3 ms·

"Their high school grades had turned out to be much better predictors of their academic ability and college potential than their SAT scores." I have to admit I

by MarkMc 7y ago

"Their high school grades had turned out to be much better predictors of their academic ability and college potential than their SAT scores." I have to admit I'm surprised by this, but it seems to be backed up by research: "As SAT scores are a more consistent indicator of aptitude, one might expect them to better predict a student’s chances of graduating college than high school GPA. But Chingos’ research shows exactly the opposite." [0] By the way, I enjoyed reading Paul Tough's previous book "How Children Succeed" - if you are curious, here is a great excerpt from that book: https://news.ycombinator.com/item?id=5721868 https://news.ycombinator.com/item?id=5721868 [0] https://www.forbes.com/sites/prestoncooper2/2018/06/11/what-predicts-college-completion-high-school-gpa-beats-sat-score/#cff67ba4b09f https://www.forbes.com/sites/prestoncooper2/2018/06/11/what-...

travisoneill1 7y ago
The SAT score buckets in the analysis max out at >1100 so not very useful. The grade buckets are also not very useful as the top range 3.67 - 4.00 probably includes > 50% of students at most schools (thanks, grade inflation). It also fails to adjust for college difficulty. Students in the >1100 bucket are obviously going to more difficult schools than the <800 bucket.
AcerbicZero 7y ago
I had a similar surprise when I first saw that data, but it does look to be solid. The conclusions drawn from it, leave a little more to be desired, as I think these newer studies seem to forget that there have been numerous studies regarding SAT/ACT's and GPA as predictors for college success. I'm not an expert on this topic, so I googled a few older studies, (and skimmed them). The trend of SAT/ACTs being relatively good predictors for college GPA's seems to continue up until around 2008, at which point it starts to diverge more appreciably. (2001) https://pdfs.semanticscholar.org/e2a8/25cfc687ce1be8be48a693f08c8caf5da4f6.pdf https://pdfs.semanticscholar.org/e2a8/25cfc687ce1be8be48a693... (2003) https://eml.berkeley.edu/~jrothst/publications/rothstein_sat_jmetrics2004.pdf https://eml.berkeley.edu/~jrothst/publications/rothstein_sat... (2006) https://files.eric.ed.gov/fulltext/ED563073.pdf https://files.eric.ed.gov/fulltext/ED563073.pdf (2008) https://www.sciencedirect.com/science/article/abs/pii/S0160289608000603 https://www.sciencedirect.com/science/article/abs/pii/S01602... There are tons of variables to all of this, but its a hard sell to suggest that SAT's have always been poor predictors of college GPA. To me, it seems more likely that changes to both the testing and to the way colleges are run over the past ~30+ years is likely the root cause. College was already like glorified HS when I was there, I can only imagine how bad its gotten over the past few years. And if college is copying high school, is it really surprising that HS GPA is aligning with college GPA's?
swasheck 7y ago
This isn't really all that surprising to me given the goal of the study is to define who are the candidates most likely to succeed at continuing to practice the same set of characteristics that got them to college in the first place (e.g. work habits and endurance).

by MarkMc 7y ago

travisoneill1 7y ago

The SAT score buckets in the analysis max out at >1100 so not very useful. The grade buckets are also not very useful as the top range 3.67 - 4.00 probably includes > 50% of students at most schools (thanks, grade inflation). It also fails to adjust for college difficulty. Students in the >1100 bucket are obviously going to more difficult schools than the <800 bucket.

AcerbicZero 7y ago

I had a similar surprise when I first saw that data, but it does look to be solid. The conclusions drawn from it, leave a little more to be desired, as I think these newer studies seem to forget that there have been numerous studies regarding SAT/ACT's and GPA as predictors for college success. I'm not an expert on this topic, so I googled a few older studies, (and skimmed them). The trend of SAT/ACTs being relatively good predictors for college GPA's seems to continue up until around 2008, at which point it starts to diverge more appreciably. (2001) https://pdfs.semanticscholar.org/e2a8/25cfc687ce1be8be48a693f08c8caf5da4f6.pdf https://pdfs.semanticscholar.org/e2a8/25cfc687ce1be8be48a693... (2003) https://eml.berkeley.edu/~jrothst/publications/rothstein_sat_jmetrics2004.pdf https://eml.berkeley.edu/~jrothst/publications/rothstein_sat... (2006) https://files.eric.ed.gov/fulltext/ED563073.pdf https://files.eric.ed.gov/fulltext/ED563073.pdf (2008) https://www.sciencedirect.com/science/article/abs/pii/S0160289608000603 https://www.sciencedirect.com/science/article/abs/pii/S01602... There are tons of variables to all of this, but its a hard sell to suggest that SAT's have always been poor predictors of college GPA. To me, it seems more likely that changes to both the testing and to the way colleges are run over the past ~30+ years is likely the root cause. College was already like glorified HS when I was there, I can only imagine how bad its gotten over the past few years. And if college is copying high school, is it really surprising that HS GPA is aligning with college GPA's?

swasheck 7y ago

This isn't really all that surprising to me given the goal of the study is to define who are the candidates most likely to succeed at continuing to practice the same set of characteristics that got them to college in the first place (e.g. work habits and endurance).