Post-II Acceptance stats are skewed

[RogueBanana]

A lot of members have been sharing statistics about specific school "Post-Interview Acceptance Percentages"

As a data-analyst by trade I was wondering, how accurate could these numbers possibly be?

For example, its pretty much accepted that ADCOMS will invite the top applicants for interviews first, early on in the cycle. They give interviews to the people they want to attend their school.

If we say a school as a 40% post-II acceptance rate, does that mean they are going to reject 60% of the applicants they want the most? I doubt it.

Wouldn't it be fair to say that the earlier on your II is, the greater your post II acceptance rate (assuming you don't say anything stupid during the interview, or just aren't a good fit for the school)

This hypothetical 60% rejection percentage would be skewed by outliers in both directions (extremely low acceptance late in cycle, extremely high early in cycle)

It'd be great to be able to crunch the numbers on this and see how much the spectrum varies over the course of the cycle.

---

Members don't see this ad.

 

[ready2go2]

They give interviews to the people they want to attend their school.

Hmmmm I'm not sure how true that statement is. I think it might be better to say they give people they want to get to know more in person an interview invite.

Also applying early may be overhyped when you think about it. It could be that quality applicants know to apply early, apply early, and therefore receive interviews earlier. But it might be true that if they applied "late" or "on-time" they might also be shown the same amount of love.

 

[RogueBanana]

Hmmmm I'm not sure how true that statement is. I think it might be better to say they give people they want to get to know more in person an interview invite.

Also applying early may be overhyped when you think about it. It could be that quality applicants know to apply early, apply early, and therefore receive interviews earlier. But it might be true that if they applied "late" or "on-time" they might also be shown the same amount of love.

Yet another possible confounding variable!

 

[kb1900]

the statistic "post-II acceptance rate" is still "accurate". Just because there may a time-dependent difference in the statistic between early and late interviews doesnt make it less accurate

Anyway, idk how you could ever quantify this.

 

[ace_inhibitor111]

You can look at UMich's graph here:

As you can see, extremely early interview dates have about a 10-15% higher chance, but the overall chances are more or less consistent even several months into the cycle. I feel this is the case for most schools. Interviewers on the first 1-2 days have a high chance of getting accepted but later interviews still have a decent shot.

 

[aalamruad]

A lot of members have been sharing statistics about specific school "Post-Interview Acceptance Percentages"

As a data-analyst by trade I was wondering, how accurate could these numbers possibly be?

For example, its pretty much accepted that ADCOMS will invite the top applicants for interviews first, early on in the cycle. They give interviews to the people they want to attend their school.

If we say a school as a 40% post-II acceptance rate, does that mean they are going to reject 60% of the applicants they want the most? I doubt it.

Wouldn't it be fair to say that the earlier on your II is, the greater your post II acceptance rate (assuming you don't say anything stupid during the interview, or just aren't a good fit for the school)

This hypothetical 60% rejection percentage would be skewed by outliers in both directions (extremely low acceptance late in cycle, extremely high early in cycle)

It'd be great to be able to crunch the numbers on this and see how much the spectrum varies over the course of the cycle.

Supposedly some schools accept over 90% of the first few groups of applicants they interview, so you're definitely right but it's not info that schools tend to release so I don't think you'll be able to find a way to crunch the numbers.

 

[aalamruad]

You can look at UMich's graph here:

As you can see, extremely early interview dates have about a 10-15% higher chance, but the overall chances are more or less consistent even several months into the cycle. I feel this is the case for most schools. Interviewers on the first 1-2 days have a ~15% higher shot of getting accepted but later interviews still have a decent shot.

Interesting. Those acceptance rates are super high compared to most schools though, so I'd expect some significantly different variations in schools with more normal post-interview acceptance rates (~40%).

 

[Shirafune]

I think it's highly school-dependent. Each school has a different way of approaching acceptances, but I would bet that most schools tend to accept early interviewees more frequently. As mentioned in this thread, utilizing post-II acceptance rates assumes that interviewees are equally desirable throughout the process. For many schools that send IIs later in the year for the waitlist, this is certainly not the case. For schools that send out small batches of acceptances throughout the year, the idea may be more applicable. Other things like acceptance-matriculant ratios (i.e. how attractive a school is to its accepted students) can confound the interpretation of post-II acceptance rates. How about medical student body fit? If a school puts a strong emphasis on finding the exactly right class profile, it might need to interview far more applicants to capture an appropriate group of acceptances. In the case of student body fit, does the "type" of applicant change over time (i.e. do early applicants tend to participate in X, Y, Z activities more than late applicants)?

It would be an incredible benefit to SDN if we could gather data on the temporal aspects of post-II acceptance rates.

 

[DingoPingo]

A study like this can only lead to more neuroticism. I vote no go

 

[aalamruad]

A study like this can only lead to more neuroticism. I vote no go

Could help people internalize the importance of applying early

 

[libertyyne]

You can look at UMich's graph here:

As you can see, extremely early interview dates have about a 10-15% higher chance, but the overall chances are more or less consistent even several months into the cycle. I feel this is the case for most schools. Interviewers on the first 1-2 days have a high chance of getting accepted but later interviews still have a decent shot.

Is this broken down by IS and OOS?

 

[Dagrimsta1]

Some people have waaaaaaayyyyyy to much time on their hands.

 

[DrHart]

I think there is some logic to your thoughts... but another point to consider. As the wise Goro says, "4.0 automatrons are a dime a dozen." It's logical to assume that the majority of candidates interviewed early in the season are the high scorers (3.9+, 520+ MCAT, etc.) There are plenty, I'm sure, that lack certain qualities and will be shot down to a waitlist or rejection (even if they interview early in the season).

In addition, there are plenty of 3.5-3.7, 32+ MCAT students who might not interview until later in the season (simply due to their scores) but have better ECs and interpersonal skills than the earlier interviewees. As the U of M graph shows, between the beginning and end of the season, the post-II acceptance rate really only goes down a little. It isn't as dramatic as you would expect if all the early candidates wowed the adcom and the follow up interviewees were second best.

No doubt post-II acceptance stats vary by school, but I wouldn't try to reason too much about it. For anyone who is granted an interview, whether in September or March, should be happy just about getting an invite.

 

[Goro]

Actually, that's exactly the case. You have to remember that people who get IIs are invited because they look good on paper. Showing up and doing well in interviews is an entirely different game.

If we say a school as a 40% post-II acceptance rate, does that mean they are going to reject 60% of the applicants they want the most? I doubt it.

Nope. Getting accepted is 100% on the interviewee.
Wouldn't it be fair to say that the earlier on your II is, the greater your post II acceptance rate (assuming you don't say anything stupid during the interview, or just aren't a good fit for the school)

This is a very human endeavor. And DrHart beat me to it!!!!

 

[FeinMS]

How accurate is the post II acceptance stats by 3rd party publishers such as US News Compass? I thought not many schools publish such data on their own...