What to make of the new MCAT score percentiles?

[Astra]

Link: https://aamc-orange.global.ssl.fast...repercentileranks_and_bar_chartsv2_merged.pdf

---

Members don't see this ad.

 

[markovchn123]

This trend is what I thought was going to happen. After taking the MCAT in January and seeing how one of my sections was at the level of section bank difficulty, I can see that AAMC is making the exam more difficult. The MCAT is going to turn less into a content based exam and more of a reasoning exam. Most scores are now up at least 1 percentile, meaning it's harder to get that score, or less people got it overall. I can see this trend continuing in the future.

 

[piii]

So if you took the new MCAT before this date, does that mean that after May 1st, your percentile is increased by 1-2 points but your numerical score stays the same?

 

[efle]

So if you took the new MCAT before this date, does that mean that after May 1st, your percentile is increased by 1-2 points but your numerical score stays the same?

Correct. Back during the first couple months a 515+ was awarded to 7% of people. In the year since, it was awarded less frequently, so now only 6% have a 515+. So your percentile climbed 93rd to 94th when they added all the new scores to the pool!

 

[CaffeineJunkie]

.

 

[XxThaDoggxX]

Does this mean my 506 may eventually be ~ a 30?

 

[efle]

In your opinion, does have any impact on admissions once these new percentiles become more publicly known? For instance, a 509 is now 82nd percentile, which is closer to a 31 than a 30 as previously reported.

I really doubt a 2 percentile difference will make or break anyone's outcomes

Does this mean my 506 may eventually be ~ a 30?

I doubt the percentiles will fluctuate that much (6 percentiles). On the old exam the yearly changes were usually less than a percentile. This one was probably the largest shift we will see since it was jumping from 2 months of data to a full year.

 

[flyingeagle]

For those (including me) taking the MCAT April 23rd, will my score be given to me as if it is before this change or after this change?

 

[efle]

For those (including me) taking the MCAT April 23rd, will my score be given to me as if it is before this change or after this change?

After, these new ranks are built from the April-Sept 2015 data.

 

[radioactive15]

Correct. Back during the first couple months a 515+ was awarded to 7% of people. In the year since, it was awarded less frequently, so now only 6% have a 515+. So your percentile climbed 93rd to 94th when they added all the new scores to the pool!

Is the new percentile for people who took it prior to May updated on the MCAT score summary or is this something adcoms will manually add to old takers?

 

[efle]

Is the new percentile for people who took it prior to May updated on the MCAT score summary or is this something adcoms will manually add to old takers?

If you have a score can't you check and see if it shows the new percentile?

 

[CaffeineJunkie]

.

 

[Cotterpin]

I'm not seeing a big change. My score and percentile are exactly the same as they were before.

 

[efle]

I'm not seeing a big change. My score and percentile are exactly the same as they were before.

Many stayed the same, others moved by 1-3 points

 

[ace_inhibitor111]

I'm confused as to whether the new percentiles include the January scores. Yes, the AAMC claims that the data is for tests administered April-September 2015, but this document says that there were 40,500 scores for tests April-September:

https://www.aamc.org/download/448800/data/validityplansforthenewmcatexam.pdf

Whereas the new percentiles includes data from 65,000 scores. There must be an error somewhere. I would think the AAMC would want to include scores from April-January to be considered a 'full year'

 

[efle]

I'm confused as to whether the new percentiles include the January scores. Yes, the AAMC claims that the data is for tests administered April-September 2015, but this document says that there were 40,500 scores for tests April-September:

https://www.aamc.org/download/448800/data/validityplansforthenewmcatexam.pdf

Whereas the new percentiles includes data from 65,000 scores. There must be an error somewhere. I would think the AAMC would want to include scores from April-January to be considered a 'full year'

Perhaps one is counting retakes and the other is not? Where does it show 65k included in new scores?

 

[ChrisMack390]

Aw mine stayed the same - was hoping for a boost 😉

 

[p0gono]

The 523-527 scores went from "100%ile" to ">99%ile". It seems like a reasonable way to categorize them, but I wonder why they saw the need to change it (or why they originally put them all as 100).

 

[ace_inhibitor111]

Perhaps one is counting retakes and the other is not? Where does it show 65k included in new scores?

https://www.aamc.org/download/454206/data/mcatatglance2015.pdf

Retakes aren't enough to explain the difference because only 7% of people had retakes. Either the dates on the PowerPoint or the dates on the new percentiles are incorrect. I think the dates on the new percentiles are incorrect, because why wouldn't you include all the scores on the previous tests in your statistics?

 

[PatchAdams25]

Wait so I'm confused.... I checked my score on AAMC and it says 507 ~74%..... should my new percentile be 76%? so essentially a 29.5 (old scale was 73% = 29 and 29% = 30)

Will it update or did I score in 74%? (January 2016 test)

 

[efle]

The 523-527 scores went from "100%ile" to ">99%ile". It seems like a reasonable way to categorize them, but I wonder why they saw the need to change it (or why they originally put them all as 100).

They can't seem to decide what they want to do at the high end. For my testing pre-2015 the max bin was 99.9th, then it got updated to 100th (they started rounding to the nearest percentile) and now it's >99 for 523-527 and 100th for 528. Hopefully now they finally settle down and stick with something.

https://www.aamc.org/download/454206/data/mcatatglance2015.pdf

Retakes aren't enough to explain the difference because only 7% of people had retakes. Either the dates on the PowerPoint or the dates on the new percentiles are incorrect. I think the dates on the new percentiles are incorrect, because why wouldn't you include all the scores on the previous tests in your statistics?

Hmm. Pretty annoying that they don't tell you the sample size on the actual percentiles document like they used to.

Wait so I'm confused.... I checked my score on AAMC and it says 507 ~74%..... should my new percentile be 76%? so essentially a 29.5 (old scale was 73% = 29 and 29% = 30)

Will it update or did I score in 74%? (January 2016 test)

Maybe they'll update it on May 1st?

 

[radioactive15]

Yup, my 9/15 score and percentiles also stayed the same

 

[deleted482591]

What's with the 100-> ">99" .... do they ever explain why they make these random changes?

 

[popopopop]

What's with the 100-> ">99" .... do they ever explain why they make these random changes?

Probably because 100 is reserved for 528.

 

[freak7]

What's with the 100-> ">99" .... do they ever explain why they make these random changes?

100th Percentile would be as good as or better than 100% of test takers. If a 523 is as good as or better than 99.5% of test takers, it'd be reasonable for them to categorize the score as >99th percentile to distinguish it from the 99th percentile.

The 523-527 scores went from "100%ile" to ">99%ile". It seems like a reasonable way to categorize them, but I wonder why they saw the need to change it (or why they originally put them all as 100).

They likely put the scores as 100th percentile because at the time, that *was* the highest score attained. After a few months, as more people took the test, the bell curve filled itself out and higher scores started trickling in. All it takes is one 528 to make the 523-527 scores all non-100th percentile.

 

[ace_inhibitor111]

100th Percentile would be as good as or better than 100% of test takers. If a 523 is as good as or better than 99.5% of test takers, it'd be reasonable for them to categorize the score as >99th percentile to distinguish it from the 99th percentile.



They likely put the scores as 100th percentile because at the time, that *was* the highest score attained. After a few months, as more people took the test, the bell curve filled itself out and higher scores started trickling in. All it takes is one 528 to make the 523-527 scores all non-100th percentile.

Previously they just rounded anything above 99.5 to 100. The old MCAT pdf had 39 as 100 when it is really 99.4-99.6 percentile. They should just give us a specific number for the top and bottom range instead of making us play Jenga with the bar graph.

 

[freak7]

Previously they just rounded anything above 99.5 to 100.

My math degree just burst into flames

They should just give us a specific number for the top and bottom range instead of making us play Jenga with the bar graph.

Agreed

 

[freak7]

I see this at the bottom of my scores: "The percentile ranks of scores are the percentages of test takers who received the same scores or lower scores than you did. The percentile ranks are updated on May 1 every year to reflect the results from previous calendar year(s)." so maybe they'll reflect the new percentiles may 1st?

That's the hope

 

[deleted482591]

So the new percentile ranks are going to be better?

 

[songl4a]

.

 

[freak7]

So the new percentile ranks are going to be better?

Better in terms of higher resolution, yes. Better in terms of "will my percentile increase?", it depends.

 

[Lucca]

Will the updated percentiles be sent to schools that are still reviewing us?

If adcoms access MCAT scores like we do (online) they should be able to see the new percentile if it has indeed been updated (as should we).

However, in all honesty, I don't think a 1-2% is going to take someone from "reject" to "accept" or even "waitlist" to accept, or "reject" to "interview"

 

[Lucca]

I think small differences can affect borderline applicants. 1% difference for me from somewhere between 32-33 to an actual 33 equivalent.

Do you think any adcom thinks that is a meaningful difference though? moving from 33 to 36 (93% to 97% - ish) is a big deal. 32 to 33? There is no real difference captured there. In fact, the confidence interval on your MCAT is +\- 2 points to begin with

 

[MEDINALDO]

They can't seem to decide what they want to do at the high end. For my testing pre-2015 the max bin was 99.9th, then it got updated to 100th (they started rounding to the nearest percentile) and now it's >99 for 523-527 and 100th for 528. Hopefully now they finally settle down and stick with something.


Hmm. Pretty annoying that they don't tell you the sample size on the actual percentiles document like they used

 

[Avicenna]

Yeaaahh no more 89 - I broke 90!!!!

 

[drechie]

Yeaaahh no more 89 - I broke 90!!!!

did you check to see if your mcat percentiel was updated on the aamc webpage?

 

[ERDOC555]

did you check to see if your mcat percentiel was updated on the aamc webpage?

They aren't updated yet, I just checked

On a side note I got an extra 1%!!! Hell yeah, that is definitely going to get me off the waitlist at a couple schools, right? (Sarcasm)

 

[kanda1o]

They won't be updated on AAMC until May 1st. My total % went up 2 points. Does anyone have the link for the score converter chart, from new to old? Curious to see if that 2 point increase changes the conversion to the old score?

 

[drechie]

They won't be updated on AAMC until May 1st. My total % went up 2 points. Does anyone have the link for the score converter chart, from new to old? Curious to see if that 2 point increase changes the conversion to the old score?

Okay that's what I was wondering because I didn't see an official change in my score. Good to know

Sent from my SAMSUNG-SM-N900A using Tapatalk

 

[ZPakEffect]

Deleted.