-
Is the source generic to all applicants or specific to your school data (such as in Naviance or SCOUR)? The stats for all students everywhere is fairly useless because results vary greatly based on your school profile. This is particularly true of general sites because they don’t take into account different GPA averages at different schools – not only that everyone weights differently but that some schools grade easier unweighted than others. Which is why the scattergram is useless unless the data is specific to your school.
-
A scattergram plots the average GPA and average test score (whether specific to your school or more generally). Which means that the true average for unhooked applicants is likely somewhere above the overall average, but the trick is you have no way of guessing how far above.
-
If you truly have access to your school specific data and for enough years and applicants to be statistically meaningful, the data can be directionally useful. However, in evaluating your chances don’t just look to see if some others have been accepted in your plot point area, but whether you are in the middle or better of the plot point for acceptances. Being near the bottom left of the acceptance cluster is not that reassuring.
Also, if you are using a scattergram at a site like Naviance, I suggest you filter out all the results except the exact scenario you are applying with. For example, if you are applying RD, then filter out all the ED, all the waitlisted and undetermined outcomes, etc. Much easier to compare the relevant outcome.