Anonymous wrote:

Anonymous wrote:

Anonymous wrote:For the top high schools—public magnets and prep schools—the colleges are familiar with how previous students from those schools have done, and that influences how they view new applicants.

Colleges absolutely know my high school and they have confidence that graduates from my school do well at their institutions. That’s one of the reasons why so many kids from my high school go to top colleges.

That’s nice for you.

I was just answering the question. Some high schools are absolutely known.

Well no sh*t. But there are tens of thousands of high schools in this country. If we rely only on “popular high schools” the top colleges will be 100% white and Asian. Which is the opposite effect of removing the sat!

Anonymous wrote:

Anonymous wrote:For the top high schools—public magnets and prep schools—the colleges are familiar with how previous students from those schools have done, and that influences how they view new applicants.

Colleges absolutely know my high school and they have confidence that graduates from my school do well at their institutions. That’s one of the reasons why so many kids from my high school go to top colleges.

That’s nice for you.

I was just answering the question. Some high schools are absolutely known.

Anonymous wrote:For the top high schools—public magnets and prep schools—the colleges are familiar with how previous students from those schools have done, and that influences how they view new applicants.

Colleges absolutely know my high school and they have confidence that graduates from my school do well at their institutions. That’s one of the reasons why so many kids from my high school go to top colleges.

That’s nice for you.

For the top high schools—public magnets and prep schools—the colleges are familiar with how previous students from those schools have done, and that influences how they view new applicants.

Colleges absolutely know my high school and they have confidence that graduates from my school do well at their institutions. That’s one of the reasons why so many kids from my high school go to top colleges.

Anonymous wrote:My high school had 5 kids attend Ivies, one attend Umich, One to Lehigh, and one to USC over an 11 year period. The rest were non selective schools. I doubt colleges were familiar with my high school.

Did you ever consider the opposite? That they are very familiar with it and that is what informed those decisions?

Anonymous wrote:Most selective private colleges have admissions counselors assigned to particular regions. Part of their job is learning what high schools in those areas typically do in terms of grading and class offerings, and also what their overall numbers look like. I worked in the admissions office at Georgetown as a student back in the 90s, and the admissions officer assigned to the NY / NJ area absolutely knew all about my high school, despite the fact that I was the only one admitted in a 3 year cycle, and only 2 out of 4 years for the whole school district.

Very good post -- admissions offices work very hard to understand all this context. Personal incredulity that they cannnot or will not does not change the fact that they do.

My high school had 5 kids attend Ivies, one attend Umich, One to Lehigh, and one to USC over an 11 year period. The rest were non selective schools. I doubt colleges were familiar with my high school.

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:With possible exceptions like the state flagship, which would have a lot of data, most selective colleges do not have sufficient sample sizes from a particular high school.

This is false.

Anonymous wrote:The performance of one or two or five students per year from a high school, in a variety of majors, should in no way reflect what a new high school senior's potential is.

And yet it is, because that's how data works.

Anonymous wrote:
I'm not saying colleges don't try to make such comparisons - I wouldn't know.

You have demonstrated that quite well.

What do you propose would be a statistically-significant sample size from a high school class of 400? How many students per year enrolled at the college x looking back how many yrs?

Over a number of years? It's easy. And remember this is not making the decision in a vacuum - it is a single data point assisting the decision process. Any amount of data is valuable and mine-able. Doesn't need to be only the state flagship. You don't think Yale does this? Or Williams? Or CMU? Or Colgate? This is how they assess college preparedness levels from different HS. It's a MUCH better predictor than a single test score.

I'll reverse the question on you: how little is NOT enough?

Most selective private colleges have admissions counselors assigned to particular regions. Part of their job is learning what high schools in those areas typically do in terms of grading and class offerings, and also what their overall numbers look like. I worked in the admissions office at Georgetown as a student back in the 90s, and the admissions officer assigned to the NY / NJ area absolutely knew all about my high school, despite the fact that I was the only one admitted in a 3 year cycle, and only 2 out of 4 years for the whole school district.

Anonymous wrote:

Anonymous wrote:With possible exceptions like the state flagship, which would have a lot of data, most selective colleges do not have sufficient sample sizes from a particular high school.

This is false.

Anonymous wrote:The performance of one or two or five students per year from a high school, in a variety of majors, should in no way reflect what a new high school senior's potential is.

And yet it is, because that's how data works.

Anonymous wrote:
I'm not saying colleges don't try to make such comparisons - I wouldn't know.

You have demonstrated that quite well.

What do you propose would be a statistically-significant sample size from a high school class of 400? How many students per year enrolled at the college x looking back how many yrs?

PP. I think the problems with testing were exacerbated by College Board's gradual move from what was more of an ability test in the 80s to the achievement test of academic skills that it is today (a purported measuring stick against Common Core), and all the murkiness in between. There seemed to be a large disconnect between its purpose (which was different things to different people with different agendas) and the reality.

Coleman explicitly made the current test more preppable via Khan, etc. but measuring high school academic skills achievement-style exacerbates the divide between students attending better and worse high schools. Makes it more difficult to find that diamond in the rough. The test is less a counterpoint to grades than a confirmation of them (or not, thanks to grade inflation).

And then there's ACT, the test of speed.

If I were College Board, I'd replace Coleman immediately. There have been more than enough disasters under his management. Hate to say it, but the way College Board could crawl back would be with another redesign, if it's clearly thought out, maybe a return to the early 1990s or something.

Anonymous wrote:With possible exceptions like the state flagship, which would have a lot of data, most selective colleges do not have sufficient sample sizes from a particular high school.

This is false.

Anonymous wrote:The performance of one or two or five students per year from a high school, in a variety of majors, should in no way reflect what a new high school senior's potential is.

And yet it is, because that's how data works.

Anonymous wrote:
I'm not saying colleges don't try to make such comparisons - I wouldn't know.

You have demonstrated that quite well.

With possible exceptions like the state flagship, which would have a lot of data, most selective colleges do not have sufficient sample sizes from a particular high school. The performance of one or two or five students per year from a high school, in a variety of majors, should in no way reflect what a new high school senior's potential is. There are far too many individual factors that would play a role, college classes/professors, who the student's high school teachers were and what their grading was like and the billion or so external reasons why some very able students might do poorly or somewhat less-able students might have top grades - this all varies widely.

I'm not saying colleges don't try to make such comparisons - I wouldn't know. I just think that it would be ridiculous and not a statistically sound approach.

To the extent that testing fades away, however, I could see this happening. Thus the attractiveness of a standardized test....

Anonymous wrote:My own college admissions counselor told me most colleges would adjust my GPA upward based on the prior strong performance in college by graduates of my high school. Do colleges really do this in a formulaic way?

I probably would be tempted to do this if I were a college admissions office but it would have a negative impact on high performers from weak high schools.

(Maybe someone should develop a standardized test to compare individual students who come from widely varying high schools.)

Colleges - in general - are like other business and know the value of data. They understand how students from certain HS perform better and use that in their decisions. Whether that manifests as a direct upward adjustment of a specific GPA would depend on the college, and I have not heard to that method specifically happening. But knowing that an A at Stuyvestant is not the same as an A at an ordinary (but fine) small rural HS is definitely a method.

My own college admissions counselor told me most colleges would adjust my GPA upward based on the prior strong performance in college by graduates of my high school. Do colleges really do this in a formulaic way?

I probably would be tempted to do this if I were a college admissions office but it would have a negative impact on high performers from weak high schools.

(Maybe someone should develop a standardized test to compare individual students who come from widely varying high schools.)