The Discussion Of Education In America Must Move To A Higher Level
Public education was created in part to be one of the mediating institutions that would mold the American character one citizen at a time. It is critical to the creation of responsible citizens capable of making informed decisions in order to produce and maintain a system of government that works. For at least a generation now, public education has abandoned the noble purpose of helping our young people understand who we are, where we came from, what we stand for and how to pass that on to our successors. Instead, it has embraced the goal of making sure that young men and women are competent at whatever they choose to do in life. Competence is important, but it does little to prepare the next generation for the job of deciding what this nation’s future will be.
If citizens are to remain citizens, and not merely consumers; if individual happiness is to be the product of more than the mere satisfaction of individual wants and desires; then the discussion of education in America must move to a higher level. It must touch upon the greater purposes that animate the nation. The advent of dot-com democracy brings with it a heightened sense of both the importance and the urgency of that discussion. We live in a time when it is possible to be all places all the time; to communicate immediately anywhere in the world; to make decisions on anything from holiday gifts to competing candidates with the click of a mouse; to create mass democracy unlike ever in the history of the world. Ironically, as we possess the technology to communicate with one another more efficiently than ever before, we run the risk of becoming a nation of strangers – each alone in front of a computer screen, talking in chat rooms, on e-mail, through the Web.
We possess the tools to transform the nature of democratic government, to make sure that democratic government responds to the wishes of the people, expressed directly by the people. The question then becomes: Do we possess the wisdom as a people to step back and ask if that is really such a good idea?
In an age of instant access, instant information and instant gratification, do we possess the wisdom to distinguish between the desire to satisfy the momentary impulse to serve popular opinion and the discipline, foresight and discernment needed to seek the long-term interests of a nation?
These are the most fundamental questions that have always confronted the American republic. For generations, educated citizens of that republic have found answers to these questions – at times through deliberation, at times through dumb luck. But the global context in which these questions are raised today is unlike ever in the world’s history, making our ability to come up with the right answers all the more important. And that means that the quality and character of the education provided the current and future generations of young minds in a democracy will be all the more critical to ensuring the future of that democracy.
While accountability for results has been an education reform slogan for some time, it is increasingly becoming a reality for schools around the nation. When states and districts create accountability systems, the first issue policymakers face is how to tell which schools and classrooms are succeeding, which are failing – and which are somewhere in between, perhaps succeeding at some things and lagging in others. This turns out to be genuinely complicated. Picking the schools with overall high or low average test scores is an obvious way to proceed, but the strong correlation between test scores and student socioeconomic background makes this problematic. Such an approach will tend to reward schools with prosperous students and punish those with disadvantaged pupils.
Most states are interested in rewarding the schools where teachers are most effective at producing student learning — that is, the schools that add the greatest value to their students, no matter where those students start or what advantages and disadvantages accompany them to school. In its simplest form, value-added assessment means judging schools and sometimes individual teachers based on the gains in student learning they produce rather than the absolute level of achievement their students reach. It turns out, however, that just as students start at different levels of achievement, they gain at different rates at well, sometimes for reasons unrelated to the quality of instruction they receive. For example, middle-class-children may be more likely to have parents help them with their homework. To identify how much value a school is adding to a student, the effect of the school on student achievement must be isolated from the effects of a host of other factors, such as poverty, race, and pupil mobility. A number of states and school districts are turning to sophisticated statistical models that seek to do just that. These “value-added” models come in two basic flavors: those that include variables representing student socioeconomic characteristics as well as a student’s test scores from previous years, and those that use only a student’s prior test scores as a way of controlling for confounding factors.
Whether to incorporate measures of student background into the model is a charged and complicated question. Those who use the first type of analytic model (including measures of student poverty, race, etc., in addition to prior test scores) do so because they find that socioeconomic characteristics affect not only where students begin but also how much progress they make from year to year. Given the same quality of instruction, low-income and minority students will make less progress over time, their research shows. If the background variables are not included, the model may underestimate how much value is being added to the students by these schools. Student background is not strongly correlated with the gains a student will make, once the student’s test scores in previous years are taken into account. If socioeconomic status indeed influences the gains made by students, as much research suggests, this raises thorny policy questions for value-added assessment. Omitting such variables from the model is apt to be unfair to schools (or teachers) with a high percentage of disadvantaged pupils.
Public education is undergoing a reformation. The future for education means transforming our static industrial age educational model into a system that can capture the diversity and opportunity of the Information Age. That means public education must reconnect with the public – the children it was intended to serve.
Effective education is not about programs and process; it’s about what’s best for your child. Some districts may deal with this dilemma by using both the level of achievement and the results of value-added analysis to identify effective schools. Another response is to assign rewards and sanctions based on value-added analysis as an interim measure until all students are in a position in which it is reasonable to expect them to meet high standards. No doubt other variations and hybrids wait to be developed and tried.
The debate over including student background characteristics in the model is important. More research is needed on how the various models perform. Today, for example, we don’t even know whether different analytic models will identify the same schools as succeeding and failing. Nevertheless, either approach gives us a more accurate measure of the contribution of a school to student learning than we would have if we looked simply at average test scores or at simpler measures of gain.
It is less clear that the models can confidently be used to identify effective and ineffective teachers. Researchers have found that teacher effectiveness (as measured by either type of model) can change a great deal from year to year. This means either that teachers often make major changes in their effectiveness or that the statistics for teacher effectiveness are not accurate. (It could be that the model does not adequately adjust for the presence of disruptive students in a class, for instance.)
Because value-added assessment for individual teachers is imperfect, many believe that it is best used as a diagnostic tool, to identify the teachers that need the most help, rather than as the “high-stakes” basis for rewards and punishments. Others contend that complicated analytical methods that leave so much to statisticians should be abandoned both for schools and for teachers in favor of simpler calculations that can be more readily understood by policymakers, educators, and citizens. Still others are content to let the marketplace decide which schools are effective. Whether these various audiences will prefer a form of analysis that is fairer or one that is more transparent remains to be seen. As the statistical techniques improve and we learn more about the accuracy of different models, though, value-added analysis is sure to become more appealing to states and districts. They can prepare to take advantage of these advances by beginning to gather the data required to make the models work, including regular test scores for all students in core subjects, and creating longitudinal databases that link student test scores over time.