I can't see the article without buying it (nope) but my first "hm, I wonder..." would be to see if they adjusted for education, income level, and parents' incom

I can't see the article without buying it (nope) but my first "hm, I wonder..." would be to see if they adjusted for education, income level, and parents' income level. This might just be saying something like "bullies don't excel in school", given what we already know about socioeconomic demographic trends WRT fertility. With the wrong study design they could even get this result if bullies have lower fertility than some SES sub-group they also tend to be a part of. One would suppose they adjusted for that, but sometimes papers make it out with some really weird oversights.