Approximate ranking percentile Forum

[Bobnoxious]

School has a mandatory curve of between a 2.7-2.8.
I just pulled a 3.4
Anyone have any idea about where that would likely put me in the rankings?
Top third?
Top quarter?
Top quintile?


Thanks

[zugzwanger]

I remember finding this on a previous thread a while back:

http://mylsn.info/rank-estimator/

[Bobnoxious]

Yeah, I've been there, thanks.

I don't have another data point, though.

[ScottRiqui]

Yeah, I've been there, thanks.

I don't have another data point, though.

No way to tell, then. Has your school ever posted additional data about previous classes? A 75th percentile value, perhaps?

[Bobnoxious]

Not that I've been able to find. It's Memphis, if anyone knows where I can find that information, I'd be grateful.

[BVest]

It seems like you'll have a second point of reference soon enough, but you won't need it.

13. CLASS RANKING
13.1 Full-Time Students
Full-time students will be ranked at the following intervals:
1st year: fall and spring semesters
. . .


13.2 Part-Time Students
Part-time students will be ranked at the following intervals with the
designated class:
1st year fall and spring semester with admission class

http://memphis.edu/law/currentstudents/ ... gs2013.pdf

[helfer snooterbagon]

Should be top half.

[History_Buff]

Possible it's top 15%.

[AReasonableMan]

I think a lot of those schools try to fail out a lot of people to keep the bar passage rates up so it's hard to guesstimate.

[arklaw13]

http://ualr.edu/law/academics/current-s ... fall-2013/

Here's an example of a ranking where the median is between 3.0-3.1. There a 3.4 would put you in the top 20% or so. So if your curve is between 2.6-2.8 I would expect to be in the top 10%.

[ScottRiqui]

http://ualr.edu/law/academics/current-s ... fall-2013/

Here's an example of a ranking where the median is between 3.0-3.1. There a 3.4 would put you in the top 20% or so. So if your curve is between 2.6-2.8 I would expect to be in the top 10%.

Unfortunately, a median by itself only tells you where the peak of the curve is; it doesn't tell you how wide or narrow the "fat" part of the curve is. You could have a school where the median is 3.0 and 3.4 is top 10%, and another school where median is 3.0 and 3.4 is only top 25%.

[arklaw13]

http://ualr.edu/law/academics/current-s ... fall-2013/

Here's an example of a ranking where the median is between 3.0-3.1. There a 3.4 would put you in the top 20% or so. So if your curve is between 2.6-2.8 I would expect to be in the top 10%.

Unfortunately, a median by itself only tells you where the peak of the curve is; it doesn't tell you how wide or narrow the "fat" part of the curve is. You could have a school where the median is 3.0 and 3.4 is top 10%, and another school where median is 3.0 and 3.4 is only top 25%.

Well in that particular example they don't have a true curve. They have a mandatory mean. The profs can give out whatever they want as long as the mean falls within the mandatory range, which is between 2.9 and 3.1 if I remember right. So the distribution will probably more normal-looking than a lot of median-based curves one could come up with. Or I could suck at statistics.

[ScottRiqui]

http://ualr.edu/law/academics/current-s ... fall-2013/

Here's an example of a ranking where the median is between 3.0-3.1. There a 3.4 would put you in the top 20% or so. So if your curve is between 2.6-2.8 I would expect to be in the top 10%.

Unfortunately, a median by itself only tells you where the peak of the curve is; it doesn't tell you how wide or narrow the "fat" part of the curve is. You could have a school where the median is 3.0 and 3.4 is top 10%, and another school where median is 3.0 and 3.4 is only top 25%.

Well in that particular example they don't have a true curve. They have a mandatory mean. The profs can give out whatever they want as long as the mean falls within the mandatory range, which is between 2.9 and 3.1 if I remember right. So the distribution will probably more normal-looking than a lot of median-based curves one could come up with. Or I could suck at statistics.

That's a good point - if you only have a mandated median, then it doesn't have to be a "curve" at all. In that case, even if you did have two or more data points, it wouldn't help you predict percentiles anywhere away from those points.

[Bobnoxious]

Thanks, all.

I feel like an idjit now. Yeah, it's a mandatory mean, not a curve.

[Bobnoxious]

Well, History_Buff damned near hit the nail on the head.

16th out of 117 w/ a 3.4

Just under the 14% mark.

Not bad for an old fart (48 years old). Not sure it'll do much for me in the way of paid summer gigs here in Memphis, though.

Now to fight off the feeling that there's absolutely no way in hell I can repeat the performance this spring.

Motivation is sorely lacking and the fatigue is debilitating.