>I'd say it almost certainly does significantly change their chance of divorce

Where's your data to break the Null Hypothesis?

Though, if the author of the article wanted to back up their claim, they would attempt to show to what degree drastic changes in career/income had ok divorce rate.