23 March 2011

The Pretense of Knowledge and Educational Testing

The protests in Wisconsin, Ohio, and other states have led to a large amount of general asininity when it comes to discussing the value of teachers and public education.  There are plenty of people who think teachers, as a whole, are overpaid relative to productivity (or, conversely, under-productive relative to pay).  There are also those who think teachers aren’t being paid enough given the demands of the job.  As someone who was home schooled and enrolled in public school, and as someone whose parents are both teachers, this debate strikes fairly close to home.

One of the biggest issues I have with this debate is the citing of standardized testing performance as a proper measure of teacher value, either in the aggregate or in the specific.  The reason for this is simply because tests do not, indeed cannot, accurately assess a teacher’s performance with any degree of accuracy.  Yet, these tests are cited as gospel, which means that these tests serve primarily as a pretense of knowledge.
There are, as I see it, two major problems with standardized testing.  One problem is that the test is completely arbitrary; another is that analysis of the test cannot even begin to account for all the variables that affect performance.

The latter problem is probably the larger problem, for test performance could be affected by what a child has for breakfast, what they’re wearing when they take the test, the ambient temperature of the room in which they take the test, the amount of sleep they had the night before the test, the amount of home preparation and study, etc.  To my knowledge, these variables do play a role in test performance, though the exact extent of which is unknown.  And there are likely other variables we have not yet even considered.  Standardized testing also tends to overlook the plasticity of the human brain; in doing so it ignores that what a person “knows” (or, more accurately, can recall) varies on a daily basis.  These sorts of things are beyond teacher’s control, and play a role in test performance.

The other problem, as mentioned, is that these tests are completely arbitrary.  I do not mean this in the sense of what sort of material is tested, even though it applies.  I mean that the test structure itself is completely arbitrary.  It is administered on a certain day at a certain time under specific (and, most likely, non-replicable) conditions.  Thus, the only thing that test results prove is that those taking the test performed at a certain (arbitrary) level on a specific (arbitrary) test under a very specific set of conditions.  To extrapolate results from this is akin to writing a biography of a random person based on seeing one picture of that person.  As can be imagined, relying on analysis of test performance is incredibly foolish and dangerous, particularly if test performance is going to be the main metric for teacher evaluation.

Now, this is not to say that students are actually well-educated and highly intelligent, nor is this to say that teachers in general are underpaid and overworked.  The only point in bringing up this critique is to point out how number fetishism is a poor method of analyzing teacher (and student) performance.  In fact, I would argue that that the reliance of statistical data is evidence of the increasing science fetishism in American society, where shoddy empiricism is used as a substitute for direct observation.  This mindset reveals a greater bias for relying on that which is perceived to be objective over that which is perceived to be subjective, but that is a subject for another post.

The flaw in this sort of thinking is that it ignores the fundamental fact that educational standards are inherently subjective.  Why should knowledge of mathematics be preferred to knowledge of history?  Why should English be given priority over civics?  There are legitimate arguments to be made for preferring one subject to another, but all arguments are inherently subjective, for value is inherently subjective, and educational standards are simply a subjective value.

This naturally begs the question:  What is the best way to determine teacher and student performance?
I believe that the best way to measure the performance of both teachers and students is to rely on the more subjective method known as “management.”  In this case, the students would be evaluated by the teacher, and the teachers would be evaluated by the principal.  There would few, if any, objective metrics with which to judge teacher performance.  And, given the inherent arbitrariness of educational standards, there wouldn’t be very many truly objective was to measure student performance.

This seems unsatisfactory for a wide variety of obvious reasons, but it’s not as bad as it seems.  In the first place, this sort of arrangement was the norm for the longest time in America.  It was not until education was federalized and increasingly centralized that number fetishism began to take root.  It should be noted that the further removed the planners are from the direct results, the more they must rely on hard data.

Also, education remains a very human activity.  The subjective nature of human experience is firmly rooted in education, and will not be removed.  Schools should not be thought of as assembly lines, but rather a place in which children can grow intellectually, in pursuit of that which interests them most.  Naturally, this requires that the school system become more decentralized and more tailored to a wider variety of scholastic subsets.  Since this requires that the government cede more of its power to citizens, you can bet this will never actually happen.

The upside to a subjective approach to education is that principals would be given far more control over school management, and would have the authority to fire bad teachers and kick bad students out of school.  Of course, the principal would be more directly responsible to parents, which would place more appropriate limits on his power.  Still, he would be authorized to act more quickly and decisively in response to problems, instead of dealing with the bureaucratic paper shuffle that is the hallmark of their job.

As should be obvious, the root of the current problem is the government and its tendency to centralize power. The solution lies in admitting that educational goals are inherently subjective and best attained by allowing the power of control to be left in the hands of those most closely connected to the students.  It worked before; there’s no reason it won’t work again.

No comments:

Post a Comment