Tuesday, January 31, 2012

Educational Assessment

College tuition is soaring five times faster faster than inflation, while 45 percent of students did not demonstrate any improvement in learning. For-profit schools are sucking up federal dollars, some of them getting over ninety percent of their revenue from federal money, while graduating only 22 percent of the students that start. The current solution to this problem that is being pushed to this problem: educational standards and assessment. It is, however, a dangerous path to follow.

Where does this come from? I'll blame Iowa's own Edwards Deming. Edwards Deming helped manufacturing industry by attempting to apply a bit of science to manufacturing. Measure the quality of what you create. Try something new. See if it helps.

The U.S. didn't like him, so he went to Japan. Guess what? Engineering your manufacturing process using evidenced-based changes did help! So now it seems like everything is made in Japan and other Asian countries. The key was to create a continuous cycle of evidence-based improvement.

Now, society wants to apply this to teaching as well.

Good teachers will naturally go through this process of assessment. They will continue to work on their class and find ways to improve it. They find out what other teachers are doing. Good teachers learn how to judge the personality of the students they work with, and learn techniques for working those students.

This process of continuing improvement is why good teachers often write their own books, and keep editing them to make them better. They create their own lesson plans and improve them year after year. But the process isn't perfect.

The problem in applying this to teaching? The impossibility of creating a common benchmark that can be used to evaluate the effectiveness of each teacher. Education is about humans. Since we are working with so many different humans, there are too many variables. Variables such as:
  • Different teachers have different styles. (The variety between teachers works well to keep students interested.)
  • Students aren't all the same, and they respond to different techniques.
  • Some students are capable of excelling in certain subject areas. Other students just move slower.
A continuous process of improvement assumes a controllable set of variables. Unfortunately this is not a situation that exists inside of academia. Particularly once the scope starts extending beyond the range of a single school, or even a single teacher.

At an individual or small-group level this continuous improvement may still be possible. Once the process expands to a larger group, it is too easy to get caught up in the paperwork and 'process.' People pay attention to the paperwork being required, and ignore the real goal of providing evidence-based improvements. A process that worked for one group may not work for more people.

This article does a wonderful job of explaining why this is bad:
Big Macs vs. The Naked Chef

(Seriously, go back and read that article. It is 10 times better than this one and you are wasting your time if you continue reading without having read that other article. Plus it has the word "naked" in the title, so it has to be good.)

Yes, that article was about software development. But the theory is the same in education.

What I hate about educational 'standards' is that it works to bring up poor performing people to the average, while bringing down high-performing teachers. We are left with an uninteresting homogeneous soup at the end.

Do we want our schools to be McDonalds? Do we want our students to be Big Macs?