RSS
Facebook
Twitter
E-PrinceEvery November and December I spend quite a bit of time writing letters of recommendation for students past and present. This activity peaks in the late fall because the deadline for most graduate school applications is the end of December ...(back to the article)
The opinions expressed here are those of the individual commenters and do not necessarily represent the views of The Daily Princetonian Publishing Company, Inc. We do not take responsibility for the opinions, facts, or claims presented by individual commenters, and reserve the right to moderate or delete inappropriate comments.
I suspect the system should be able to work, *if* a recommender can establish a reputation as an unbiased and reliable evaluator. So, if you consistently rate students as you think they fit within the group at Princeton (not necessarily along all those dimensions, granted, but the important few), and these recommendations are consistent with how these students later do, you word will begin to carry weight. In fact, since your retirement date is uncertain, this qualifies as a repeated game with an infinite horizon, where your threat to stop giving quality recommendations would convince those who ask to respect the ones you do give.
Some people *are* objectively better than others; it's not all fuzzy wuzzy. At times a single numerical rank may not capture all dimensions of the variation among people, but that's not the same as claiming that the idea of ranking is flawed. If anything it means that we need more ranks, to capture more of the differences in ability between people.
Guys, guys, maybe you haven't learned this yet because one of you is a sophomore, but a topic in almost every seminar and lecture I've been in at Princeton is "The Persistence of Unavoidable Bias." Think about it.....history, art histort, English, psychology, the sciences. No human carries out any action without a PERSPECTIVE and thus even lifting a glass of water to your mouth is done as it is influenced by the rest of your life. We are not robots. No one is unbiased.
If I can obtain a list of numerical ratings for each of the applicants to my program, then I can create a simple numerical algorithm for choosing the ones I accept. And if I am questioned about bias in my selection process, then I can simply produce the numbers supplied to me by others, along with the algorithm. Furthermore, the hard work of actually thinking critically about the applicants is delegated entirely to the distinguished faculty providing the information.
Seriously, I don't think it has come to this. But at some level this view probably explains the increasing demand for numerical ranking to some extent.
See I do not believe that an evaluator can be made unbiased. What incentive does an evaluator have to postulate some algorithm that would evaluate students on multiple characteristics - surely none. If anything, putting a check in the higher boxes is not only easier, it may offer them some sort of prestige to know that their student is filling some top-notch position. And, besides, even if you could pin someone against their peers you are still left with having to pin their peers against everyone else that could quite possibly be competing for the job.