Despite some recent media reports about declining student test scores, some with problematic methodologies, the author of this piece suggests such reporting has obscured some real successes and that we lose sight of the gains taking place in education.
Average test scores, despite the play they get in the media, have very little meaning as measures of student achievement, because they can hide gains made by increasingly large minority populations. This is known as Simpson’s paradox, which is explained here. For example, the twelfth grade NAEP data for English language learners showed a slight increase in the last 15 years (though not by a statistically significant amount). ELL students, who tend to score lower, make up a much greater percentage of students today than in 1992, and thus have the effect of “bringing down” the average despite stable or improved performance.
As Michael Martin, a research analyst at the Arizona School Boards Association, has noted, today’s population of students are much harder to teach. Minority populations, English language learner populations, and populations of students from single-parent or foster homes have all increased. In addition, more students today are in college preparatory programs; many of the students who are today achieving lower scores would have, in past decades, been tracked into non-college preparatory classes such as vocational education.
Finally, Martin also notes that graduation rates for students of all races and backgrounds are significant higher today than in previous decades and are increasing. Lower twelfth grade test scores, he argues, are the result of our successes in keeping students in school who would otherwise have dropped out. We are looking at the wrong measures of success, he argues, and we are therefore drawing the wrong conclusions.
When we judge the achievements and failures of students based only on a few selected test scores, we lose sight of real measures of success in education and do ourselves and our children a disservice. If we want to grade our schools, we should do it based on figures that matter – graduation rates, preschool enrollment, or other real measurements of the quality of education schools are providing. In these areas, states around the country have seen marked improvement, and that is something worth talking about.