Abstract
The state of Georgia initiated the use of a criterion-referenced test to identify students in need of academic remediation. Remediation efforts were to be mandated to foster performance on the state's minimal competency test, which is required for graduation. This paper examines the relations among criterion-referenced test scores, subsequent predicted remediation need decisions, and actual performance, without remediation, on the minimal competency test. The results indicate that an excessively large number of students would be required to be remediated who subsequently (without remediation) successfully passed the minimal competency test. Special emphasis is placed on the procedures used to establish cutoff standards and the occurrence of classification errors. Recommendations concerning the use of criterion-referenced tests for classification decisions are presented.
Get full access to this article
View all access options for this article.