Shameless cross promotion - you should check out my thread on how to estimate admissions chances:
http://talk.collegeconfidential.com/what-my-chances/1795057-chance-your-own-danged-self-heres-how-p1.html
One of the biggest misunderstandings I see on this board (especially in the “Chance Me”
threads) is a failure to consider GPA as important as standardized test scores. There is ample
evidence that most colleges consider GPA and test scores equally, and plenty of evidence that
GPA is a superior predictor of college success.
I think the main reason that people focus on test scores to the detriment of GPA is simply
structural - there is a framework (the 25th/75th percentile reports of incoming freshmen) for
evaluating test scores and the information is widely available. For GPA, the admitted freshmen
data is inconsisently reported, which makes quick comparisons more difficult. However, I think
there is data available that can be used, and if we want an accurate estimate of admission
chances, it should be used, even if it might require some estimation/guesswork.
There are several pieces to the puzzle:
- Convert GPA (unweighted) to a SAT score.
- Estimate the 25th/75th GPA for the incoming freshman class at a school using the available
data. - Account for “course rigor” for the applicant.
- Guess-timate what a particular schools incoming freshmen “course rigor” might be.
A quick note on why I am using the SAT scale and not the ACT:
- Despite the gains made by the ACT relative to the SAT, the SAT remains the choice at most
top schools. - The resolution is higher - the SAT is a 60 points scale versus the 36 used by the ACT.
- The effective resolution is higher - about 53 for the SAT and about 25 for ACT (this is the
range from the 1% point to the top) - It’s just my preference - my post, my method, it is easy to convert my results back to an
ACT score if that’s what you prefer.
Ok, let’s get started with converting raw unweighted GPA to the SAT scale.
Back in the days of old, before grade inflation, the generally accepted method (still the
basis for the University of California method) was to simply multiply GPA by 200. Boom, you
are done, there’s still time for tea! So, you ended up with the following table:
4.00 800
3.75 750
3.50 700
3.00 600
2.50 550
2.00 400
However, this was devised back in a strange and magical time when students actually got C’s on
a regular basis, and the median high school grade was a B-/C+. Obviously, this is not the case
today, and so we need to adjust the conversion to reflect the current environment. My goal was
to come up a simple formula such that the GPA percentiles lined up reasonably well with the
SAT percentiles.
I derived the second column of the following table from a variety of sources, but primarily
from a series of studies done by the ACT people on grade inflation:
4.00 99% 800
3.75 98% 725
3.50 90% 650
3.25 75% 575
3.00 50% 500
2.75 28% 425
2.50 8% 350
2.25 2% 275
2.00 0% 200
The formula for the third column is GPA800 = 200 + 300 * (GPA - 2.0)
This actually lines really nicely with the equivalent SAT percentiles.
So, that’s 1) from the list above. In the next post, I will outline how to estimate the
25th/75th percentile GPAs from the limited / inconsistent data that schools provide.