In the comment thread on the post on public vs. private delivery of pre-k, there is some discussion of the broader issue of using NAEP to measure the effectiveness of universal pre-k. NAEP is an acronym for the National Assessment of Educational Progress. NAEP has provided assessments of student achievement in different subject areas for samples of U.S. students since 1969. Since 1990, some results have also been reported for some states, with all states required to participate since the passage of No Child Left Behind.

The comment thread considers whether trends in NAEP test scores by state can be used to evaluate the effectiveness of universal pre-k. More specifically, the comment thread discusses whether we can use trends in NAEP test scores to evaluate the universal pre-k programs in Oklahoma, Georgia, and Florida.

To summarize my comments in this thread: NAEP trends as of now only add limited information to our evaluations of universal pre-k. The information from NAEP is limited because of the timing of various tests, effects of preschool on student retention in grade, and other causes of NAEP trends such as composition of the student population, changes in social problems, and changes in K-12 school quality. Better information is provided by well-done comparisons across individuals rather than across states, in which we compare individuals who participated in preschool versus individuals in a well-chosen comparison group. Having said that, the information that is provided by NAEP trends is consistent with universal pre-k in Oklahoma and Georgia having positive effects on student achievement.

For more specifics, see my replies in the comment thread.