GiveMeGrayIcons wrote: oldercoldervoice wrote: GiveMeGrayIcons wrote:
Our conclusions would be the same
GPA is not a good indicator of LSAT scores. That's what regressions are for, predictive analysis. However, I do think that having an actual versus predicted LSAT score graph would be kinda cool
I realize you haven't been ITT for long but I do actually know what I'm talking about. There's a pretty big grE
y area between "these two datasets have absolutely nothing to do with each other" and "this is a perfect 1:1 correlation." You don't just throw your data in the trash if the R squared is less than 95% lmao.
We both know what we're talking about. And I didn't say that we should throw our data into the trash. I just don't think that since our R squared is low our data is suddenly trash. It could be that GPA is a poor indicator of LSAT scores. And also--feel free to call me a noob--what is ITT???
I just don't think GPA is such a completely poor indicator of LSAT that 3% would really be the R Squared value for accurate data. It was a small sample from an informal survey which asked for predicted rather than actual LSAT scores, hence why I think the data is at least a little bit terrible. No offense Vulpes.
But I really do want that actual vs. predicted LSAT graph. I'm gonna need everyone ITT (in this thread) to make a commitment to answer that survey when it comes out, capisce?