Preliminary percentiles

[MCATforsale1]

Hey guys, few questions about MCAT preliminary scores.

1. What are preliminary percentiles?
2. How soon after our test do we get them?
3. Are they accurate?
4. If they can provide us estimates, wouldn't that mean our test is scored? Why not just release the score?

Thanks

---

Members don't see this ad.

 

[a_borgia]

1. Percentile ranges that your actual score will most likely fall into. 10% ranges for the total and 15% ranges for sections.
2. If you took june, today. If you are taking a later test, no ranges will be released, you just have to sit tight for scores.
3. Yeah, sort of. 90% of actual scores will fall into the prelim range given them. That means there's a 1/20 chance of doing worse and 1/20 chance of doing better.
4. Not really, they have some granularity on the data, but they are still in the process of throwing out questions and such. These are very indicative but the AAMC wants to be 110% sure they don't mess anything up.

If you want detailed statistical/numerical breakdowns on what these numbers mean, join us over at the score compilations thread!

 

[MCATforsale1]

1. Percentile ranges that your actual score will most likely fall into. 10% ranges for the total and 15% ranges for sections.
2. If you took june, today. If you are taking a later test, no ranges will be released, you just have to sit tight for scores.
3. Yeah, sort of. 90% of actual scores will fall into the prelim range given them. That means there's a 1/20 chance of doing worse and 1/20 chance of doing better.
4. Not really, they have some granularity on the data, but they are still in the process of throwing out questions and such. These are very indicative but the AAMC wants to be 110% sure they don't mess anything up.

If you want detailed statistical/numerical breakdowns on what these numbers mean, join us over at the score compilations thread!

Thanks for the reply.
Do you think the fact that they're still filtering out the garbage from the test could cause infalted scores early on (i.e. june and earlier)?

 

[a_borgia]

Thanks for the reply.
Do you think the fact that they're still filtering out the garbage from the test could cause infalted scores early on (i.e. june and earlier)?

Inflated? Math says it shouldn't because everyone gets the garbage questions nixed, but in reality the AAMC is erring on the side of caution.

https://www.aamc.org/students/download/434504/data/percentilenewmcat.pdf

Look at that link. On the second page, you can see how verbal, an untouched section, has a perfectly smooth bell curve. BB and PS have jagged curves with a dip at 50% because they pushed any borderline scores to the next highest one rather than simple rounding down OR up (probably). This means there's a little bit of inflation, but it's super unlikely that it's net more than 1 points worth on any given person's exam.

 

[MCATforsale1]

Inflated? Math says it shouldn't because everyone gets the garbage questions nixed, but in reality the AAMC is erring on the side of caution.

https://www.aamc.org/students/download/434504/data/percentilenewmcat.pdf

Look at that link. On the second page, you can see how verbal, an untouched section, has a perfectly smooth bell curve. BB and PS have jagged curves with a dip at 50% because they pushed any borderline scores to the next highest one rather than simple rounding down OR up (probably). This means there's a little bit of inflation, but it's super unlikely that it's net more than 1 points worth on any given person's exam.

If they throw out items though, would a person who got that item correct still get credit? This is where the inflated scores could come into play, don't you think? I understand what you're saying btw. Statistically, it's not going to be a major shift because it's such a high stakes situation. They'd control for variance way ahead of time. My point is just to suggest that some people could benefit at the early moments.

 

[a_borgia]

If they throw out items though, would a person who got that item correct still get credit? This is where the inflated scores could come into play, don't you think? I understand what you're saying btw. Statistically, it's not going to be a major shift because it's such a high stakes situation. They'd control for variance way ahead of time. My point is just to suggest that some people could benefit at the early moments.

No, throwing it out means that your score if you got it right goes from, say, 47/59 to 46/58 and if you got it wrong, it would go from 47/59 to 47/58. It shouldn't really hurt as long as you are solid on concepts, because everyone is, in the end, scored on the exact same things. We just might have fewer questions counted, which reduces granularity but does not inflate scores. Erring on the side of caution, however, will inflate scores, but everything else is controlled for through percentile based scoring.