Lord (1980) proposed formulas that provide direct relationships between IRT discrimination and difficulty parameters and conventional item statistics. The purpose of the present study was to determine the robustness of the formulas beyond the initial and restrictive conditions identified by Lord. Simulation and real achievement data were employed.…

The recent increase in the use of constructed-response items in educational assessment and the dissatisfaction with the nature of the decision that the judges must make using traditional standard-setting methods created a need to develop new and effective standard-setting procedures for tests that include both multiple-choice and…

Ability estimates yielded by the one- (1PL), two- (2PL), and three-parameter (3PL) models and the nominal response model (NRM) were compared with the number-right (NR) scoring model using items not susceptible to test-wiseness (NTW) and items susceptible to the ID1 test-wiseness strategy. These items were contained in grade 12 diploma examinations…

The estimation accuracy of procedures based on classical test score theory and item response theory (generalized partial credit model) were compared for examinations consisting of multiple-choice and extended-response items. Analysis of British Columbia Scholarship Examination results found an error rate of about 10 percent for both methods, with…

Thirty-six test-wise and 41 test-naive high school seniors completed a 14-item test designed to assess their abilities to apply selected test-wiseness principles. Results indicate that, before students can profitably apply test-wise skills, they must first possess knowledge about the content of the stem or options. (Author/KS)

Research findings suggest that teachers are not competent in student assessment and that teacher education programs do not provide adequate training in assessment techniques. A survey of 33 Canadian universities found that most prospective teachers did not complete a measurement and evaluation course. Contains 44 references. (SV)

Evaluates the equivalence of translated achievement tests administered to 4,400 English- and French-speaking sixth-graders. Items displaying differential item functioning were flagged using three statistical methods; results were relatively consistent across methods, but not identical. Substantive review of French items via back-translation to…

An "ill-defined" informal performance task was constructed to evaluate secondary students' problem solving in mathematics. The task for 170 students in grades 9 and 10 was to evaluate other students' solutions to two mathematics questions. Higher-achieving students generally preferred responses reflecting multiple approaches to problem solving,…

Described in this paper are the first three activities of a research program designed to assess the differential validity and utility of successive and simultaneous approaches to the development of equivalent achievement tests in the French and English languages. Two teams of multilingual/multicultural French-English teachers used the simultaneous…