Sunday, May 25, 2014

An Opening for a Student Entrepreneur?

The Crimson has a follow on story about the decision not to publish the difficulty scores for courses. In a year when students have been pretty apathetic -- the student body did not come together around the Occupy movement, or fossil fuel divestiture, or even the controversies surrounding sexual assault -- it seems that everybody is angry about the Q guide.

The comments on the original story include some proposing that students take control. Unfortunately some of the history is misstated. Take this comment, for example:
What current students may not know is that the Q guide used to be a student run operation that printed candid and unsympathetic evaluations, with minimal censorship and often a great deal of humor. It did not protect the faculty from the truth being told about the quality of their teaching and course materials. But then the faculty took over the Q, and under the guise of improving the evaluations, created all kinds of avenues for faculty to opt out of letting the students know the truth about their courses. Now, faculty with a bad review are often not named as instructors. When the student comments are generally negative, they aren't printed. Some prominent life sciences courses, despite public assertions to the contrary, are so poorly regarded by students that their comments haven't been printed for years. The students would be best served by once again taking ownership of the evaluation process - by collecting and publishing their own evaluations.
 The actual history goes like this. The Crimson used to publish a "Confidential Guide," a.k.a. the Confi Guide. It contained advertising and was sold for a few bucks. It was random in its coverage, irreverent, humorous, and sometimes cruel. I include for illustrative purposes the review of my own course from the 1978-79 Confi Guide.

(I can't deny it all. Yes, I used to do things in class for which I would probably get arrested today. But I am sticking to my story that I had absolutely nothing to do with my TFs wheeling that keg of beer into the Science Center lecture hall on one festive occasion in the spring of 1981 and passing cups of brew out to the whole AM 110 class.)

In 1975, Harvard went into competition with the Confi Guide, publishing student course evaluations under the auspices of the Committee on Undergraduate Education. The volume was called the CUE guide as a result. In its early editions (CUE is a student-faculty committee, and I was on it for several years starting in 1975), the CUE guide included direct quotes from student questionnaires, typically zingers the editors had picked out to spice up the prose. Some faculty protested, and the FAS was paying the bills, so no more direct quotations. Also, originally individual professors could opt out, but the faculty legislated mandatory participation just a few years ago. So the above suggestion that faculty can opt out isn't true, but it's also true that the prose is now unspeakably dull, since it is for all intents and purposes mechanically generated from the numerical summaries of the questionnaires.

The CUE guide eventually put the Confi Guide out of business. And eventually control of the guide passed out from the authority of the CUE so the guide was renamed the Q.

Over the years there were charges of administrative censorship, which is of course exactly what was going on. It was one of the thankless jobs of the dean of undergraduate education to get the Guide produced by editors who were supposed to be producing a student guide but were in fact under administrative control. Was that censorship? "If it walks like a duck and acts like a duck, I guess it’s a duck," said one dean. (Note: sone of the dates in that story are plainly wrong.)

Is there really enough student interest to produce a meaningful guide outside official control? Another commenter is skeptical and has an interesting suggestion about how to piggyback on the Q data collection, with which which students are incentivized to cooperate because they get to see their grades a few days earlier if they complete their questionnaires.
Unfortunately, there's no way to make a survey with as high of a participation rate as the Q guide. A student-run evaluation couldn't incentivize so many students to report in the same way the Q can, since the College can keep us from seeing our grades. [a CS student] suggested a Chrome extension that registers a person's input to the Q guide - they'd only have to fill it out once, but ALL data (including student comments) would be recorded and could be made public. Can someone make that and publicize it, please? It'd be anonymous, and a student would only have to add it once to Chrome.
It's an interesting idea, but I wouldn't really recommend it. Sounds sketchy to me, and you'd have to get massive compliance with installing the extension, and would have to get new students to do it every year.

But in the Internet era, it shouldn't be too hard for for some enduring institution (The Crimson? The UC?) to provide enough small incentives to get students to fill out a simple survey about the courses they took and to web-publish a quick summary of the results. If they wanted to provide a professional service, they could even improve on the Q by asking students to evaluate their courses a second time a year later. The quality and value of a course are likely better judged with the perspective of some passage of time. Anyway, people probably thought Angie's list and Amazon reviews would never work either. And I have previously documented how Facebook itself came into being when Zuckerberg decided to do something Harvard had thought of doing but hadn't gotten around to doing. What evaluation services people would find useful enough to help build, I certainly wouldn't want to predict.

In other words, it's just possible that Harvard's move to control the difficulty information could prove to be the straw that broke the back of the already overweight censorship quality control mechanisms that have been put in place on the Q over the years. In the era of cheap web publishing, there may be an opportunity there for some enterprising student.


  1. I've heard that IBM and other companies knew that search was going to be important and even developed some search engines but never tried to make a product out of it because they didn't know how to make money off of it. While Google likely had a better product, the true genius may have been figuring out how to make money.

    Same here- can someone get the students to fill out the forms (perhaps with a small financial inducement) and also make money off of this?

    1. The problem is that the only people who would be interested are Harvard undergraduates and that is a few thousand people. But the Confi Guide managed to survive on that readership for years, and it was more expensive to put out than an electronic document would be.