<p>Some educational reformer, who didn’t like the US News and World Report ratings of colleges suggest that the best way to evaluate a college is to measure how much learning went on. Although this sounded great at first, my first thought was ," how can this be possible"</p>
<p>After giving this some thought, I may have a partial solution: Take the GRE subject area results for kids taking the GRE in particular majors. Thus, use Physics GRE results for each school, Math GRE results for each school.This would seem to measure actual learning in their majors!</p>
<p>Admittedly there are several drawbacks to this. One drawback that I can think of offhand would be that not every major has a GRE such as accounting, visual arts, and business management, engineerig etc., although CPA passing rates can be used in accounting.</p>
<p>Secondly, this might encourage schools to focus on what the GRE tests and not on the education, which is always a fundamental problem. However, I do think that this would be a better measure than what US News is doing.</p>
<p>Another alternative is to have faculty members rate schools based on what they know of their major! This would seem better than simply using college presidents in the “peer review” process.</p>
<p>Any method to rank colleges is problematic. Clearly the USNews methodology is seriously flawed for any number of reasons. However so is the methodology proposed by taxguy. Is there any doubt that the average GRE score of a physics major at P’ton will be better than the average at Penn State given the significant difference in the quality of the students both institutions admit? </p>
<p>In fact the exhaustive study by Kreuger and Dale suggest that career success is insignificanty impacted by the college a person attends but is very dependent on the talent and abilities of the student.</p>
<p>I have studied the post graduate success of students based on PhD statistics. It is not surprising to find the likes of Amherst and P’ton high on the list. However why does Oberlin top the list year after year? And even more, how does one explain the performance of St Olaf and Wooster grads? And at the university level, while MIT unsurprisingly tops the list in engineering, why does Rensselaer outperform Cornell by a slim margin? And in all disciplines, why does Dartmouth not crack the top 25?</p>
<p>And what about curricula? A son’s friend is majoring in Product Design and Innovation(an interdisciplinary architecture/art/mech eng program) which cannot be found at any other university. How does a ranking figure in that academic opportunity?</p>
<p>And who is to say that academics alone define the worth of a college to students? Might not having the opportunity to explore an interest in crew or sailing be as valuable as a senior seminar in Serbo Croation literature?</p>
<p>And much is made of the characteristics of the student body. Who is to say that my late nite discussions with Bruce Villanch(one of Bette and Whoopie’s comedy writers-you’ll remember his stint on Hollywood Squares) or my pickup game with future NBA’er Jim Cleamons were any less memorable than those of friends who attended ND or Cornell?</p>
<p>No, the worth of a college is far too complicated to allow for some statistical evaluation or ranking. Use them as a very general guide and for the raw information they sometimes provide, but aviod the pitfalls of then attempting to arrange them in some bogus numerical order.</p>
<p>You’d need a pre-test if you want to measure “learning.”</p>
<p>Imagine two institutions whose recent grads scored similarly on GRE or other subject matter tests for Physics. One might conclude that about the same amount of “learning” took place. </p>
<p>However, what if College A had a hihgly selective student body comprised of well-prepared students who had attended math and science high schools, had access to courses like AP Physics in high school, were participants in math olympiads, and so on. And College B didn’t. College B has students from average high schools, who had less enrichment before getting on campus. Now where do you think the most learning took place?</p>
<p>I agree with hoedown, you would need to pre-test to get the real outcome. The difficulty in doing this is that the SAT does not measure all subjects. The schools with the students going in with high SAT scores have an advantage. I would think that a school with mid-level SAT scores that produced high GRE scores would actually show the better outcome . . . not sure there is a way to actually implement this though conceptually that is the information all of us parents would like to see.</p>
<p>BTW, I like the NRC rankings but they are too spread out to be applicable today. I think the last one for math was 1993.</p>
<p>As it is, we muddle through by observing what all the other parents think is worth their hard-earned dollars and their children’s time. That’s not foolproof, but it applies a lot of brains and real-world experience to the problem.</p>