4 ms·

> with similar test scores > with similar test scores > WITH SIMILAR TEST SCORES

by Robotbeat 4y ago

throwaway81348 4y ago
Your point?
- UncleMeat 4y ago
  Presumably test scores are a not terrible proxy for qualification. So if black and hispanic students of similar qualification are not being admitted into these programs, then the programs are not actually equitably enrolling people.
fein 4y ago
I had to look around for the source of this, and I can't find the "with similar test scores" part in the Vanderbuilt site article on the Vanderbuilt study this is based on. It says 66% less likely, but the "similar test scores" part isn't in the same paragraph. I really dislike that it is so difficult to find the source of these statements in modern media, as the news articles I read didn't even cite the study, I had to go googling for it. What I found: https://news.vanderbilt.edu/2016/01/18/teachers-race-affects-gifted-program-selections/ https://news.vanderbilt.edu/2016/01/18/teachers-race-affects... and of course the link to the gov data source in that article is broken: http://https//nces.ed.gov/ecls/ http://https//nces.ed.gov/ecls/ Edit: I see where they pulled the line from. Misleading at best as the "tests" they are talking about are not gifted assessment tests, but rather standardized testing which I don't think is that great of an indicator. I was in the gifted program during middle school, and that was predicated by a few admittance tests, one of them being IQ.
- femiagbabiaka 4y ago
  If you read three paragraphs of the article you linked you can see the statement written in plain english: > However, controlling for math and reading scores did not have the same effect for black students. In fact, black students continued to be assigned to gifted programs half as often as white peers with identical math and reading achievement. If you read to the end, the paper is literally linked: https://cdn.vanderbilt.edu/vu-news/files/20190417222934/Grissom_AERAOpen_GiftedStudents1.pdf https://cdn.vanderbilt.edu/vu-news/files/20190417222934/Gris...
- UncleMeat 4y ago
  > Misleading at best as the "tests" they are talking about are not gifted assessment tests, but rather standardized testing which I don't think is that great of an indicator. I find this to be frustrating. How many times have we heard from people defending the existing systems that standardized testing is an appropriate method of stratifying people. When universities decide to minimize the relevance of the SATs we see howls of complaints. Now suddenly these tests are of minimal utility if they demonstrate that qualified black and hispanic students are being turned away?
  - femiagbabiaka 4y ago
    The goalposts will be moved until the desired outcome is achieved.
  - hackinthebochs 4y ago
    The test matters. You don't find the top 2% of ability by a test designed to measure competency, for example. This distinction may or may not be relevant in the context of gifted programs within a single school, but it is certainly relevant when comparing the utility of state competency tests and SAT tests.
- kube-system 4y ago
  > Misleading at best as the "tests" they are talking about are not gifted assessment tests, but rather standardized testing which I don't think is that great of an indicator. I was in the gifted program during middle school, and that was predicated by a few admittance tests, one of them being IQ. Yes, a student who studies hard may get good academic test scores, but may do poorly on gifted admittance tests. They're testing two different things. The former tests ability to understand class content, the other tests the ability to solve problems which are not part of school curriculum.
jaywalk 4y ago
I realize it's just a blog post and not a scientific paper, but the author didn't even cite a source for that statistic. My gut reaction is that it's probably incredibly misleading if not outright false.
xienze 4y ago
Yeah, it's very easy to draw conclusions that suit your agenda when you use slippery terms like "similar." A score of 90 and a score of 100 are "similar" in the sense that they're both an A, but one is a bit more impressive. And that's not even getting into the possibility that the author is using an even broader range of test scores, like "students who make all As and Bs on tests." For all you know, he could then be comparing white students making all As (and being accepted into a gifted program) with black students making all Bs (and not being accepted). Having the raw data matters.
- femiagbabiaka 4y ago
  The raw data is publicly available: https://nces.ed.gov/ecls/dataproducts.asp https://nces.ed.gov/ecls/dataproducts.asp
bgandrew 4y ago
- femiagbabiaka 4y ago
  Incredible. Not any statement, but left oriented, because of course lying and half-truths are limited to people with the political orientation you don't agree with...
- concordDance 4y ago
  EVERY uncited statement needs double checking. Even on non-contentious topics. Cited statements also need double checking. Frequently there's a game of telephone going on and the original research has been distorted. This is not just a left wing phenomena, it's a phenomena anywhere you get people who aren't completely dispassionate observers (this excludes all advertiser funded media as they are incentivized to make things interesting and engaging).

> with similar test scores > with similar test scores > WITH SIMILAR TEST SCORES

throwaway81348 4y ago

Your point?

UncleMeat 4y ago
Presumably test scores are a not terrible proxy for qualification. So if black and hispanic students of similar qualification are not being admitted into these programs, then the programs are not actually equitably enrolling people.

fein 4y ago

I had to look around for the source of this, and I can't find the "with similar test scores" part in the Vanderbuilt site article on the Vanderbuilt study this is based on. It says 66% less likely, but the "similar test scores" part isn't in the same paragraph. I really dislike that it is so difficult to find the source of these statements in modern media, as the news articles I read didn't even cite the study, I had to go googling for it. What I found: https://news.vanderbilt.edu/2016/01/18/teachers-race-affects-gifted-program-selections/ https://news.vanderbilt.edu/2016/01/18/teachers-race-affects... and of course the link to the gov data source in that article is broken: http://https//nces.ed.gov/ecls/ http://https//nces.ed.gov/ecls/ Edit: I see where they pulled the line from. Misleading at best as the "tests" they are talking about are not gifted assessment tests, but rather standardized testing which I don't think is that great of an indicator. I was in the gifted program during middle school, and that was predicated by a few admittance tests, one of them being IQ.

femiagbabiaka 4y ago
If you read three paragraphs of the article you linked you can see the statement written in plain english: > However, controlling for math and reading scores did not have the same effect for black students. In fact, black students continued to be assigned to gifted programs half as often as white peers with identical math and reading achievement. If you read to the end, the paper is literally linked: https://cdn.vanderbilt.edu/vu-news/files/20190417222934/Grissom_AERAOpen_GiftedStudents1.pdf https://cdn.vanderbilt.edu/vu-news/files/20190417222934/Gris...
UncleMeat 4y ago
> Misleading at best as the "tests" they are talking about are not gifted assessment tests, but rather standardized testing which I don't think is that great of an indicator. I find this to be frustrating. How many times have we heard from people defending the existing systems that standardized testing is an appropriate method of stratifying people. When universities decide to minimize the relevance of the SATs we see howls of complaints. Now suddenly these tests are of minimal utility if they demonstrate that qualified black and hispanic students are being turned away?
- femiagbabiaka 4y ago
  The goalposts will be moved until the desired outcome is achieved.
- hackinthebochs 4y ago
  The test matters. You don't find the top 2% of ability by a test designed to measure competency, for example. This distinction may or may not be relevant in the context of gifted programs within a single school, but it is certainly relevant when comparing the utility of state competency tests and SAT tests.
kube-system 4y ago
> Misleading at best as the "tests" they are talking about are not gifted assessment tests, but rather standardized testing which I don't think is that great of an indicator. I was in the gifted program during middle school, and that was predicated by a few admittance tests, one of them being IQ. Yes, a student who studies hard may get good academic test scores, but may do poorly on gifted admittance tests. They're testing two different things. The former tests ability to understand class content, the other tests the ability to solve problems which are not part of school curriculum.

jaywalk 4y ago

I realize it's just a blog post and not a scientific paper, but the author didn't even cite a source for that statistic. My gut reaction is that it's probably incredibly misleading if not outright false.

xienze 4y ago

Yeah, it's very easy to draw conclusions that suit your agenda when you use slippery terms like "similar." A score of 90 and a score of 100 are "similar" in the sense that they're both an A, but one is a bit more impressive. And that's not even getting into the possibility that the author is using an even broader range of test scores, like "students who make all As and Bs on tests." For all you know, he could then be comparing white students making all As (and being accepted into a gifted program) with black students making all Bs (and not being accepted). Having the raw data matters.

femiagbabiaka 4y ago
The raw data is publicly available: https://nces.ed.gov/ecls/dataproducts.asp https://nces.ed.gov/ecls/dataproducts.asp

bgandrew 4y ago

femiagbabiaka 4y ago
Incredible. Not any statement, but left oriented, because of course lying and half-truths are limited to people with the political orientation you don't agree with...
concordDance 4y ago
EVERY uncited statement needs double checking. Even on non-contentious topics. Cited statements also need double checking. Frequently there's a game of telephone going on and the original research has been distorted. This is not just a left wing phenomena, it's a phenomena anywhere you get people who aren't completely dispassionate observers (this excludes all advertiser funded media as they are incentivized to make things interesting and engaging).