The Two Cultures of Educational Reform – NYTimes.com

August 30, 2013 at 10:44 am 3 comments

(Shoot — I meant to put this on “draft” and come back to it, but hit the wrong button. Sigh.)

Here’s what I thought was interesting about this piece: I agree with Fish’s depiction of “data and experiment culture” about education, and the “ineffable culture,” too.  But his alignment of MOOCs with “data and experiment culture” of MOOCs seems wrong.  Our data about MOOCs says that they’re not working. So, belief in MOOCs is “ineffable.”  It’s about having warm feelings for technology and the hopes for its role in education.

About halfway through his magisterial study “Higher Education in America,” Derek Bok, twice president of Harvard, identifies what he calls the “two different cultures” of educational reform. The first “is an evidence-based approach to education … rooted in the belief that one can best advance teaching and learning by measuring student progress and testing experimental efforts to increase it.” The second “rests on a conviction that effective teaching is an art which one can improve over time through personal experience and intuition without any need for data-driven reforms imposed from above.”

Bok is obviously a member of the data and experiment culture, which makes him cautiously sympathetic to developments in online teaching, including the recent explosion of MOOCs (massive open online courses). But at the same time, he is acutely aware of the limits of what can be tested, measured and assessed, and at crucial moments in his analysis that awareness pushes him in the direction of the other, “ineffable” culture.

via The Two Cultures of Educational Reform – NYTimes.com.

Entry filed under: Uncategorized. Tags: , .

Defining an alternative to MOOCs with a better chance of engaging women Call for Papers: Special Issue of CS Ed for Doctoral Research

3 Comments Add your own

  • 1. BenK  |  August 30, 2013 at 11:51 am

    To reply to something which is still a ‘draft’ may not be quite fair; but perhaps it offers a chance for revision earlier in the process.

    To say that the data shows that MOOCs are ‘not working’ is a very serious and ultimately ‘ineffable’ claim. I can state just as definitively that they have already succeeded. Someone who attends to ‘data and experiment’ cannot baldly state ‘the hypothesis is false’ without stating the hypothesis! Further, in this case, as is so often the case in research, a hypothesis-testing orientation is a fundamental misunderstanding of the role of the data in the research and development process.

    MOOCs are not a static entity. Instead, they are both a product and an engineering platform, an experimental system. To argue that the data currently available from MOOCs demonstrates that they are a failure and should be abandoned is a perversion of the rhetoric of ‘data and experiment.’ It is a result of predetermining the limits of the engineering effort and predicting, ineffably, that there is no possibility of success. The data being produced from MOOCs is a ‘baseline’ for steering further design changes, refinements and so on. This is a valuable – and in fact, very large – aspect of a culture of ‘data and experiment.’

    Reply
    • 2. Mark Guzdial  |  August 30, 2013 at 12:15 pm

      That’s fair, Ben. Let me try to be more clear.

      Hypothesis: MOOCs are as effective as face-to-face courses, where “effective” is measured as net learning gain from Day 1 of a course to the last day of the course.

      The data that we have so far does not support that hypothesis. If 90% of a course drops out, then the net learning gain for those students is small (depending on how much of the course that they sat through). Given that so many people taking MOOCs are “refreshing” their existing knowledge, the net learning gain for them is also small.

      Of course, MOOCs are still in an experimental stage. I’m glad that people are still pushing on them and trying to make them work. But those doing so are driven not by data and results, but by an “ineffable” faith in technology and its potential role in learning.

      Note that I did not state my hypothesis as “can be as effective.” That’s a hypothesis that can be easily supported with an existence proof — you just find one MOOC that works as well or better than a face-to-face, and the hypothesis is proven. I’m interested in systemic change. Here at Georgia Tech, we’re building a whole degree with MOOCs. To do that, you need to support the strong form of the hypothesis that I describe above.

      Reply
  • 3. Rob St. Amant  |  August 30, 2013 at 12:41 pm

    On data and MOOCs, I was disappointed to read Sebastian Thrun’s recent post about the SJSU pilot.

    This summer, we ran the second instance of our pilot. While in the Spring, we actively sought out underserved high schools from low-income areas in California, this time we simply opened up enrollment to anyone… In short, pass rates are up and match more closely those in SJSU on-campus classes…

    He outlines some of the changes between the spring and summer classes, and says that these undoubtedly improved the success rate, but he also says that almost half of the summer students had a bachelor’s degree or better, in contrast to the spring’s mix of high school students and SJSU students. I think this makes the comparison of aggregate pass rates pretty meaningless–even if almost 50% more students passed college algebra, we don’t know if those students are are just brushing up their past experience. It would be nice to know what works and what doesn’t.

    Reply

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Trackback this post  |  Subscribe to the comments via RSS Feed


Recent Posts

August 2013
M T W T F S S
« Jul   Sep »
 1234
567891011
12131415161718
19202122232425
262728293031  

Feeds

Blog Stats

  • 1,267,825 hits

Enter your email address to follow this blog and receive notifications of new posts by email.

Join 4,562 other followers

CS Teaching Tips


%d bloggers like this: