Sample AAMC vs. Real MCAT

[aldol16]

Advertisement - Members don't see this ad

Hi guys,

Yesterday I uploaded a graph showing Kaplan FL scores vs. MCAT for those who took the exam in Jan. 2016. I took the time today to do the same for AAMC Sample exam vs. MCAT. It's posted below. n ~ 65 this time and again, I only included the most recent test-takers' data. r^2 was 0.62, so a much better predictor of how you will do on the real thing than the Kaplan FLs.

 

[FutureFloridaDr]

Niceeee! That feel when you're in the 64%-ile range though

 

[salemstein]

Is this based on %correct on test or percentiles?

 

[aldol16]

This is based on % correct. Percentile rankings are not available on the AAMC Sample.

 

[GCSDOC]

Hi guys,

Yesterday I uploaded a graph showing Kaplan FL scores vs. MCAT for those who took the exam in Jan. 2016. I took the time today to do the same for AAMC Sample exam vs. MCAT. It's posted below. n ~ 65 this time and again, I only included the most recent test-takers' data. r^2 was 0.62, so a much better predictor of how you will do on the real thing than the Kaplan FLs.

View attachment 201319

Any chance you can do this with aamc test #1 vs. actual MCAT?

Sent from my SM-A300H using SDN mobile

 

[aldol16]

http://forums.studentdoctor.net/threads/aamc-fl-1-vs-mcat.1190348/

 

[NextStepTutor_2]

Hi guys,

Yesterday I uploaded a graph showing Kaplan FL scores vs. MCAT for those who took the exam in Jan. 2016. I took the time today to do the same for AAMC Sample exam vs. MCAT. It's posted below. n ~ 65 this time and again, I only included the most recent test-takers' data. r^2 was 0.62, so a much better predictor of how you will do on the real thing than the Kaplan FLs.

View attachment 201319

Are these based on official reports, or just SDN submissions like previous attempts to analyze new data?
Be wary of reading much into any analysis with a sample size this small (and anonymously self-reported) and no useful controls for test results when making ANY conclusions or plans for your test date on the basis of these graphs.

No one test, AAMC or not, will tell you how prepared you are. It is anecdotally interesting though, and I admire your commitment to MCAT education @aldol16 .

Good luck to all!

 

[aldol16]

Are these baaed on official reports, or just SDN submissions like previous attempts to analyze new data?
Be wary of reading much into any analysis with a sample size this small (and anonymously self-reported) when making ANY reliable conclusions or plans for your test date.

These are based on SDN self-reported statistics from the January administration of the new MCAT. Although SDN is a biased sample compared to the pool of MCAT takers, the idea is that it might be a better sample of SDN users who are likely to be on SDN in the first place. Another concern is reporting error, where only people who are satisfied with their scores report them on here, which would tend to increase the disparity between their FL scores and their actual scores. This is the bias to keep in mind when assessing these data.

No one test, AAMC or not, will tell you how prepared you are. It is anecdotally interesting though. Good luck to all!

In terms of preparation, the AAMC exams (a combination of both of them) would likely give you the best estimate of how prepared you are. It's case of having to choose the best option from a pool of bad options. To date, most test prep companies exhibit more variability in their predicted scores as compared to the AAMC practice exams - again, an analysis that could be biased due to the nature of people reporting. However, the residuals are much smaller in the AAMC plots as compared to the Kaplan one I've done.