Sign Up For Free

**Join for FREE**,
and start talking with other members, weighing in on community discussions,
and more.

Also, by registering and logging in you'll see fewer ads and pesky welcome messages (like this one!)

- Reply to threads, and start your own.
- Post reviews of your campus visits.
- Find hundreds of pages of informative articles.
- Search from over 3 million scholarships.

We've just launched our new college search tool at http://www.collegeconfidential.com/schools. Use the following form to provide feedback as we continue to work toward a more robust solution to best meet your needs: https://www.surveymonkey.com/r/CCSearchSurvey

CollegeBoundSoon
Registered User Posts: **30** Junior Member

Every school reports the "middle 50%" for SAT scores. What exactly does this mean? Does this mean 50% of all scores, or 50% of all students? Because if it's 50% of all students and not scores, there are probably a lot more scores at the lower end of the range. For example, it a school says the middle 50% is 1200 to 1350, is it likely that most of the scores within this range are closer to 1200 than to 1350? Does my question make sense?

Post edited by CollegeBoundSoon on

This discussion has been closed.

## Replies to: What does middle 50% really mean?

93Junior Member592Member17,472Super Moderatorhttp://collegesearch.collegeboard.com/search/CollegeDetail.jsp?collegeId=1251&profileId=6

SAT Critical Reading: 700 - 800

SAT Math: 700 - 790

SAT Writing: 690 - 790

ACT Composite: 31 - 35

http://collegesearch.collegeboard.com/search/CollegeDetail.jsp?collegeId=3261&profileId=6

SAT Critical Reading: 460 - 580

SAT Math: 460 - 590

SAT Writing: 430 - 550

ACT Composite: 20 - 25

The meaning is that at Harvard, a quarter (or more) of all enrolled students in the current freshman class had SAT critical reading scores of 800 (whew!), a quarter (possibly NOT the same quarter) of students had an SAT math score of 790 or higher, a quarter (again, maybe not the same quarter) had a writing score of 790 or higher, and a quarter (as before, not the same quarter as for the three other reported test sections) had an ACT score of 35 or higher. And, on the other hand, at Harvard one quarter of all enrolled freshmen this year have critical reading scores of 700 or below, and one quarter have math scores of 700 or below, and one quarter have writing scores of 690 or below, and one quarter have ACT scores of 31 or below. In general, most enrolled students at Harvard have high scores, and those who have the lowest scores are probably amazing in other ways.

By contrast, at the University of Nevada Reno, with a different size freshman class, one quarter of the enrolled class (which thus is a different number of students from one quarter of Harvard's class) have critical reading scores of 580 or above. It is perhaps possible that EVERY student at UNR has a lower score than ANY student at Harvard. Similarly, one quarter of UNR students have a math section score of 590 or above, and one quarter have a writing score of 550 or higher, and one quarter have an ACT composite of 25 or higher. I think you can work out for yourself what the bottom quartile scores mean. You are correct in figuring that it's theoretically possible that even at Harvard there could be some students with a section score on the SAT as low as the scoring scale goes, with the interquartile ranges still being true as reported above, but I doubt that's what actually happens at Harvard. The 25th percentile does NOT set a lower bound for the lowest possible score with which an applicant might be admitted.

75Junior Member30Junior Member626Member3,937Senior MemberCollegebound: I doubt that info is published for any school. It might follow a bell curve, but that is purely speculative.

17,472Super Moderator988Member34,105Super Moderator988Member2,298Senior MemberIn most cases, the number refers to ENROLLED students who took the test. One would think that the enrolled student population would be somewhat lower than the ACCEPTED student population (i.e., the middle 50% of students accepted would be higher than the numbers reported).

If it's a normal distribution, you would expect to see most scores near the median and smaller numbers at either end, but without seeing the actual data, who knows.