The author didn't considered some obvious correlations, maybe because it would be more difficult to get some data about it. It looks like that occupations that

The author didn't considered some obvious correlations, maybe because it would be more difficult to get some data about it. It looks like that occupations that promote more social interactions are more prone to divorce. I also spotted that occupations with weird or irregular work hours also have higher divorce rates. I didn't understand what the x-axis means.