Jersey Boys (and Girls) Have Lowest Divorce Rate in US

Yesterday's New York Times was all over a recent data from the recently released American Community Survey 2011 showing that the "divorce rate" is lower in New Jersey than in any other state. Who would have guessed? Well, actually, anyone who has been observing the data over time would have seen that divorces per person have been low in New Jersey for some time. In general, they are lowest in the northeast and highest in the south. An important reason for that is that you have to get married in the first place in order to get a divorce and marriage rates are lower in the northeastern states than most elsewhere in the country. I put "divorce rate" in quotations to remind you that when the Census Bureau calculates its divorce rate, the denominator is simply the number of people aged 15 and older. However, a true divorce requires that the denominator be the number of married people--the people actually at risk of getting a divorce. The fewer people married (as in New Jersey), the lower the divorce rate will automatically be. Those marriage data are harder to come by than they used to be, however. The Census Bureau notes that:

Historically, data on marriages and divorces in the United States were collected from marriage and divorce certificates filed and collected at the state-level through the National Center for Health Statistics (NCHS) vital statistics system. In 1996, the Centers for Disease Control and Prevention (CDC) and the NCHS discontinued the collection of detailed state-level vital records data from marriage and divorce certificates. Beginning in 2008, questions about marital events were added to the ACS to collect national and state-level marriage and divorce data. These new marital events items fill a void in the marriage and divorce data collected in the United States.

I will say that the New York Times did a good job of rounding up a star-studded cast of American sociologists to comment on these results and the statistical issues were at least alluded to, if not exactly made specific in the article.