The Magical Thinking of State-Level Education Goals
There's quite a controversy brewing over Florida's different academic proficiency targets for different ethnic groups. While most educators will be familiar with NCLB-style disaggregated student achievement goals (which are based on improvement over past scores, not lowered expectations for some groups), apparently such racially disaggregated goals at the state level are something new, part of Florida's NCLB waiver. Many are crying foul, labeling the goals "the soft bigotry of low expectations," to borrow a phrase from former President Bush.
I'll go a step farther: setting goals at the state level is silly, period. States have a very, very indirect impact on the performance of students, which explains why states tend to monkey with the tests, not make educational changes, in order to raise scores.
Districts—especially large ones—also have very little direct impact on the academic achievement of students, so it's little surprise that huge improvements (absent some obvious logical cause) are often a sign of cheating, as they were in El Paso, or of demographic changes, as Diane Ravitch argues in The Death and Life of the Great American School System.
Under what conditions does goal-setting work? In other words, when can goals actually lead to improvement? I can think of several necessary factors.
First, goal-setting must be tied to a theory of action so that the impact of specific actions can be assessed. Yet there is virtually no effort to conduct this kind of analysis or distinguish true impact from coincidence.
So superintendent contracts include clauses on raising test scores. But can they do so? The literature on the superintendency, with few exceptions, answers "yes" to the question. When writers, policy makers, and administrators mention successful school chiefs they point to increasing scores on standardized achievement tests, high percentages of graduates entering college, and National Merit Scholarship finalists.
Yet when superintendents are asked how they get scores or graduation rates to go up, the question is often answered with a wink or a shrug of the shoulders. Even among most researchers and administrators who write and grapple with this question of whether superintendents can improve test scores, there is no explicit model of effectiveness.
Perhaps rising test scores are nothing more than luck, Cuban continues:
Without some model by which a superintendent can be shown to have causal effects, test scores going up or down remain a mystery, a matter of luck that the results occurred during that school chief's tenure.
Second, for goal-setting to be effective, the area of improvement must be something that effort, practice, resource allocation, or other actions can actually impact.
I can set a goal to lose weight, but not to grow taller. Setting a goal to improve something you can't actually improve is called magical thinking.
Policy is a blunt instrument, and state-level policy is surely no exception. When state legislators or education officials make no substantive changes from year to year, why should they expect to see any change in scores for which they can take credit?
The classroom door is nearly impenetrable, as any experienced educational leader can tell you. If you can get teachers to open their doors, they can make change happen on an impressive scale, but most state actions have a vanishingly small impact on actual classroom practice, with the notable exception of high-stakes testing.
When it comes to disaggregated goals, why on earth would the State of Florida expect to see specific growth among specific groups absent any targeted action to address their needs? Are they putting serious money behind detailed plans to achieve these goals? Taking no action and expecting big change is called delusion when individuals do it, but apparently it's called goal-setting when states do it.
Third, if we're not on track to achieve our goals, we need to be able to make adjustments along the way. To the extent that the goals depend on the performance of other people, there must be latitude to make more dramatic adjustments to other factors to compensate for this limited degree of control. If a CEO has a division that is unprofitable, he can increase its R&D and marketing budget, fire the VP in charge, or make other similarly bold changes. If the quarterback keeps throwing the ball away, the coach can call a time-out and develop a new plan.
Usually there is no such latitude in education at the state level, which partially explains why we're seeing increasingly desperate moves such as the present efforts to tie educator pay to student performance. If state policymakers want to actually set and achieve goals, they need resources that they can flexibly deploy to address emerging needs. In a two-year budget cycle, this tends to be impossible.
Let's put this in perspective: as a principal, I can't make a kid do his homework or master a concept, despite the wide range of motivators and conditions I can influence. Do we really believe that Florida education officials can have a specific impact on how many Latino students graduate from high school?
Given these limitations on goal-setting at the state level, I'm not getting too worked up about the targets any particular state sets for its students. Goal-setting is mostly a waste of time.
Instead, states should decide what they want to happen, figure out the most efficient ways to make it happen, and relentlessly execute specific, focused plans to achieve viable targets. Let's say you want to increase the graduate rate among Latino students. Great. How about investing in 8th graders in the areas with the highest dropout rates, and seeing if you can move the needle for them over a 5-year period? You can then figure out what works and how much it will cost, and do more of it.
This is what great teachers do. They figure out how their students are currently doing, and design appropriate plans of adequate intensity to get them to where they need to be.
This is what states should be doing, not setting silly goals. If states want to make wild predictions about things outside of their control, it would be easier to stick with fantasy football.