Mathematicians Deconstruct US News College Rankings

<p>[Science</a> News / Rating The Rankings](<a href=“http://www.sciencenews.org/view/generic/id/37207/title/Math_Trek__Rating_the_rankings]Science”>http://www.sciencenews.org/view/generic/id/37207/title/Math_Trek__Rating_the_rankings)</p>

<p>

</p>

<p>Here’s the actual paper</p>

<p><a href=“http://arxiv.org/PS_cache/arxiv/pdf/0805/0805.1026v1.pdf[/url]”>http://arxiv.org/PS_cache/arxiv/pdf/0805/0805.1026v1.pdf&lt;/a&gt;&lt;/p&gt;

<p>

</p>

<p>Duh. :rolleyes:</p>

<p>I don’t understand why they said they had to reverse-engineer four pieces of data. There are about 15 components in the ranking, and USNews publishes 12 of them. The researchers said they didn’t know retention score. Really?</p>

<p>The results are interesting–it says that if you use these data, and assume they are accurate, then these are all the possible rankings an institution could have. You can alter the weights, but at the end of the day, using these data, an institution will only rise so far or drop so low.</p>

<p>I don’t understand the geometry part…</p>

<p>I do understand lists of rating criteria with associated scores, and weighting. e.g.</p>

<p>Published Criteria Data:
Peer Assessment: 8%, 85/100
Peer Assessment in one or two probable majors: 16%, 94/100
Admissions Selectivity: 11%, 91/100 etc. etc.
Median SAT Score (adjusting for a Single Sitting -> Superscore) which I doubt they do…
Median GPA
Research $ per student
Teacher/Student Ratio
Study Abroad Opportunities
Alumni Giving (adjusting for Private vs. Public entity) which I doubt they do
Campus Safety
Internship Opportunities (within Major)</p>

<p>Personal Criteria:
Weather
Distance from home
Campus Attractiveness
Social Style
Athletics
Quality of potential mates
… etc.</p>

<p>It seems all one would need is a spreadsheet, 1-100 scoring on the published criteria, and the applicants OWN 1-100 scoring wrt the personal criteria (weather, social, campus safety, etc,) and % for weighting. Voila, a personal ranking.</p>