The ink isn’t even dry on Kentucky’s new results from the eighth and tenth grade ACT-based EXPLORE and PLAN tests, but the Kentucky Department of Education (KDE) already spun out a somewhat misleading news release about the scores.Since the composite scores didn’t change at all for Kentucky’s eighth grade EXPLORE, and went down a tenth of a point for the 11th grade PLAN, such spinning is probably inevitable. Certainly, the KDE news crowd lost no time creating a muddled image of the true national trends on this test in an attempt to hide Kentucky’s lackluster performance.

Here are some problems with the KDE’s EXPLORE and PLAN news release.

Problem 1) The KDE release includes two score tables, one for EXPLORE and the other for PLAN, for the fall 2006 and 2007 administrations of the test. These tables list Kentucky’s scores for both years next to two more columns that supposedly separately list national comparison scores for both 2006 and 2007. The impression given by this presentation is that the national scores remained perfectly flat during this two-year period. That impression is absolutely incorrect.

You have to look at a small asterisk note found only with the EXPLORE scores (the asterisk is totally absent from the PLAN table) to learn what is really going on. In fact, there actually are NO national scores for either 2006 or 2007.

Instead, there is just one set of national comparison scores for a sampled test that was conducted back in 2005. Thus, the KDE news release’s tables are seriously misleading, strongly implying there are separate national reference scores for both 2006 and 2007 when there is in fact nothing of the kind available. The image of flat performance across the nation is thus nothing but a mirage.

Never the less, we do know that Kentucky’s 2007 PLAN and EXPLORE scores are below the nation’s 2005 performance in every area, rather notably so for our high schools. That leads to the second news release problem.

Problem 2) The release claims that when the number of test takers is large, the scores tend to be lower than elsewhere.

That is generally true, but this does not fairly apply in this particular case. The national comparison scores are from a specially selected group of students that is supposed to fairly and randomly represent all students in school, not just the “upper crust” that usually take the ACT college entrance test.

As a result, these national scores for 2005, as best as ACT could conduct their sample, approximate the overall performance of all students, with one exception. The national test sample only included students who took the test in the standard time limit. The sample did not include special education students who get to take the tests with testing accommodations. A considerable number of Kentucky’s learning disabled students do get to take the tests with extra time allowed, however, and that may inflate their scores.

The full impact of those testing accommodations simply is unknown. In any event, for the present the excuse for lower scores due to more kids taking the test cannot be supported with data — there simply isn’t any to cover the special learning disabled issue, and the special sampling used by ACT for the national comparison in 2005 eliminates this concern for all the rest of our students.