what's the deal with msar median gpa?

[nycfella]

---

Members don't see this ad.

why not just print the average gpa's? I know a median is a better measurement in that "it gets rid of" outliers... but still most of the time averages are used...

 

[BigRedPremed]

You just said medians are better indicators. 😕

 

[braluk]

It's a good point. Sometimes I wished that the MSAR would print an average, provide a range of GPAs, all alongside a median. However, keeping in mind that the AAMC is a business as well, I would understand that these statistics may also dissuade applicants from applying to schools or use up the resources of the already bogged down adcoms by having to screen out even more students who were misled to believe that they are more easygoing on applicants with lower numbers within the given range set. There's something about leaving applicants in the dark, a little bit, that may be helpful for a business. Granted, a range of GPAs might actually help boost applications but I think there are unforseen effects on students applying to medical school (dissuasion or persuasion) that as a business, would be a risky move given that their chunk of change is from the AMCAS. But it's all a mystery when it comes to anything AAMC related

 

[bahblacksheep]

Statistically speaking, both average and median are measures of a center of a distribution. However, as someone already mentioned, the median would be best to eliminate outliers or a non-normal distribution that would misrepresent the center of the distribution. If the distribution does not have outliers and is normally distributed, then the median and average will be roughly the same. All to say, I'd trust the median more than the average if you don't know what the distribution is going to look like, just like you'd pick a nonparametric test over a parametric test.. Of course, the absolute best would be a histogram.

 

[estairella]

Of course, the absolute best would be a histogram.

I disagree. The absolute best would be a website where you submitted a digital photograph and it gives you an automatic acceptance, waitlist or rejection at every AAMC-listed school. 😛

 

[bahblacksheep]

Ah not for me. Not very photogenic. Or visogenic. LOL. I'd probably get a rejection from every school... but that wouldn't be so different than now.

 

[Cirrus83]

Well, I don't agree with the suggestion that posting the average GPA would dissuade more people from applying. If you look at the MSAR Stony Brook's median GPA is like a 3.72, but their average GPA is actually a 3.6 (according to both the admission office when I was there for my interview, and according to US News, who got the data from Stony Brook anyway).

So in that case the median would probably get less people to apply than the average would.

To be honest I think the MSAR people probably just felt the median was a more useful number. But seriously though, if I had only seen that 3.72 median I think I would have been quite a bit more scared since my own GPA is only a 3.55, so thank goodness I saw the 3.6 average first =P (not that I didn't apply to many schools with higher grades than what I have anyway, but still).

Anyways it'd probably be best to have both, but for what it's worth you can get the data if you want it bad enough, since the data schools give US News has the GPA average and not the median, so combined with the MSAR you'd have both (at least for schools that reported to US News)

 

[Wordplay]

Statistically speaking, both average and median are measures of a center of a distribution. However, as someone already mentioned, the median would be best to eliminate outliers or a non-normal distribution that would misrepresent the center of the distribution. If the distribution does not have outliers and is normally distributed, then the median and average will be roughly the same. All to say, I'd trust the median more than the average if you don't know what the distribution is going to look like, just like you'd pick a nonparametric test over a parametric test.. Of course, the absolute best would be a histogram.

I believe the MSAR has lots of distributions, including GPA (with graphs of applicants, accepted, not accepted).

 

[Captain Fantastic]

I believe the MSAR has lots of distributions, including GPA (with graphs of applicants, accepted, not accepted).

Yes, but they are not by school. The histograms are for all applicants across all schools. It would be awesome to see how my scores sat in the distribution for a particular school.

 

[Spathi]

Medians are better if you are a white mle without connections. Medians eliminate legacy policy people and minorities and people who paid their way in etc.

 

[Droopy Snoopy]

Medians are better if you are a white mle without connections. Medians eliminate legacy policy people and minorities and people who paid their way in etc.

What?

 

[jennifermov]

I disagree. The absolute best would be a website where you submitted a digital photograph and it gives you an automatic acceptance, waitlist or rejection at every AAMC-listed school. 😛

brilliant!!! i would even get my hair done for the picture.

 

[jojo_1981]

brilliant!!! i would even get my hair done for the picture.

Totally! Glamour Shots, here I come! 😀

 

[UMP]

1st and 3rd quartile numbers would be best