Correlation Between Time of Secondary -> Interview Invite and Acceptance Decision?

[slippydubloons]

---

Members do not see this ad.

Hi All,

(I want to preface this post by saying that I think stats is a cool field and am interested in gleaning relationships between sets of data)

I've seen a bunch of posts around trying to correlate random things with acceptance decisions and I recognize that due to different workflow reasons among schools, not much is really concrete. However, I was wondering whether there is a correlation between the timing of secondary submission -> interview invite and the final decision reached on an application.

I imagine that someone who gets an interview from School A within two weeks of submitting secondary is considered by admissions to be a good fit at that school and likely to get accepted there, whereas if they get an interview invite from School B four months after submitting their secondary they may not be as desirable a candidate/fit for School B and therefore be declined (assuming the schools are either both rolling or non-rolling).

I'm not asking whether it's "more likely" whether someone gets in or not based on the timing of their interview (e.g. whether there is an inherent boost to the application and/or if there is a bias in admissions between early and later interviewees), I'm just wondering whether there could be a statistical correlation derived from this data. And while I recognize that schools do things differently and look for various things when calculating interview order, I still wonder whether there's any useful info that could be gained from this.

Any thoughts?

 

[Lawpy]

What data are you going to use to determine correlation? SDN posts are a bad source since they are self-reported and we have to assume they're genuine (i.e. nothing stops SDNers from lying to protect themselves)

 

[libertyyne]

Hi All,

(I want to preface this post by saying that I think stats is a cool field and am interested in gleaning relationships between sets of data)

I've seen a bunch of posts around trying to correlate random things with acceptance decisions and I recognize that due to different workflow reasons among schools, not much is really concrete. However, I was wondering whether there is a correlation between the timing of secondary submission -> interview invite and the final decision reached on an application.

I imagine that someone who gets an interview from School A within two weeks of submitting secondary is considered by admissions to be a good fit at that school and likely to get accepted there, whereas if they get an interview invite from School B four months after submitting their secondary they may not be as desirable a candidate/fit for School B and therefore be declined (assuming the schools are either both rolling or non-rolling).

I'm not asking whether it's "more likely" whether someone gets in or not based on the timing of their interview (e.g. whether there is an inherent boost to the application and/or if there is a bias in admissions between early and later interviewees), I'm just wondering whether there could be a statistical correlation derived from this data. And while I recognize that schools do things differently and look for various things when calculating interview order, I still wonder whether there's any useful info that could be gained from this.

Any thoughts?

Some of the internal variables still apply that would make this meaningless. Your individual reviewer might be efficient, the committee may meet once time a amonth and you right before they meet. If you are a superstar pr a rare commodity you will receive quick review and quickly be put through the process. Dr.bob from earlier this cycle was an example. The dude got IIS without even completing secondaries. Each school may also prioritize different aspects of the application or may go in order received. It's like trying to read the tea leaves or goat entrails, and the biggest deterrent to this exercise is the lack of reliable data.

 

[RogueBanana]

I've actually posted before wondering about this same topic.

Some schools have tickers on their website that track how many IIs of the total have been offered. You could examine these and see if there is a significant disparity between the beginning of the cycle and the end. From what I have seen there is a small downwards trend after august, a lull around thanksgiving and a slight upwards trend in winter.

From there, you could assume that people who submitted secondaries in early had to have been a majority of those IIs in the beginning. Though over time the degree of correlation would weaken as the number of secondaries submitted increases and more "stellar" applicants enter the pool later on.

Either way there's really no way to be sure without having full access to any particular school's adcom data or somehow convincing the AAMC to include a question about date submitted/interviewed/accepted on their matriculant questionnaire

 

[slippydubloons]

Some of the internal variables still apply that would make this meaningless. Your individual reviewer might be efficient, the committee may meet once time a amonth and you right before they meet. If you are a superstar pr a rare commodity you will receive quick review and quickly be put through the process. Dr.bob from earlier this cycle was an example. The dude got IIS without even completing secondaries. Each school may also prioritize different aspects of the application or may go in order received. It's like trying to read the tea leaves or goat entrails, and the biggest deterrent to this exercise is the lack of reliable data.

I don't necessarily think it would be meaningless, per se. I think trying to draw any conclusions about causality would likely be futile, yeah, but in terms of a purely data-driven approach it could still be an interesting metric to look at.

What data are you going to use to determine correlation? SDN posts are a bad source since they are self-reported and we have to assume they're genuine (i.e. nothing stops SDNers from lying to protect themselves)

I think this is probably the issue, I would've liked to have put up a self-reporting survey about this response time and admissions decisions, but perhaps you're right in that the nature of an anonymous SDN population filling this out may be a no go for credibility.