,
18 tweets,
5 min read
Read on Twitter
Lots of questions today about my work on the ECD after that WSJ article. I'm really proud of the work and hope I can clarify any misunderstandings. I've been working for a long time on admissions & contexts for low-income students. (thread & AMA below) #emchat #admission
This work was done in collaboration with Nick Bowman (Iowa), @kkglasener @KC_Deane & Jandi Kelly at Michigan. The published papers available here: www-personal.umich.edu/~bastedo/topic…
Holistic admissions has been done for a long time, which means (ideally) reading applications in the context of the opportunities available in high schools and neighborhoods. But admissions officers had much more refined data on applicants than they had on their contexts.
A lot of offices relied on their own impressions during college visits or school profile sheets provided by the high schools themselves. These were often unavailable, incomplete, and had no standard format.
I gave a presentation to CB researchers in 2015 where I laid out this research and said they could be the natl org with the kind of resources and relationships to be able to fix this.
Our first paper (a lab experiment) showed that just improving context data increased propensity to admit low SES applicants by 25%. www-personal.umich.edu/~bastedo/paper…
Admissions officers who did contextualized holistic review were even more likely to admit the low-SES applicant: www-personal.umich.edu/~bastedo/paper…
CB researchers ran with this and developed the ECD as a *dashboard* of information that admissions officers could use to better understand a student's contexts using reliable, national data on their high school & neighborhood. It is not just one adversity score, as reported.
Our first paper on the 8 pilot institutions shows that ECD adopters are significantly more likely to admit low-income applicants, particularly from non-feeder high schools. It is under review now but you can read here: www-personal.umich.edu/~bastedo/paper…
You can see a version of the ECD here: collegeboard.org/membership/all…
SAT is just one part of the ECD, to contextualize the score. The ECD is being used in test optional schools. The SAT could end tomorrow and the ECD would still be useful, reliable, robust info. You can hate standardized tests and love the ECD (and vice versa, I suppose)
The ECD is just one part of the application. So race isn't on the ECD, for example, because it's already on the application in the applicant's info. It can be used at schools with race-conscious admissions and schools under AA bans.
ECD is not "part of" the SAT. It is ported to admissions offices and it doesn't matter if they use SAT, ACT, or no tests at all. The ECD also does not have any student-level information, just aggregate data from the neighborhood and high school.
Not everything in ECD is what I would choose to do. It is a CB product, and they have a million stakeholders. But it is a *huge* improvement over the data AOs had before. I think if we care about using evidence to improve equity, we should support the ECD.
The ECD is not a panacea. Applicants will have other stressors and adversities in their lives (disabilities, death of parent, many things) that have to come through the rest of the application. That's holistic review. This is only trying to improve one part.
I also get there are a lot of people suspicious of CB for lots of reasons. I would say orgs are complex, and you can support some things they do without supporting everything they do. CB *must be* transparent about the formula being used to calculate the adversity scores.
If you read this far, thank you! I'll answer any questions I can about the research below. I've never posted a thread before and generally nervous about Twitter, so please be kind.
More detailed description of the ECD data just posted: professionals.collegeboard.org/environmental-…
