I've been jumping around reading Caplan's posts on your link in my free time today and I've found him very convincing. However, I know very little about economics. Could you recommend a good overview article on signalling/ability bias/human capital in higher education? I am sincerely quite interested in this stuff.
I did some googling and found this blog post that might be a more accessible overview of Caplan's position. I'm not an economist either and my understanding is that of a layperson.
Right now, the inaugural class of Minerva Schools at KGI (part of the Claremont Colleges) is finishing up its first semester of college. I use the word "college" here loosely: there are no lecture halls, no libraries, no fraternities, no old stone buildings, no sports fields, no tenure... Furthermore, Minerva operates for profit (which may raise eyebrows), but appeals to a decidedly different demographic than DeVry etc; billed as the first "online Ivy", it relies on a proprietary online platform to apply pedagogical best practices. Has anyone heard of this before?
The Minerva Project's instructional innovations are what's really exciting. There are no lectures. There are no introductory classes. (There are MOOCs for that! "Do your freshman year at home.") Students meet for seminar-based online classes which are designed to inculcate "habits of mind"; professors use a live, interactive video platform to teach classes, which tracks students' progress and can individualize instruction. The seminars are active and intense; to quote from a recent (Sept. 2014) Atlantic article,
It sounds to me like Minerva is actually making a solid effort to apply evidence-based instructional techniques that are rarely ever given a chance. There are boatloads of sound, reproducible experiments that tell us how people learn and what teachers can do to improve learning, but in practice they are almost wholly ignored. To take just one example, spaced repetition and the testing effect are built into the seminar platform: students have a pop quiz at the beginning of each class and another one at a random moment later in the class. Terrific! And since it's all computer-based, the software can keep track of student responses and represent the material at optimal intervals.
Also, much more emphasis is put on articulating positions and defending arguments, which is known to result in deeper processing of material. In general though, I really like how you are called out and held to account for your answers (again, from the Atlantic article:
Their approach to admissions is also interesting. The Founding Class had a 2.8% acceptance rate (a ton were enticed to apply on promise of a full scholarship) and features students from ~14 countries. In the application process, no consideration is given to diversity, balance of gender, or national origin, and SAT/ACT scores are not accepted: applicants must complete a battery of proprietary computer-based quizzes, essentially an in-house IQ test. If they perform well enough, they are invited for an interview, during which they must compose a short essay to ensure an authentic writing sample (i.e., no ghostwriters). After all is said and done, the top 30 applicants get in.
Anyway, I am a student and researcher in the field of educational psychology so this may not be as exciting to others. I'm surprised that I hadn't heard of it before though, and I'm really curious to see what comes of it!