How about broader based recruiting supplemented with test scores, that ensures the students recruited and selected, stay at the university longer. It's not an either or argument.
TU #195 US News 2024 rankings.
- Thread starter lawpoke87
- Start date
Well I have a giant hammer and post secondary education is my nail. I am digging in.
Dad brag time...my son was accepted into the University of Michigan College of Engineering on Early Action. I asked him if he needed to learn the words to "Hail to the Victors"
Adds: just found out he was also accepted to Illinois (early action) College of Engineering and NC State college of engineering (early action). I think he expected the NC State one today. He applied there because he was able to visit during a soccer trip and really like the campus, the engineering school/program, and the soccer facilities. Illinois has an excellent program for what he's trying to do. He now has options.
And someone I told about the Michigan thing asked if he would still go there or enter the portal now that Harbaugh was leaving. LMAO!!!🤣
And someone I told about the Michigan thing asked if he would still go there or enter the portal now that Harbaugh was leaving. LMAO!!!🤣
Dartmouth requiring SAT moving forward.
Cliff notes: “That is, contrary to what some have perceived, standardized testing allows us to admit a broader and more diverse range of students.”
admissions.dartmouth.edu
Cliff notes: “That is, contrary to what some have perceived, standardized testing allows us to admit a broader and more diverse range of students.”
Update To Testing Policy
Standardized testing like SAT is good. It tells you something about your prospective students using a common yardstick. I would love to see it accompanied with other types of standardized tests that get more at work habits and applied social skills. In work psychology, you get better prediction of success by looking at the whole person using standardized and reliable metrics. No reason to avoid the same approach with university admissions IMO.Dartmouth requiring SAT moving forward.
Cliff notes: “That is, contrary to what some have perceived, standardized testing allows us to admit a broader and more diverse range of students.”
Update To Testing Policy
drboobay, would be interesting to get your thoughts on the Yale and Dartmouth announcements.
My read is they are (a) renorming based on "local" scores rather than the national norm set, so each person's result is rescored to show their position relative to other students at their school rather than the national sample, and (b) banding, with everyone above it seems the 75th percentile at their school going into a "will succeed" band. They seem to mean that test score correlation with performance is logarithmic rather than linear, so a kid with a 34 isn't more likely to succeed than a kid with a 32 and so the 34 shouldn't get more consideration. I'm not sure what the re-normed score tells you but it's not a pure traditional intelligence or knowledge measure. Or maybe it's a more pure measure. IDK.
This is a really interesting use of the assessments that seems specifically intended to benefit kids at bad schools. It seems the re-normed standardized tests give schools a (constitutional?) and practical justification to say that a kid at Podunk HS or Inner City HS who took the only 2 AP classes his school offered is a better candidate than a kid at Holland Hall who took 15 AP classes. The Podunk High kid could get a 30 on the national norms but score higher on the re-normed local basis than a kid who scored 32 at Holland Hall since the HH kid will be competing with "local" norms that are way above the national norms. It's sneaky smart if so.
Last edited:
Aha! Now you have me interested. I will look carefully and reply back when not using mobile.
For a lot of schools, test optional is as much a cultural/political issue as an admissions issue. It's a red flag for the anti DEI crowd that some schools don't need.
I incorrectly assumed that "do you require tests?" and "what do you do with the scores?" were the same question, but they are not. Schools are finding innovative ways to use scores to support non-traditional candidates. Gone are the days where "higher score means higher chance of admission". Scores seem to work as a flexible floor to ensure that people who are admitted will succeed - everyone agrees you don't want to admit people who will fail out, but that doesn't mean that higher is better anymore.
Yale btw also accepts AP/IB test scores instead of SAT/ACT. Again, they are looking for indicators that students have a minimum skill set to succeed.
Test scores are unlikely to have an impact on the Holland Hall type kids - schools have enough data on those schools, they could probably predict the scores anyway. Test scores are information poor for most kids from established schools. They just help push kids from other schools over the top where the college might otherwise not have a basis to accept them based on a fairly limited background.
I incorrectly assumed that "do you require tests?" and "what do you do with the scores?" were the same question, but they are not. Schools are finding innovative ways to use scores to support non-traditional candidates. Gone are the days where "higher score means higher chance of admission". Scores seem to work as a flexible floor to ensure that people who are admitted will succeed - everyone agrees you don't want to admit people who will fail out, but that doesn't mean that higher is better anymore.
Yale btw also accepts AP/IB test scores instead of SAT/ACT. Again, they are looking for indicators that students have a minimum skill set to succeed.
Test scores are unlikely to have an impact on the Holland Hall type kids - schools have enough data on those schools, they could probably predict the scores anyway. Test scores are information poor for most kids from established schools. They just help push kids from other schools over the top where the college might otherwise not have a basis to accept them based on a fairly limited background.
This is interesting, and harkens back to how pre-employment assessments were scored prior to the 1991 Civil Rights act - in other words, the first few years I was in graduate school. Back then, within group norming was common - black candidates compared to other black candidates, white candidates to other white candidates, woman candidates to other woman candidates, etc. It was an easy way to avoid disparate impact. The results certainly provided utility in predicting job performance - just not as much utility as they would have if scored using a common norm - because there is so little research suggesting meaningful differential validity for most tests in employment contexts. Of course, the 1991 Civil Rights Act made this practice illegal - it is explicitly taking race into account when making employment decisions.
Of course, this practice does not EXPLICITLY take race into account when making college selection decisions. However, I will suspect that is occuring de facto in many settings based on the predominance of a particular racial group in some schools. The interesting thing for me would be to know more about the research Dartmouth is citing. I don't accept it at face value without seeing exactly what they did and the soundness of their conclusions. Did they find that within group norming by school improved validity in comparison to national norms, with respect to some important outcomes such as college grades, retention, graduation? If so, this would imply some type of differential validity based on socioeconomic factors, and their approach could be very sensible. Personally, I have conducted research where within-facility standardization across business locations has improved validity, so it's not completely far fetched to me. Did they publish their research in a peer reviewed journal? If so, and you can point me to it, I would be happy to review it and give an objective take on what I see.
Last edited:
Super interesting. If they are re-norming within high school, it is not explicitly race based and should be legal under recent SCt precedent that focuses on facial equity and not outcome equity. But I haven't read those cases, just reports on them. There is a case in VA where a high school has taken an equal number of students from each of the middle schools in the district. Families at the "better" middle schools have objected bc they would get more kids in if everyone was put into one pool. It hasn't gotten very far so far. I wouldn't be surprised if the courts ruled that you look only at facial equity if outcome equity help minorities but look at outcome equity if that help the majority.
On this - "Did they find that within group norming by school improved validity in comparison to national norms, with respect to some important outcomes such as college grades, retention, graduation?" My sense based on just reading the reports is they basically defined the DV as being dichotomous rather than continuous. E.g., you graduate with a 3.0 GPA or above, or you don't. That makes the validity question a lot easier.
Either way, what it means to be "test mandatory" means a lot different outcome than what I've always thought it meant.
If they found that renormed scores within HS predicted ultimate college GPA better than the national norms, then what you're testing is quite different than intelligence or education. E.g., you scored a 30 on the national norms but 95th percentile in your high school and had a higher GPA than me at Dartmouth even though I scored a 32 on the national norms but 80th percentile in my high school. That would be crazy, I really doubt that was what they found but maybe.
I don't know if their research has been published or not. I just read the releases on the Dartmouth and Yale pages.
Last edited:
Is is super hard to understand exactly what Dartmouth found. Yale says that test scores in HS are the best predictors of GPA but not much else about what they found.
I also agree that much depends upon what you are shooting for - GPA, retention, etc. Also, if diversity is an end in itself, I have to say it is super weird we have to use proxies because it's something you can select for directly. If diversity in SES is what you are shooting for, then use it directly and select those of highest ability within SES bands.
Of course, the piece that's missing here are qualities like conscientiousness and social skills. Universities should use situational judgment tests alongside cognitive tests IMO to get at "knowing the right thing to do and how to treat others." I know that about a decade ago, there was an effort to add an SJT component to the LSAT and it appeared to be predictive of important outcomes as lawyers, and not just in law school. Shelly Zedeck was involved in the effort and has talked about it with me. But it never was implemented for reasons that are unclear to me.
I have never once seen a study showing that clinical judgment trumps an algorithm though. So I'm doubtful this "holistic" talk track really amounts to anything but making parents feel better, and introducing bias and error.
I also agree that much depends upon what you are shooting for - GPA, retention, etc. Also, if diversity is an end in itself, I have to say it is super weird we have to use proxies because it's something you can select for directly. If diversity in SES is what you are shooting for, then use it directly and select those of highest ability within SES bands.
Of course, the piece that's missing here are qualities like conscientiousness and social skills. Universities should use situational judgment tests alongside cognitive tests IMO to get at "knowing the right thing to do and how to treat others." I know that about a decade ago, there was an effort to add an SJT component to the LSAT and it appeared to be predictive of important outcomes as lawyers, and not just in law school. Shelly Zedeck was involved in the effort and has talked about it with me. But it never was implemented for reasons that are unclear to me.
I have never once seen a study showing that clinical judgment trumps an algorithm though. So I'm doubtful this "holistic" talk track really amounts to anything but making parents feel better, and introducing bias and error.
Why elite colleges are bringing the SAT back
Yale and Dartmouth are bringing testing back — but thousands of other schools aren’t.
www.vox.com
The key takeaway seems to be that underrepresented students don't realize they are being assessed on a localized scale and thus are less likely to submit scores because all they see is their nationally normed score. Other research shows that tests mandatory means underrepresented people are less likely to apply at all - now they apply but hurt themselves by not submitting scores that they don't realize are competitive. Using test mandatory in this context is a lot different than based solely on national norms and makes a lot of sense if they can get kids to apply in the first place. I don't know if it's legal though, which might be why most schools still don't do this.Why elite colleges are bringing the SAT back
Yale and Dartmouth are bringing testing back — but thousands of other schools aren’t.
Based on Figure 2, there seems to be a lot more error in the correlation bn score and gpa for disadvantaged students than advantaged students. I wonder why that is?
"Moreover, Dartmouth Admissions reads SAT/ACT scores “in context,” or in relation to the local educational
environment. For example, an applicant with an SAT score of 1400 has a higher probability of admission if from a high school where average SAT scores are relatively low. Under a test-optional policy, these students are less likely to be identified and admitted."
"For example, at an SAT score of 1400, less-advantaged students have twice the probability of admission of more-advantaged students. These relatively high-achieving less-advantaged students likely should submit their scores, as their score would benefit their application."
@chito_and_leon What I see here is the common pattern with cognitive tests in employment. There is no slope bias - the plotted lines are roughly parallel. The regression line for more advantaged students is above the one for disadvantaged students. This pattern would suggest that use of a common regression line, mixing the groups, will result in OVER predicting less advantaged students' performance on the criterion (first year GPA at Dartmouth, it seems). So there is no clear evidence of statistical bias disfavoring the disadvantaged group. it's pretty much a textbook example of nearly any differential prediction analysis I've ever seen comparing White versus Black job applicants and how cognitive scores predict job performance.
You raise another issue, which is possible differences in prediction error for the two regression lines. It's not a topic I have reviewed in detail before, but I would expect a steeper line if prediction was more accurate for the advantaged group. Visually, I don't see that.
Of course the right technique to analyze this would be moderated multiple regression. I am sure they would find significant intercept differences, reflecting the overprediction for less advantaged students by using a common regression line. But the illustration would suggest there is not a statistically significant interaction term (i.e., they have similar slopes).
English please? 😂 😂 😂
Statistics wasn't my jam. Calculus fine, but statistics sucked!
In other news, my son was just named one of the valedictorians at Union for this year. And to be open about how they do it, any student with a weighted GPA above 5.0 (on a 5.0 scale) is name a valedictorian. It encompasses about the top 1% of students in this class. Basically they wanted it to be less cutthroat and to reward the kids who excelled in the highest level classes offered. AP classes and what they termed "college level" classes were awarded 6 GPA pts for an A. It allows kids who do more than just academic classes as part of school to be recognized. Extra curriculars are no longer "extra" but are regarded as classes and only awarded 4 GPA pts for an A so kids in band or athletics really are hampered by being involved in those activities if they are also an academic kid (like my son). I told the story about how the valedictorian and saludatorian in my HS class was decided by .0002.
And we also discussed the topic above about how my son's ACT score in Oklahoma (and at Union) is likely to be viewed very favorably by some of these high academic schools because of the comparison.
This is how they did it at Jenks when I graduated, but that was bc we didn't have AP or IB classes so a 4.0 was the highest GPA you could get. As I vaguely recall, there were 8 or 9 of us but the class back then was only maybe 350 kids.
Yeah, my kid has a class of almost 1100 at Union and he will graduate roughly 10th or 11th in his class. He's taken 7 AP classes plus 3 add'l classes they treat as college level classes (Calc 1, Organic Chem, and Differential Equations). I'm pretty sure the chemistry teacher wishes he were teaching and doing research in college. And he teaches his HS classes like they were college classes, which I appreciate because it has forced my son to adapt to that style and actually think when doing homework or taking an exam instead of just regurgitating facts. The same with the Calc teacher (who teaches Calc 1, AP Calc BC, and Differential equations...and having 1 teacher who teaches the advanced math like that has been beneficial for the students. This is why I get defensive when someone says Union should split its HS into 2 and make them smaller. Why? My kid has every advantage and all the support he would ever need just as much as any other student at Union. Those resources would be split and watered down if you split the school up. If you can do what you do well with what you have, what would be the reason to split the school?
My kids' HS is a little smaller but still big. The thing I like isn't that they throw tons of resources at the "smart" kids - every wealthy suburban school district takes care of those kids - it's the great programs for kids that don't want to go to a 4 year college. They provide work experience in the trades and kids graduate with advanced skills certifications or slot easily into a community college program to get certifications quickly. There are a lot of great jobs as electricians, welders, machine workers, etc. and preparing kids for those jobs is fantastic. They recently added aviation so kids can accelerate into aircraft mechanic and pilot programs. If the school broke in half, the kids applying to Northwestern or MIT wouldn't be affected, but the kids who want to be a machinist would be.
The top kids are going to be fine no matter where they are at. It is the middle and lower parts of high schools that get lost in these monster schools. Sorry dude. Your bias towards the monsters is wild.
And Union supports all of these kids as well. My son's class size has changed very little from when he was a freshman. When I was in HS, I started with a class of 1200 and ended up about 1000. In MA, once a kid hits 16, they can drop out of school without needing a parent's permission. I don't know what the law is in OK, but at least at Union you're not seeing a lot of kids dropping out and the HS has a stated goal of 100% in college or work ready when they leave. They have partnerships with TCC and Tulsa Tech for this part and I'm here for it. I think last year, out of the 17 seniors on the soccer team, all were either accepted to college or had a trade job lined up and probably 3-4 of them had started their own businesses.
You're clown side is coming out. Unless you have a kid in the school you have no idea what services and support they offer and what's available. I don't comment on Jenks, Bixby, Owasso or BA because I don't know for certain what they offer their students. I talk about Union because I know for certain what they offer. I get the daily emails, I get the CCC emails and notifications daily, I've met the counselors, I've met the teachers.
At the end of the day, I don't care about athletics or football and those things shouldn't drive decisions like this. I know Union's support of all of its students and available programs would suffer if it were to split. Stop being the jealous west side dad whose schools struggle to compete and want to blame the split. If one of the football coaches really wanted to, they could go and grab all of the best players in Edmond and consolidate them into one school because OK has an open transfer policy.
And yes, I know, my kids will always do well no matter what school they are in. I do believe public ed is a partnership between the school and the parents and we've always been involved. The kids that don't thrive in public ed are usually the ones whose parents/guardians aren't invested or don't have the time to be invested. There's not an answer to that issue. You can split the school up 4 times and you still won't solve that part of the equation.
Sorry I get very excited when I see something I actually know! I really know little about football or basketball beyond being a fan and watching.
It's hard to know too much by just looking at the graph in the Dartmouth report. They leave out the statistics thout would be definitive, so I am inferring a lot.
But what I see is the standardized test predicting first year Dartmouth GPA really well. The two parallel lines on the graph would suggest overprediction of GPA for the low SES group if you mixed the two groups together (the way that standardized tests are usually used - without separating out SES by group).
In other words, if your goal is to predict first year GPA, the standardized test is NOT BIASED against the lower SES group, If anything, it is biased against the higher SES group.
I happen to be somebody who has looked again and again and again at standardized cognitive tests in employment. And it's the same story. I've never once seen standardized test scores show bias that disfavors minorities when predicting job performance. Commonly, the tests will overpredict performance.
Bottom line - the graph and paper suggest no bias in predicting GPA. The bigger question, as @chito_and_leon has pointed out, is what your "criterion" or goal is as a university. Is it to get a student population that gets the best grades possible? If so, nothing is better than the SAT or ACT. It works much better than high school GPA, and it's not biased for that purpose. However, if you want other things, like well adjusted, hard working students, it doesn't tell you much. If you value diversity as an end in itself, obviously SAT and ACT work against that objective - unless maybe you mess around with how it's being used (which maybe Dartmouth is doing) and not just say "higher is better."
Yes. There is considerable investment in “how it is used” to increase, not decrease diversity and target specific under represented demographics amongst POC.
They say this without saying it ..... Their challenge is communicating this to the pool that they want to overselect without getting sued. It's interesting that Yale and Dartmouth have gone test mandatory (though not ACT or SAT mandatory at Yale) but Harvard, Stanford and Michigan among others have not.
"For example, at an SAT score of 1400, less-advantaged students have twice the probability of admission of more-advantaged students. These relatively high-achieving less-advantaged students likely should submit their scores, as their score would benefit their application."
Again I am not a lawyer here. But if you speak plainly that you are looking to address socioeconomic disparities - that they are recognizing some of us are born into this world with more resources and opportunity than others - and stay away from talking about (and using) race directly, I would think this would work. In other words, really do this and don't use it as a pretext.
That’s exactly the law. Race can no longer be a pre-text or the sole deciding factor in admissions decisions. Something that’s been the law in California for decades btw and the Supremes just got around to forcing the Ivys to do.
You can and should continue to ask questions of students about life experience and world view in admissions decisions with the aim of a diversified student body. As part of that inquiry, race can be considered and discussed. To the extent those discussions are used to balance racial goals at the expense of some “more qualified” candidates remains to be seen. Schools may continue to decide solely on race but with life experience as a pre-text. We just don’t know. It wouldn’t be the first court decision ignored by ideologues, the states or the marketplace.
I think it's too early to say. I suspect we'll end up with some prohibition on pretextual decisions that will almost exclusively be used to strike down programs that promote diversity. We have a "you can have your cake and eat it too" Supreme Court. Those challenges are coming already. They probably won't focus on outcome/effect as that would open the door for challenges to many facially neutral policies that have a strong discriminatory effect, but there probably will be a review of whether the policy makers intended the policy to have a race effect. I suspect the Harvards and Stanfords are trying to get these outcomes in a less explicitly stated way.
Appeals Court Upholds Admissions Policy at Elite Virginia High School
BTW let me put my cards on the table. I am an old fashioned believer that a color blind society is an end goal. Just like there is little if any pejorative to being of Italian or Irish descent now ideally we get to a place where being of African descent yields the same kind of reaction. Not there yet but a lot closer now than when I was born. Hoping these kinds of pretexts become irrelevant.
As you know with your psychology background, we're evolutionarily designed to create in groups and out groups and to think our group is better than the out groups. Who occupies the out group may change but discriminating based on some out group feature won't go away. You might as well try to convince people not to procreate or eat. The best we can do is protect the out groups but the law is clearly moving away from that (though it has accidentally made it easier for now, which will change). We had a nanny from Bulgaria and she absolutely hated gypsies. I didn't even know gypsies were a thing anymore. It's always going to be someone.
I don't like cold water splashed in my face at bedtime!As you know with your psychology background, we're evolutionarily designed to create in groups and out groups and to think our group is better than the out groups. Who occupies the out group may change but discriminating based on some out group feature won't go away. You might as well try to convince people not to procreate or eat. The best we can do is protect the out groups but the law is clearly moving away from that (though it has accidentally made it easier for now, which will change). We had a nanny from Bulgaria and she absolutely hated gypsies. I didn't even know gypsies were a thing anymore. It's always going to be someone.
Antiziganism, or prejudice against the ethic Romani peoples is a significant problem in both Eastern and Western Europe. Indeed, polite society, even on the left, in large numbers, in Germany, Italy, and France will casually mention their desire to see Romani people removed or marginalized from their society, oblivious that their words align with Nazi state policy, who sent hundreds of thousands to be deported and then later on to concentration camps when other nations would not accept them. Some estimate the number killed at over 1.5 million.As you know with your psychology background, we're evolutionarily designed to create in groups and out groups and to think our group is better than the out groups. Who occupies the out group may change but discriminating based on some out group feature won't go away. You might as well try to convince people not to procreate or eat. The best we can do is protect the out groups but the law is clearly moving away from that (though it has accidentally made it easier for now, which will change). We had a nanny from Bulgaria and she absolutely hated gypsies. I didn't even know gypsies were a thing anymore. It's always going to be someone.
This isn’t a small problem. There’s about 10 million “Gypsies” in Europe and a majority of Europeans view them unfavorably across wide number of polls going back decades across demographic and political party lines. Your average Italian or Czech citizen is likely to say some pretty vile things about them and would do nothing if they witnessed a beating or a theft if the victim is Romani. They are prohibited in state schools in some instances. Lots of other petit bigotry.
Europe isn’t the social paradise a lot of liberals in the United States want to think it is.
Last edited:
I knew gypsies were targeted by the nazis but didn't realize that was still a thing. "Social paradise" is relative, I think... That said, I think most people realize that Europe is not a monolith socially, culturally or politically. The 3 examples you give in particular have large segments of non-progressive people (Marine Le Pen got 42% of the vote in France and the Italian PM is head of a party that's basically a more racist version of MAGA). Of course the gypsies have some decidedly unliberal cultural traditions like child marriage and subjugation of women that also probably don't help them in the eyes of progressives.
Similar threads
