I just love statistics like this... In order to have a high divorce rate, you have to have people who are married to begin with, as opposed to a higher percentage of residents who engage in long term co-habitation and having children outside of wedlock.
The last two items caused National Institutes of Health to revise their analysis a few years back to start to factor in co-habitation when looking at vital statistics such as birth and death rates compared to marriage/divorce rates.
Not surprisingly, "blue states" (typically with large urban concentrations) had higher occurances of co-habitation, and higher percentages of single parents who never married. Taken a step further, if you apply the frequency of divorce per 1000 married individuals to the co-habitation population, and add that rate into divorces per 1000 residents (married and unmarried), it flattens out the ratio enough to say there's practically no correlation between political leaning and divorce/breakup rates.