Talking about Pritchett’s new book, which I really liked. Here is an excerpt from the podcast.
I do mention that Clay Christiansen has this idea of disruptive innovation. Which is where you actually moves to something that looks like lower quality but then rebuild a higher quality on top of that. The classic example of course is the PC (personal computer), which in computing terms when it came out in 1980 was a garage hobbyist toy that no serious computer engineer would pay any attention to. And all the firms that ignored the incipient disruptive innovation of the PC got themselves blown away by this, at the time, low-quality alternative. So I do think technology is going to change the way classrooms are managed in ways that are going to look disruptive, in the sense that they may appear to be de-skilling the classroom. But I think that in the long run there will be a disruptive innovation in the developing world that will rapidly accelerate the rate at which they can close on these higher levels of schooling. But when I hinted at this chaos–it’s going to be very chaotic. It’s going to be lots of people doing things that don’t look like finished classrooms, but produce incredible gains, and they are going to reconstitute a new way of doing education.
Keep in mind that I believe in the null hypothesis, which is that no education technique makes a big difference in terms of outcomes. Pritchett’s book actually offers a lot of support for that hypothesis, in that many results that he reports show little or no difference. However, he does offer one example, from Pakistan, in which giving parents “report cards” on school performance puts pressure on schools to improve and leads to some significant gains in the context of a controlled experiment.
“…moves to something that looks like lower quality…”
There’s an economics lesson hiding in that sentence, one which positively begs for a closer look. Namely, “quality” is multidimensional, and some of the dimensions are not defined in those terms that computer engineers, or engineers in general for that matter, deal in professionally.
At the time, the alternative to the “garage hobbyist toy” was the central computer or minicomputer – an early version of “the cloud”, if you will. Generally, you could become a User and access it only if you were a Very Serious Person with Very Serious Funding, or else a student paying handsome tuition. Such access originally entailed tasks such as “keypunching” holes into “IBM cards”, lugging “decks” of them to an office to “run” the “job”, and finally, at long last, getting the results hours later or the next day.
With time, Teletype terminals became more available. The service was often slow and atrocious except maybe between midnight and 4AM. You got only a minute fraction of the central computer, since dozens or even hundreds of other terminals were logged in. That could be less than you got out of a “garage toy” when you had the whole “toy” to yourself. (This is rather like the slow, ponderous, unreliable, overcrowded city bus that only runs at certain arbitrarily chosen hours, as against even a beater of a car that you can drive anywhere at any day and hour you that you might need.)
In other words, the “high” quality alternative was in many respects – but non-technical ones – of abysmally low quality indeed. And therein seems to lie an important lesson about how to frame and think about disruptive innovation.