> “If fathers were really more likely to take off because they preferred sons, surely they wouldn’t wait 13 years to do so,” reasons Dr Kabatek.
It's pure speculation: no data, no model, nothing. But even if dr. Kabatek is right, it only takes a fraction of the fathers to obtain that 1.8% difference. You can't really explain such small differences with sweeping explanations (raising daughters is more stressful than raising sons, abortion of sons conceived into tenuous marriages, or your mother/daughter friction) without adding many, many more factors that mitigate them.
And let's not forget that the data is difficult to interpret. The (preprint of the) article's abstract even starts like this:
> Evidence from the U.S. that couples with daughters are more likely to divorce than couples with sons has not been found for other Western countries.