National report: don't stretch the test

Suzannah Herrmann, Ph.D.

Two very important questions face educators in determining when an Ohio high-school senior should become an Ohio high-school graduate. First, how much does that student know? And, second, exactly how should that question be answered?

With educators in many states, including Ohio, looking at using the ACT and SAT as high-school graduation tests, a new report says they should proceed with caution.

The Report of the Commission on the Use of Standardized Tests in Undergraduate Admission, issued in September, says it would be wrong to stretch these tests, twisting them to measure knowledge or ability for which they are not designed. The report comes as Ohio educators are increasingly chit-chatting about using the ACT to replace the Ohio Graduation Test (OGT). Certainly, something new is needed, if for no other reason than the OGT measures only what a high-school student may know up to the 10th grade. Gov. Strickland is interested. In his series of meetings with education stakeholders, the governor has hinted at using the ACT-in combination with other measures-to replace the OGT (see here).

In 2004, the Stark Education Partnership conducted a study (see here) that recommended that the state should grant waivers to districts to use the ACT and/or ACT WorkKeys in place of the Ohio Graduation Test in conjunction with end-of-course tests or in conjunction with the social studies component of the OGT. In 2007, Kentucky began using the ACT and its related exams to help inform decisions students make about high- school coursework. High-school juniors are required to take the ACT. Eighth graders take Explore, a curriculum-based assessment designed to help students make the most of high-school opportunities. Tenth-grade students take PLAN, an assessment which, in part, explores career/training options (see here).

Any discussion on how the ACT is used in Ohio, however, needs to account for the new report from the National Association for College Admission Counseling (see here). Chaired by the dean of admission and financial aid at Harvard University and joined by colleagues from places such as Georgetown, Columbia, and Achieve Inc., this report examines how college readiness tests such as the SAT and ACT should be used. Several of the recommendations explicitly address state issues. Particularly pertinent recommendations are to:

  • Refrain from using admission tests as measures of achievement for accountability purposes. Don't stretch the tests for uses other than which they were designed. They are designed to measure students' abilities to complete college work, and nothing more (or less for that matter), unless big technical changes are made to the test.
  • Ease the K-16 transition. Make the transition between primary and secondary school and college easier by having a college-readiness assessment that is aligned to high-school coursework.
  • Help verify the research on the value of test preparation. It's not clear how useful test-preparation programs are in helping students. We need to know more.

Of course, the report is not the end-all, be-all for how standardized admission tests should be used for accountability purposes by states. But it should give advocates of using the ACT or the SAT as a high school graduation test pause. It would be peculiar not to notice what is written in this national report, especially if choices are made to use these tests for multiple purposes. Regardless, what it simply goes back to is what constructs do we want to measure at the end of high school, what rigorous tests do we want to select (or develop), and what can we buy for these constructs and tests. The ACT may or may not rigorously measure what we want. That's to be determined. But it shouldn't be selected just because it's there and we want to keep up with the Joneses in Kentucky.