3 votes

Cost Matters: Why Lambda School should have a lower success rate than college

3 comments

  1. [2]
    Deimos
    (edited )
    Link
    This is an extremely sterile analysis that doesn't even seem to consider anything about the students or their situations. Many of the people that are attracted to Lambda School can't just afford...

    This is an extremely sterile analysis that doesn't even seem to consider anything about the students or their situations.

    Many of the people that are attracted to Lambda School can't just afford to be down ~$22,500 if their "gamble" doesn't pay off, and they certainly aren't thinking about it with a gambling mindset when deciding to enroll.

    Treating it as an abstract math problem like this only makes sense if all you care about is aggregate numbers, and not the students as individual human beings with lives. And comparing an investment of thousands of hours to how often you can find something on Wikipedia is just absurd. You can't just shrug and try again repeatedly if your education fails.

    This post mostly just makes me upset about how unfeeling and oblivious the author is. Education shouldn't be analyzed this way.

    5 votes
    1. skybrian
      Link Parent
      The basic concept makes sense to me but I see now that it's not explained that well. Too much math, not enough explaining the assumptions. For example, that $22,500 isn't a real cost. Opportunity...

      The basic concept makes sense to me but I see now that it's not explained that well. Too much math, not enough explaining the assumptions.

      For example, that $22,500 isn't a real cost. Opportunity costs can be strange. Like, what's the opportunity cost of free college, if you flunk out? Or taking a gap year to travel? That depends what job you think you would have gotten otherwise. If work is hard to find, maybe not much. If you would have gotten a good job, wasting your time screwing around in school (or doing anything else than going to work) can be very expensive.

      We still do it though, because there is more to life than working all the time.

      It may be that the experience of learning something isn't a total loss, even if you don't get a better job out of it? At least, that's the line that many colleges will try to sell you on. Maybe we should offset that opportunity cost with the benefit of the experience?

      It seems to me that it's a scandal how many students (and parents) don't do the math, but Lambda school isn't a particularly bad example of this. Unlike college, the cost of failure isn't a hard cost, it's wasting your time. I think it's one thing to weigh opportunity costs and the benefits of learning against each other, and much worse to take a big gamble that puts you in debt afterwards.

      And that's a weakness of the blog post - it treats soft and hard costs the same.

      Also, the success rate of a school depends on how selective it is. The way to make sure everyone succeeds is to avoid admitting anyone who might not make it. This is denying them the opportunity to try. It's like denying a loan to someone who wants to start a small business.

      Maybe saying "no" to unprepared students might be for the best? It would increase profits for Lambda School and increase their success rate if they were more closed-minded about who they'll accept. It's the opposite of what non-selective schools do, though.

      3 votes
  2. skybrian
    Link
    From the blog post:

    From the blog post:

    On the topic of educational institutions, consider my favorites: Wikipedia and Duckduckgo. These institutions have very high failure rates for me; quite often, I search for something and do not find what I'm looking for. Obviously, Wikipedia is not in any sense bad for consumers simply because I search for lots of things I don't find.

    And the reason for this is obvious; Wikipedia is free. If it's useful 20% of the time and takes me 5 seconds to search for something, that's an average of 25 seconds of search per useful result. That's a win.

    In contrast, a college with a 20-30% success rate (and they do exist) is taking tens of thousands of dollars from students and providing very little value in return. This is much worse even though the success rate is the same.

    Lambda school operates somewhere in between the extremes of Wikipedia and College. It's a lot cheaper than college - particular for the people who don't succeed. This means that the socially optimal result (from the perspective of a student) is that Lambda School should admit more students until its success rate drops significantly below college.

    I understand the sentiment that high failure rates are a problem, but hopefully the example of Wikipedia provides intuition about why a low cost method of education can be valuable even if its success rate is low.