So the standard practice on this research question has become to always control for household instability and thus make the unacceptable finding of worse outcomes go away. But that is like showing (as a hypothetical example) that attending public schools in richer neighborhoods does not really cause more access to admissions to “better” colleges because if one controls for student SAT scores and impressiveness of résumés and application essays, then neighborhood wealth and school resources become statistically insignificant—which, any sociologist would know, is ridiculous. But apparently
...more
