SAT-MCAT correlation

[panicking]

---

Members do not see this ad.

This is the formula I posted 3 months ago.

MCAT = 7.89 + .0182*SAT

Add a point or two if your Math score is about 20 points stronger.
Subtract a point or two if your Verbal(SAT) score is about 20 points stronger.
My SAT score is 690M/710V and my MCAT score is 11V,11P,10B,Q.
I am less "panicking" now.

 

[owen_osh]

Wow, this is pretty accurate for me. How did you come up with the equation?

 

[JulianCrane]

Yeah, that's a good question. Just exactly what is the derivation behind this formula? Do you factor in if someone had an off day while taking the SAT??

 

[Daredevil]

Okay according to the formula, I was supposed to get about a 36 on the MCAT. I think I will have to take it again to prove it right. HAHA. 😛 :wink: 😀

 

[pocwana]

I think panicking meant:

MCAT = 7.89 + .00182*SAT

But, since the coefficient for SAT is so small, it doesn't seem to be significant.

if someone got 500 on the verbal, his expected score is 8.8. if someone got 800, her expected score is 9.3. There's not much difference with the equation.
The point panicking made about the point difference between Math and Verbal seems valid and may be a significant factor in the ultimate score. Panicking, where did you come up with that?

 

[PADoc2be]

This is my first ever post! Yippee!!! Well, I think Panicking has the right formula. It turned out pretty accurate for me!!! 🙂

 

[panicking]

For composite SAT and MCAT scores, the formula is
MCAT = 7.89 + .0182*SAT
for verbal scores only, the formula is
Vmcat = -.83 + .01665*Vsat

Read the chapter Modelling of Data and pay attention to section Robust Estimation in the book Numerical Recipes. This formula is a lot more accurate than the linear regression. The program is smart enough to throw out irregular scores.