Today, , and I posted new work on the important and understudied phenomenon of retaking college entrance exams. "Take Two! SAT Retaking and College Enrollment Gaps" NBER version: Ungated version:
“Eliminating disparities in [SAT] retake rates could close up to 20 percent of the income gap and 10 percent of the racial gap in four-year college enrollment.”
has nice figure of our instrument for SAT retaking (left-digit bias, h/t ). Helping disadvantaged students retake can close college enrollment gaps, so says our paper.
Eliminating disparities in SAT-retake rates could close up to 20 percent of the income gap and 10 percent of the racial gap in four-year college enrollment, from , , and
New NBER working paper by , , & finds that low-income and minority students are far less likely to retake the SAT than other students.
This SAT test retaking paper is really cool - need to find a way to work into my classes!
Small changes in the college going process can make a big difference. New WP from & : retaking the SAT can substantially improve performance & help close income & racial gaps in 4-yr college enrollment
New WP by , , on retaking SAT and college enrollment, esp for low income kids. My mom got low-income fee waivers & made me retake—retakes should be encouraged and benefits (and waiver info) should be shared often
FINALLY (it was a short list of papers today)... Q: Why do people retake the SAT? Is retaking wise? A: People retake it because they have round-number-goals and get ticked off when they just-barely undershoot, and yes, it's good to retake it. #NBERday
This paper by et al is clever. I suspect the implications drawn are too clever by half, but I need to reread to think through this clearly (not cleverly).
Plus we now have clear evidence that retaking the SAT can lead to substantial score gains! Nearly one-fourth of those who retake see score gains of 150 points or more. So her gain is large but not suspicious.
Retaking the SAT improves scores, but underrepresented groups tend to retake it less
Presented this paper yesterday : Next day discovered a nearby restaurant named after the paper. Does this citation make it to Google Scholar?
For example, here's the one slide version our recent SAT retaking research (with and ).
Interesting: 10% of racial college enrollment gap may be attributable to gaps in *re*taking tests.
Or just retaking: "For students who initially score in the lower half of the SAT distribution, retaking once boosts superscores by nearly 0.4 standard deviations (120 points)" (on old 2400 point scale)