The average SAT score has remained in the middle 1,000’s for decades. Meanwhile, the number of kids scoring 1400+ has increased dramatically. Does that indicate the “dumbing” down of the test or point to the fact that many kids are now intensively prepping (pretty much unheard of in the 1980s when I took it) and taking the test multiple times (instead of once or twice)?
It maybe sensitive and it would be wrong in lot of cases. Pitt has 24k undergrads and its top quartile is 720/750/33. In other words, at least 6,000 students on campus are pretty darn smart.
It’s all about finding your type of people.
The strongest kids in the country gravitate towards the nationally highly ranked schools, if they can afford them. barring kids that want to stay in-state. Especially in some majors. That’s just a fact of life. “Sometimes” there isn’t even contention between something high ranked like CMU in CS vs someplace like Harvard for CS. I am sure CMU would lose quite a lot of those cross-admit kids. That’s just life.
1 Like
I wonder if there is a good source for actual cross-admit data. Parchment has a comparison tool here, but I am not really sure what their data is based on: Parchment comparison
My version of chatgpt built into my search engine (called Neeva) gives me this:
"The data for the cross-admit comes from the high schools that subscribe to Parchment, which offer their students the option to release that data to them. This is not based on the public student-submitted data.
"
Ah that’s interesting. So the tool is based on legitimate data that comes directly from high schools?
That’s the claim. Yes. This declaration is probably buried deep inside the website somewhere. chatgpt surfaced it. I haven’t dug through the parchment website myself.
That’s pretty cool then.
Edited: Based on posts below, and some more digging around I did on the Parchment site, it looks like this is just student reported data. Too bad! It would have been nice to find reliable cross-admit data.
Yes. I noticed that :-). I wish you and your son the best !!
Yield primarily relates to 3 factors – selectivity, use of early admissions restrictions (ED/REA/SCEA/…), and for a lack of better word uniqueness of the school. For example, kids who prefer UMass to Harvard generally don’t apply to Harvard as a backup in case less selective UMass rejects them. That wouldn’t make sense. Instead they only apply to UMass and not Harvard, so they don’t hurt Harvard’s yield. However, kids who prefer Harvard to UMass may apply to less selective UMass as a backup , so it hurts UMass’s yield. Yield is not a good measure of actual student preference. Parchment cross admits stats are even worse due to being self reported, outdated, and often having a small sample size.
2 Likes
Do you have information on exactly how Parchment gets its cross admit stats? Are there better cross admit stats available somewhere?
It’s mostly self reported. For example, if a kid attending Grand Canyon (for profit college in a lot of TV ads) says he also got in to Harvard, but chose not to attend, it counts as a cross admit choosing Grand Canyon over Harvard. The result is Parchment reports 58% of cross admits choose Grand Canyon over Harvard, as seen at Compare Colleges: Side-by-side college comparisons | Parchment - College admissions predictions.
I guess I still don’t understand how the data is reported to Parchment. Do the students themselves report it? How?
Since Parchment actually works with high schools and colleges, moving documents around when students apply and attend, it seemed like a potentially more promising source of data than (for example) something like Niche. But if it’s just random student reports, then it’s just like Niche or whatever…
With the Grand Canyon vs Harvard comparison that you linked, my understanding from the Parchment tool was that if the two percentages weren’t in color, they didn’t consider it statistically significant (presumably because there were too few reports).
I suspect the schools themselves know this information.
There was a document that was circulated around decades ago showing that Harvard won 60+% of cross-admits vs Stanford.
But nobody other than Stanford or Harvard themselves know cross-admit figures. But Larry Bacow said that no school wins in cross-admits against Harvard a few years ago in an interview.
In the past Parchment primarily seemed to be a site for students to support college applications. Students would enter where they planned to apply for college, and Parchment would list predicted chance of admission, sort of like Naviance. Students would later report their decisions – which colleges accepted them, which colleges rejected them, and where they are attending. These student reported decisions from previous years became the basis for the cross admit tool. Some HSs also used Parchment in place of Naviance, but my impression is the majority was self reported.
The schools will likely know. When you turn down Harvard (or any other college), they ask you where you are going. You can either choose to reply or not.
Of course the cross admits were fundamentally unsure about where they want to go to begin with. There is an earlier decision making that happens where kids ED or EA into schools that are not called Harvard.
1 Like
I don’t doubt that Harvard wins the majority of cross-admitted students. But if my kid had applied and gotten in to Harvard in addition to his other schools, he still would be going to a state school, because we can’t afford Harvard. (My guess is that he also wouldn’t prefer Harvard, but who knows. He has surprised me lots of times already.)
We know kids with similar admits whose parents hoped for financial aid that didn’t materialize, and the kids are going to state schools instead of the full price privates where they were also admitted…
So it’s plausible to me that some kids everywhere would choose a state school over Harvard or any other expensive private.
1 Like
Many kids would prefer an in-state public to Harvard… I’d expect the vast majority of kids in the US. However, those kids that prefer the in-state public to Harvard generally don’t apply to Harvard, so they don’t show up in yield or cross admit stats.
1 Like
How would they know this? They have no visibility to what other schools accepted their students. Perhaps they survey their matriculants or those admitted who don’t attend, but that would be student self reported data, if they even choose to answer the question.
Well, Stanford’s Faculty Senate circulated a document a few years ago showing cross-admits by university.
So I’m assuming they would know otherwise I wouldn’t know how it could have been on the agenda in 2010 or so.
But it’s certainly going to be the most accurate piece of data there is.