Monday, August 2, 2010

The Frustrating Reality of Test Scores

Yesterday's New York Times had an article on plummeting test scores in New York City. The state has recalibrated and made tougher its expectations on state English and math exams. As you might expect, the results have been dramatic. Schools in New York City that were flying high saw their passing rates plunge. Not a pretty picture.

Teachers and principals interviewed for the article put a game face on the situation, and what else can they do? Any school looking for state or Federal funds has to play according to state or Federal rules, which means measuring results on test scores and pass rates. Of course, one of the many challenges of this dynamic comes in situations like these, when those measurements change and schools and districts scramble to face a new reality.

One solution, as discussed in a recent Time cover story, The Case Against Summer Vacation, is to do away with the notion that public schooling is a nine-month affair. In this scenario, June, July, and August are culprits in the unintended crime of helping students to forget, during the summer months, much of what they learned the previous year. Keep 'em in school year-round, this argument goes, since we don't need them out in the fields helping with the farm anymore (the original logic behind a September-June school year).

Perhaps the answer is indeed more summer school, or versions of summer school. But as another New York Times article notes, "The scoring adjustment could raise questions about the precision of educational testing." In a word, precisely. Let's use as a baseline the assumption that there are many components to improving success in public schools. Better materials that reflect specific student populations, smaller class sizes, improved teacher pay, more coordination and sharing among educators -- the list can grow, but it's hard to argue with the basics.

Living (and sometimes dying) by the score, however, can be a hazard. It's a natural human desire to look for some objective measurement against which we can compare ourselves to one another. That makes sense in sports, in science, and in other fields. But in education, the variables are so many, and they so deeply depend on the human response under varying conditions, that a strict numerical assessment can be limiting.

Especially when Those On High alter the rules by which the assessments are made. That's what happened in New York, and it regularly occurs across America. If there is a lesson or a moral to be derived from all of this, it's that there is no sure, single solution to educational progress. Students need many approaches, many avenues to success. What works for one often doesn't work for the next. And measuring success is rarely as easy as pointing to a test score.

No comments: