might want to read the next paragraph
" Democrats have had less time to divorce because they marry later. And because the GSS data are not longitudinal, we do not know if the Republican rate of divorce for any given marriage is indeed higher. Additional research will have to determine, conditional on length of marriage, if Republicans or Democrats have greater risks of divorce."
basically Rs have been married longer, thus the possibility of divorce is greater. also that quote you use, and the article itself use several terms I couldn't find definitions for that seem like they need to throw out some qualifiers for.
"ever-married" vs "shared married". they say Rs are more likely to be "ever-married", so beyond the Rs being married longer, and more being married, it seems like there is some qualifier they are throwing in with the part you quoted.
here are some other qualifiers they throw out. it sounds like they aren't tracking which way someone leaned when they divorced, they are just looking at the current status. the implications that divorced men are joining Rs would throw the metrics.
"his may be partly because
divorced men have moved in large numbers into the Republican camp in recent years. It may also be due to the fact that Republicans marry at an earlier age, which may increase their lifetime risk of divorce. "