Would you contribute to an admissions database?

[FranklinR]

The APA requires that schools publish some data profiling who got in to their programs. This data is quite limited, frequently out of date, and only tells part of the story.

In particular, while it provides (sometimes) median values, it doesn't provide ranges. It also doesn't speak to letters of recommendation, POI contact, whether or not you already have an MA, publications, etc.

And, significantly, it doesn't tell us who applied but didn't get an interview, and who applied, got an interview, but didn't get an offer, and who got an offer but walked away.

What this would give us is a more complete profile for a school - does the school have a GRE/GPA floor? Do GREs really matter to this particular school? What percent of those who got interviews had prior contract with a professor?

This would require (anonymous) contribution of a lot of data from many, many applicants, preferably over time, since it would be easiest to collect data at several points in the application cycle. Which means an actual database, with user ids and passwords, not just a survey. Which we almost have here in this forum. (not that this is necessarily the best place for something like this, although it does have a certain appeal.)

I don't have a proposal for making this work. I'm just wondering how many of you would a. be willing to contribute (Grades, GPA, MA status, Publications, POI contact, titles of letter of rec senders, GRE, schools applied to, possibly more) and b. be interested in the results. and c. if someone knows how to put it off, well then, all the better!

---

Members don't see this ad.

 

[lbug312]

I think it sounds like a decent idea. I know a data base like this exists for Dental School. Here is the website, you can check it out and other people can see how it would work.

http://predents.com/

 

[cara susanna]

That's a great idea! If I get in anywhere, I will certainly contribute. Though I can contribute the place where I got an interview, though it remains to be said if I'll get an offer.

I know some HTML, but probably not as complicated as you would need for this. I wonder if there's a website that would have a service like this.

 

[FranklinR]

This would suggest to me that if I did this, I should require contribution of data to see the results. The predent database is pretty thorough - perhaps it could be modified to include psychology results...

 

[irish80122]

I pitched this idea when I applied two years ago and few were interested. Now that I am going through my second interview cycle on "the other side" I understand why. If you make the cutoffs for the school/prof (some profs like their applicants to be higher) then it really is all about match with the program and what the program needs. If one applicant is a great match with a 3.6 GPA and a 1150 GRE and a second is an iffy match with a 4.0 and 1450, from what I have observed, my program would take the first candidate.

It is a great idea, don't get me wrong, but the numbers aren't what drive the process, it is the match. The main thing is that your numbers are in-line, and the programs' websites and disclosure data can tell you that.

 

[cara susanna]

I agree that match is the most important, but it would be nice to know GRE range etc, and how many people who were invited for interview ended up getting in. Stuff like that.

 

[Thrak]

Honestly, I'm not sure it would help anything. The thing about the selection process for Psych PhDs is it seems to be qualitative, rather than quantitative. If it were law school or business school (or in a much better example, undergraduate colleges), it would make a lot of sense. The quantitative data is probably very predictive, even through the higher end of the spectrum (a generic student with 170 LSAT and 3.75 GPA will more likely than not get him/her into law schools ranked X through Y on any given year).

For Psych PhDs, you're not only looking at universities, you're looking at professors, and you're looking at given years. A 1350 GRE/700 Psy GRE/3.75 GPA will get you to the interview stage at Generic U if you're looking to work with Professor X, but only if you're interested in working in the Smoking Cessation via Alchemy lab, and only if the 8th-year grad student finishes his dissertation by the Fall. Otherwise, you're out of luck. If the 8th year does finish, you might lose out to someone with a 1200/650/3.6, if that person is interested in that lab and you're not.

If the 8th year doesn't finish, nobody gets accepted to work with Professor X, regardless of scores.

I don't think this kind of data can be ballparked properly. The range is probably too great, and based on too many qualitative factors, to be predictive.

Just my opinion, though.

 

[FranklinR]

Honestly, I'm not sure it would help anything. The thing about the selection process for Psych PhDs is it seems to be qualitative, rather than quantitative. If it were law school or business school (or in a much better example, undergraduate colleges), it would make a lot of sense. The quantitative data is probably very predictive, even through the higher end of the spectrum (a generic student with 170 LSAT and 3.75 GPA will more likely than not get him/her into law schools ranked X through Y on any given year).

For Psych PhDs, you're not only looking at universities, you're looking at professors, and you're looking at given years. A 1350 GRE/700 Psy GRE/3.75 GPA will get you to the interview stage at Generic U if you're looking to work with Professor X, but only if you're interested in working in the Smoking Cessation via Alchemy lab, and only if the 8th-year grad student finishes his dissertation by the Fall. Otherwise, you're out of luck. If the 8th year does finish, you might lose out to someone with a 1200/650/3.6, if that person is interested in that lab and you're not.

If the 8th year doesn't finish, nobody gets accepted to work with Professor X, regardless of scores.

I don't think this kind of data can be ballparked properly. The range is probably too great, and based on too many qualitative factors, to be predictive.

Just my opinion, though.

Sometimes it seems like it's all about the match - and the match may be why a specific person gets in. Other times, it seems like it's all about the numbers. My feeling is that numbers will keep you out, where a match will get you in. And in order to actually attend, you have to both get in and not get eliminated.

What you don't get from the schools published data is the range of scores accepted. Is it broad? For a program like Penn, I kind of doubt it, but I don't know. We could know, if we had the data, but right now it's all speculation.

For that matter, it can't be all match - I can write a proposal that matches a professor's interests perfectly, pitch my prior research as if it was a perfect blend - and not get in. Match is (in theory) about interests, and it's hard to believe that we're not all pretty good at pitching our interests to match the professor.

I'm floating this as a trial balloon, since for it to work it would require that you submit your data to the system and then specify how the school responded. If you're not sure the data would be useful to you, would you still submit yours?

 

[psychanxiety]

I might participate, but the thing that concerns me is getting too specific. I don't want to be identifiable, so if I post exactly which schools I applied to, where I get offers, where I accepted, my interests, publications, scores, etc. it adds up to identifiable information. Part of the value of these online forums is the freedom to express critical opinions, which is only possible because of anonymity.

Edit: If this was conducted by a trusted institution like Psi Chi, APA, or a respected university then I would be completely on board. But not just SDN, sorry.

 

[FranklinR]

I might participate, but the thing that concerns me is getting too specific. I don't want to be identifiable, so if I post exactly which schools I applied to, where I get offers, where I accepted, my interests, publications, scores, etc. it adds up to identifiable information. Part of the value of these online forums is the freedom to express critical opinions, which is only possible because of anonymity.

Edit: If this was conducted by a trusted institution like Psi Chi, APA, or a respected university then I would be completely on board. But not just SDN, sorry.

My presumption would be that there would be no connection between your identity here and the data you submit. So while it might be conceivable to identify you based on the combination of your test scores, schools, etc., there wouldn't be any connection to anything you post here.

I don't really know who would conduct this. I am associated with a university, but I don't know if they or I would want for them to be associated with a data collection project like this.

I was thinking more like the predents.com model, except not using SDN user names. What would you think?

 

[irrealised]

If psychanxiety's concerns are the same as mine, I think the issue IS that there will still be one name linking us to all of the schools that we applied to (not to mention the rest of our stats). I wouldn't do it.

Now, if there was a way to enter information specifically for each school with no other information (i.e. no user name), then I would definitely add mine. I would even be willing to have an "invisible" user name email that was kept by the database owners (presuming this all seemed legit) as long as none of that info was publicly available. I'm assuming anything less would just lead to a useless database full of spam and completely unreliable info, but that's as much as I would be comfortable with.

 

[FranklinR]

In my head, it wouldn't be possible to find out all the schools that a particular person applied to, or even to get a profile of a particular person. It would be school based: a profile school by school, of the scores & data of who got in, who got an interview, who got rejected. Really, organizing the information by person isn't particularly useful.

It would indeed be best if the APA did this, but clearly the schools don't want to publish this data, or they would. I'm not entirely sure why they don't, other than perhaps it would highlight the fairly random nature of admissions. But schools don't really deny that, so maybe it's only a logistical issue. Except a lot of schools use the same application backend, so it would be fairly easy...

Interesting, the dentists don't seem to have a problem with personally identifiable information on predents. Interesting, indeed.

 

[doktor yoyo]

Deleted By User