Presents the preliminary results of a comparison of Computerized Adaptive Testing (CAT) and conventional testing of English competency for students with limited English proficiency. The CAT procedure, which uses computers and a measurement theory known as Item Response Theory (IRT), provides several benefits that traditional paper and pencil…

As part of an effort to develop strategies for measuring institutional effectiveness, a study was conducted at West Virginia University at Parkersburg (WVU-P), a two-year institution with an enrollment of nearly 4,000 students, to determine the most appropriate instrument for measuring gains in general education between incoming freshman (i.e.,…

Two approaches to testing student achievement are discussed: (1) traditional standardized tests; and (2) testing achievement on an instructional unit. Content validity is a major problem with standardized achievement tests. Even though test publishers typically consider a number of curriculum guides and textbooks in producing a test which is…

Compares grades awarded by 5 examination boards for 11 subjects at A levels. Uses data from the A Level Information System (ALIS) project for 1989. Finds discrepancies among boards not statistically significant. Fails to identify grading as consistently severe or lenient. (NL)