What to infer (and not to infer) from 25th-75th score ranges

<p>What I'm assuming</p>

<p>OK to infer:
- The 25th percentile score of a subsection of a test means that 25% of the class has scores at that score or lower. 75th percentile score means 25% are at that score or higher.</p>

<p>NOT OK to infer:
- Get the 75th percentile score for each subsection, add them up, and assume that the top 25% of the class has a total score of that or higher.
- Adding up the 3 50th percentile scores gives you the average composite score for the college</p>

<p>Anything else I should know? I'm just curious because some schools only give 25th-75th score ranges for subsections and not the entire test.</p>

<p>All correct. I don’t think there is anything else to get out of them, except you can infer that anything below around the 25th percentile is hooked (URM, Legacy, Development, Athlete)</p>

<p>You might also want to be careful about using the score data for any school that is test optional, because the scores that are reported are probably higher than those that aren’t reported.</p>

<p>Also, I have a pretty strong suspicion that for a white or asian student from a middle class or higher background, scores in the second lowest quartile (25th - 50th) are probably not very competitive unless the student brings more to the table – more could be some kind of academic or athletic hook, talent, or even just the ability to be full pay. In particular, girls who fit in the 25th-50th percentile profile who don’t have some other significant factors may find it tough sledding at selective LACs that are already swinging female in their enrollment.</p>

<p>A question: If you get all your scores above the 75th percentile in each category, you are pretty much in, right?</p>

<p>^at highly competitive places, no</p>

<p>^^at state schools and publics yes</p>

<p>

</p>

<p>At all but the top 5% of colleges this is highly likely unless there is an issue. At the top 2% the theory does not work at all.</p>

<p>so top 2% translates to, what, top 10, top 20 colleges?</p>

<p>2% of roughly 4,140 = 83.</p>

<p>More like 30-40 or so. 0.7 to 0.9% sounds about right.</p>

<p>And only 20 or so schools that must include EC for admittance.</p>

<p>I’d agree, it’s really top 1%.</p>

<p>“^^at state schools and publics yes”</p>

<p>Unless your applying to UT and not in the top 10%!</p>

<p>Unless it’s OOS at a highly competitive public, or a highly competitive private, then yes (as long as you don’t have something that is highly negative elsewhere, such as a criminal record or GPA of 2.0).</p>

<p>At quality publics, no. SAT/ACT scores are just one factor. You’ve got to have GPA/class rank to match, and the best of them will evaluate the strength of your HS curriculum. This applies equally to in-state and OOS applicants. It’s just flat wrong to lump all publics together on this. And it’s way wrong to assume that SAT/ACT scores alone will get you into any school in the US News top 50, whether it’s a research university (“national university” to use USN’s terminology) or LAC.</p>

<p>Actually SAT/ACT and GPA gets you into all but 30 schools in the nation.</p>

<p>Is it reasonable to say that the average composite score would actually be lower than the 50th CR + 50th W + 50th M?</p>

<p>Anyone know the answer to this?</p>

<p>
[QUOTE=Me, Myself, and I]
Is it reasonable to say that the average composite score would actually be lower than the 50th CR + 50th W + 50th M?

[/quote]
</p>

<p>ofcourse… not everyone who gets 800 on CR gets 800 on math too.</p>

<p>True, but we’re not talking about the 75th+ range.</p>

<p>It’s not out of the question for many people to have 3 x 720.</p>

<p>Example:
subject A B C D E
Math 70 50 50 25 55
Reading 70 20 70 40 50
Total 140 70 120 65 105
Mean scores are 50 for Math, 50 for Reading, but 105 for total.
When it comes to average, total average is always equal to sum of average of all.</p>

<p>Maybe, but the 50th percentile score is the median, not the average.</p>