For those of us who haven't bought the MSAR yet

[Drogba]

http://www.washington.edu/uaa/gateway/advising/downloads/gpamcat.pdf

enjoy.

Edit: HMS website says their avg GPA was a 3.76 thought it is reported as 3.87 in the above list. The first number is for matriculated students and for 2005 and the latter is for accepted from 08/09 MSAR. That is a huge difference between the two averages...

---

Members don't see this ad.

 

[Phenol312]

http://www.washington.edu/uaa/gateway/advising/downloads/gpamcat.pdf

enjoy.

Edit: HMS website says their avg GPA was a 3.76 thought it is reported as 3.87 in the above list. The first number is for matriculated students and for 2005 and the latter is for accepted from 08/09 MSAR. That is a huge difference between the two averages...

actually its median GPA not average GPA.

edit: sorry im not trying to split hairs or argue semantics but the language can be misleading.

 

[Drogba]

actually its median GPA not average GPA.

edit: sorry im not trying to split hairs or argue semantics but the language can be misleading.

Do you think the mean and median for this dataset would be that different? Since the median is near the max for the top schools you would expect the skew of the mean to be downwards for the class averages. Since the number of matriculants is even smaller than acceptees I guess the skew has a potential to be even greater. Considering the range of GPAs though is relatively small at these schools I'm not sure how much of a difference there would be. You aren't splitting hairs, the difference between the two measures can be significant in many cases, I just don't know if this is one of them.

 

[Phenol312]

Do you think the mean and median for this dataset would be that different? Since the median is near the max for the top schools you would expect the skew of the mean to be downwards for the class averages. Since the number of matriculants is even smaller than acceptees I guess the skew has a potential to be even greater. Considering the range of GPAs though is relatively small at these schools I'm not sure how much of a difference there would be. You aren't splitting hairs, the difference between the two measures can be significant in many cases, I just don't know if this is one of them.

it probably wouldn't be but since we dont have the raw data who knows.

how does this sound

if a school reports a median gpa of 3.6 that means a statistically significant number of individuals got in with a 3.2 as did a number with a 4.0. however, the majority of accepted students might have only had a 3.5 so instead of reporting the mean gpa of 3.5 they report the median gpa of 3.6. I think this tactic would be especially useful with science GPAs. Reporting the mean and not the median can make the school more competitive based on numbers alone.

Think of it if one med school had a mean gpa of 3.4 wed all probably be applying there and the school would be swamped with applications. but if they report the median they can just look like all the rest and maybe the applications would get spread around.

I'm just thinking of this now off the top of my head im sure there is some glaring flaw with my logic i just cant find it

anyone who does gets a cookie

 

[0Complications]

it probably wouldn't be but since we dont have the raw data who knows.

how does this sound

if a school reports a median gpa of 3.6 that means a statistically significant number of individuals got in with a 3.2 as did a number with a 4.0. however, the majority of accepted students might have only had a 3.5 so instead of reporting the mean gpa of 3.5 they report the median gpa of 3.6. I think this tactic would be especially useful with science GPAs. Reporting the mean and not the median can make the school more competitive based on numbers alone.

Think of it if one med school had a mean gpa of 3.4 wed all probably be applying there and the school would be swamped with applications. but if they report the median they can just look like all the rest and maybe the applications would get spread around.

I'm just thinking of this now off the top of my head im sure there is some glaring flaw with my logic i just cant find it

anyone who does gets a cookie

The only real reason one would use a median rather than a mean is too account for outliers in a data set. It is just the middle point, if the data set has an odd number it is the actual score of the middle point. If the data has an even number then the two center points are averaged.

So, yes. This could make a huge difference for a medical schools accepted GPA. But you also know that 50% of the people accepted were above that median point and 50% below it, so the difference could be minimal or it could be drastic. That's the beauty of statistics. You only know what the author wants you to know.

 

[Drogba]

it probably wouldn't be but since we dont have the raw data who knows.

how does this sound

if a school reports a median gpa of 3.6 that means a statistically significant number of individuals got in with a 3.2 as did a number with a 4.0. however, the majority of accepted students might have only had a 3.5 so instead of reporting the mean gpa of 3.5 they report the median gpa of 3.6.

If the majority of accepted students had a 3.5 it is unlikely that the median would be higher. Reporting GPA by mean would almost universally have the effect of a downward skew because of the range of GPA and the fact that the average/median GPA are near the maximum (4.00). This is not necessarily true, and I can't prove it without the raw data but it sounds reasonable to me. I doubt the mean of any GPA set for acceptees would be higher than the median.

Edit: I just ran a quick analysis on MDApp data (something close to a raw dataset of GPAs) and the median GPA for any accepted person was 3.68 vs a 3.64 mean.

 

[0Complications]

Edit: I just ran a quick analysis on MDApp data (something close to a raw dataset of GPAs) and the median GPA for any accepted person was 3.68 vs a 3.64 mean.

This means absolutely nothing. Only that you did a lot of work to get a set of numbers that only a small percentage of the applying/accepted students posted.

Just understand what median GPA means and use common sense.

 

[zmeister22]

Dang, yo. Are you trying to make me feel bad? I'm gonna go drink now...

 

[Drogba]

This means absolutely nothing. Only that you did a lot of work to get a set of numbers that only a small percentage of the applying/accepted students posted.

Just understand what median GPA means and use common sense.

Actually, it only took <5 minutes to do that. It means given a set of applicants' GPAs the mean is likely to be lower than the median. Whether this is true of each med schools' GPA set or not is a different story.

I'm not sure what your last comment is in reference to since this is merely a question of a statistical phenomenon and not supposed to shed some great light on the application process...