<p>
75th percentile…of course. ;)</p>
<p>barrons-
I read the article from the Wisconsin website. Yes, I can see that you reported their findings accurately. The main purpose of the article was to discover whether lowering course loads would increase GPA. ACT was an incidental variable. ACT was a stronger predictor for higher GPAs (3.5 or above) than lower GPAs (2.0 or above). The researchers used binary logistic regression as their statistical analysis which means they had to break down GPA into dichotomous groups such as above or below 2.0. I am not sure this is the best approach because it does not take advantage of the finer gradations in GPA outcomes that are available (maybe to the hundreths place).</p>
<p>But, my initial suspicion was correct. They analyzed individual students rather than groups of students. They were only able to account for about one third of the explanation for GPA among individual students using a combination of factors. That’s not bad for individual students.</p>
<p>However, if they had used group averages and percents they would have found that ACT alone accounts for about 70% of the explanation for retention percent. I did this analysis using IPEDS data but did not have access to GPA information.</p>
<p>Another way to say this is that it is hard to know how an individual student will perform but it is much easier to predict how a group of students will perform on average. If you were to increase ACT scores for the freshman class, it might be difficult to know which individual students will perform better but you can be quite sure that SOME students in the group will perform better.</p>
<p>25% - 75% SAT</p>
<p>1220 1400 New College of Florida</p>
<p>25% - 75% ACT</p>
<p>27 31 New College of Florida</p>
<p>NCF is a state university (part of the State University System of Florida) despite its small size (800 students) and despite its exclusive focus on undergraduate education.</p>
<p><a href=“http://www.collegeview.com/school/school_hub.jsp?scid=5500701&location=1500&pAreaId=null[/url]”>http://www.collegeview.com/school/school_hub.jsp?scid=5500701&location=1500&pAreaId=null</a></p>
<p>OP corrected 
to include New College of Florida in Top 10 Public Universities</p>
<p>Sorted first by SAT 25th percentile, then by SAT 75th percentile.</p>
<p>1250 1420 Georgia Institute of Technology-Main Campus
1240 1450 College of William and Mary
1230 1440 University of Virginia-Main Campus
1230 1430 University of Michigan-Ann Arbor
1220 1400 New College of Florida
1210 1390 University of North Carolina at Chapel Hill
1200 1460 University of California-Berkeley
1200 1430 University of Illinois at Urbana-Champaign
1200 1390 University of Maryland-College Park
1190 1350 SUNY at Binghamton</p>
<p>Second post corrected to include New College of Florida in Top 10 Public Universities</p>
<p>Sorted first by SAT 75th percentile, then by SAT 25th percentile.</p>
<p>1200 1460 University of California-Berkeley
1240 1450 College of William and Mary
1230 1440 University of Virginia-Main Campus
1230 1430 University of Michigan-Ann Arbor
1200 1430 University of Illinois at Urbana-Champaign
1250 1420 Georgia Institute of Technology-Main Campus
1220 1400 New College of Florida
1170 1400 University of Wisconsin-Madison
1150 1400 University of California-Los Angeles
1210 1390 University of North Carolina at Chapel Hill</p>
<p>ACT Post corrected to include New College of Florida in Top 10 Public Universities sorted first by ACT 75th percentile, then by ACT 25th percentile</p>
<p>27 32 College of William and Mary
27 32 University of Virginia-Main Campus
25 32 University of California-Berkeley
28 31 University of North Carolina at Chapel Hill
27 31 Georgia Institute of Technology-Main Campus
27 31 New College of Florida
27 31 University of Michigan-Ann Arbor
26 31 University of Illinois at Urbana-Champaign
25 31 Missouri University of Science and Technology
24 31 University of California-Los Angeles
26 30 Colorado School of Mines</p>
<p>Not sure about the intended aim of the “Top 10” list. The schools are public, but they have differences that can distort the implied comparison of schools based on SAT/ACT. It seems some of the school scores are inflated because they don’t offer education for “lower prestige” areas, such as agriculture. Also, it seems some school scores are inflated because they limit or don’t attract many international students. A school with a higher score doesn’t necessarily mean it is better. A school score not in the “Top 10” may be reflecting a more diverse education program; and it can have departments higher ranked than those in the “Top 10” group. If I were seeking a chemical engineering education, I’d welcome Minnesota, Wisconsin and Texas over most of the schools shown on the “Top 10” list.</p>