Monday, January 24, 2011
Gordon Smith has posted the final rankings of his crowdsourced survey about law schools. Perhaps the single most important thing it teaches us is: brand matters. We all believed this already. But either accidentally or intentionally, Gordon tested this theory with his new rankings by including one law school under two names: Franklin Pierce and the University of New Hampshire. The two schools partnered this year and as of this past fall, the Pierce name was replaced by UNH. So which school does the crowd think superior? No surprise, State U fared far better than Franklin Pierce. UNH scored a 43; Pierce a 33. UNH jumped 20 spots in the reputation rankings simply by changing its name.
Of course, there might be substantive reasons for this. For example, perhaps the subjects understood that Pierce would be improving in the future due to support of a big university. But more likely they responded to their overall impression of the University of New Hampshire brand. If you think Pierce is pleased by this reputational lift, imagine the excitement among folks at Southern New England School of Law (aka UMass Law) as they imagine the incremental benefit of their merger!
ADDITIONAL COMMENT BY BRIAN LEITER: Dan beat me to the punch on this one, so I'll just add my comment here. Gordon's exercise seems to me a persuasive reductio of the 'method,' and not only because of the amusing result noted by Dan. Because Gordon's version was linked by the trashy "Above the Law" blog, there was an enormous influx of voters with dubious knowledge and motives, which might also explain the massive strategic voting (over 10,000 votes from Ann Arbor, Michigan, but less than 2,000 from Brooklyn, N.Y., that's not so good; and the nearly 20,000 votes from Washington, DC probably has more than a little to do with the rather robust performance of certain DC schools). All surveys are predicated on the idea that aggregating the partial information of a lot of respondents will produce meaningful results, but this looks more like aggregating the misinformation and bias of a large number of respondents.