ess #1: factors affecting educational achievement
this is my first of several posts about sessions i saw at ess.
this session only had 2 papers, both quantitative. the first was by ann owens at harvard on neighborhoods as contexts for adolescents’ school success, using the add health data. she constructed an index of “disadvantage” in the census tract – % single mothers, % in poverty, etc.; and an index of “neighborhood status” – % with a BA, etc. she found no significant effect of relative neighborhood position within the school – that is, there’s no additional disadvantage to students from poor neighborhoods who attend mixed-SES schools. this is good news for proponents of (class) desegregation because it means students aren’t hurt by the relative deprivation or what have you.
unsurprisingly, having ambitious peers and having a high status neighborhood intensifies college attendance.
actually, the most interesting thing about this session was the interaction between the audience and the speakers (both of whom were grad students). during this 1st talk, 2 people* were kind of muttering back and forth, and ended up leveling some pretty sophisticated critique of the statistics.
the critique was basically that neighborhood effects might be artifacts of individual level measurement error (eg, kids’ reports of their parents’ educations, which are probably systematically biased). this seems to be a really important issue to me. i’m not sure what else to say about it, but it will be on the back burner.
the second critique was that school effects might be artifacts of selection effects. i think, especially if we take a liberal definition of selection, this is probably true. that is, most people don’t live in particular neighborhoods by accident. there’s a lot of evidence that affluent folks choose where to buy a house based on the schools, and there’s also a lot of evidence that poor folks don’t often have a lot of choice where to live (which under my broad definition is selection too). but on the other hand, these non-random outcomes may still turn out to have compounding effects – the whole larger than the sum of its parts, i mean, leaving some real neighborhood or school effects.
anyway, owens gave a really remarkably poised response. the guy sitting behind me whispered, “i’d be terrified to go next!” i agreed – and so did the next presenter, a first year student who urged us to “be nice to the young-un”.
the amazing part is – everyone was! her project was very preliminary, but she got feedback and advice that was perfectly suited to her level. i was shocked, not only in contrast to the (good) sophistication of the comments to the 1st presenter, but also that 2 presenters got quite good feedback at the appropriate level. it made me sort of optimistic about the field in a way i didn’t expect, and set a nice tone for the rest of the conference.
*one of whom turned out to be jaap dronkers, who did a similar thing in a later session. i don’t know who the other chap was.