. As just discussed, drug or alcohol abuse, depression, and financial difficulties might both lead to worse outcomes for children and increase the probability of divorce. Even if the divorce is simply the result of the selfish, narcissistic preferences of the parents, it is likely that kids being raised by such adults will have worse outcomes compared with kids raised by a devoted altruist. Accordingly, even if policy forced the parents to remain married, we might not see better outcomes for these children.
The reason it can be so difficult to disentangle causation from correlation is because divorce does not happen randomly in the population. Rather divorce occurs in couples who are typically worse off than the average married couple before the divorce occurs. We would all like to believe that we can be safeguarded against disease, death, poverty, unhappiness, and insanity by finding someone to love and holding on tight. But the data suggests that while love may conquer some adversities, it flourishes without them.
This brings us to an even harder question: What are the differences between married and separated life for those whom policy can affect? It is clear that children do better in a loving marriage than a messy divorce. But few families face this choice. When asking whether the children of divorced parents would have been better off if their parents had not divorced, we are usually asking whether conflict or separation is in the kids’ best interests. Thus far, the empirical research provides little evidence that stopping these divorces would benefit the children. Research by Mark Hoekstra at the University of Pittsburgh followed families on the brink of divorce comparing those who filed and then reconciled with those who filed and then divorced. He found little difference in the academic and behavioral outcomes of the kids in each of these types of families. While this may not be the perfect test — after all these marriages were pretty far along the process before they reconciled — it does indicate that at least for some couples reconciling is not very likely to help the kids.
While advocates hide behind correlations, the evidence that preventing divorce would benefit children is weak at best. Given that there is little evidence that the government can do a better job determining which marriages will be best for the children, a reasonable default rule might be to leave Mom and Dad to make the best decisions for their kids. As others in this forum have noted, altruism is an important part of families and parents who love their children just might make the right decision.