Saturday, April 6, 2013

The Grey Zone Part 1: Unintended Negative Incentives

It's time to look at some issues in the grey zone.  It's good for us to push ourselves a bit.  Why?  Any strong system must have the strength and courage to look at itself as it is.  We can only grow out of our weaknesses, if we study them and learn from them. Just as we tell our students that we learn most from mistakes, we must also tell and believe in this ourselves.  Let's get on with the learning!

I also point out that the goal isn't to make anyone feel bad or guilty, but to open up a deeper discussion about what we're really in this for.  Further I emphasize strongly that no sane person intentionally wants students to fail or for our system to have the failings that it currently does.

Here's the starter topic for this series:  Atlanta schools are caught in a large, shameful cheating scheme to boost test scores.   

Atlanta is but one example, by the way.  Other districts have been caught tampering with test scores.  Why did Atlanta schools cheat?  I can't know exactly all the reasons why, but certainly it does not help when people's jobs were essentially tied to test scores.  When $$$$ and test scores are tied together tightly, then there exists incentives that encourage people to make unintended choices.  We see this on wall street and the banking/financial crisis.  We see this is sports, where the incentives to be juiced are apparently worth the risk to some athletes for the financial and social gain.  School life is a subset of our larger culture and not completely immune to some of our failings.

One notion that we don't talk about in the U.S. enough is balancing accountability (testing) with responsibility.  What I mean by this is that we are overly concerned with accountability, whether it is test scores, covering all the material, or getting good teaching evaluations.  If incentives are too "high stakes" or put another way if there's too much emphasis on accountability, then we are susceptible to opening a Pandora's Box of unintended consequences.

The Atlanta cheating scandal is a big, headline worthy example of unintended consequences of high-stakes testing, but there are other more insidious and frankly unwelcome versions of this at the college level.  One example is the coverage issue in freshman Calculus.  I'm not saying that cheating and coverage are the same issue, but they point to an underlying issue.   Instructors probably gripe about one thing the most:  coverage.  Calculus courses are jam packed with content.  Implementing active, empirically-validated teaching methods is hard work, and one of the main reasons why instructors do not implement modern teaching methods is time.  "I would like to do that, but there's no time..." The analogous issue in K-12 is "I would like to do that, but it's not on the test..."  In K-12 external accountability pressures that are too great push teachers away from modern pedagogies.

It's clear I believe in active, student-centered instruction.  Let's put that aside and think about the larger issue logically.  Whatever your take is on teaching, coverage should not be a core reason why we choose to use a teaching method or not.  We should use the methods that produce the best learning outcomes, based on the evidence available.  Moreover  focusing on covering a list of topics is only the start of a discussion about real education.  It is noted that missing from a list of topics are problem-solving ability, critical reasoning, communication, curiosity, attitudes about mathematics, and so on.

The part that is especially uncomfortable for us in the teaching profession boils down to this.  When we say, "I would like to do that, but I don't have time..." one could argue that this is like saying "I know this would help my students, but it's my job.  It's how the system is set up..."  I don't like the sound of that.  Being pinned down by coverage is parallel to the excessive accountability in K-12 schools in the sense there exists unintended incentives for not doing the right thing.

Let's turn this around to the positive direction...  we can start talking about coverage as a real issue and deal with it through having productive discussions and seeing what we can learn from one another, especially by learning from successful programs and instructors.  Ideas are out there for improving learning outcomes without sacrificing our standards that simultaneously improve areas such as problem solving and attitudes about mathematics.

The cheating scandal and being trapped by overly long syllabi are two examples of a by product of unintended negative incentives.  Cheating comes from excessive accountability, and slow uptake of empirically validated teaching methods is affected by excessive content demands.