Curriculum Bias

Abstract

The present study was designed to investigate the potential bias of' a commonnli developed certifi'ing examination administered to students enrolled at the parent and regional schools of a basic medial science program. Data were collected from 260 first-year basic medical science students who took a basic science comprehensix'e in June of 1977; 180 students were enrolled at the parent campus and 80 were enrolled at the regional campus. Items were cross-classified by the discipline to which they were relevant and hi the (ampus affiliation of the author. Two total scores were generated for each student: one represented the student's per brmance on items authored b h faculty at the parent campus, while the second represented performance on items authored byf(culty' at the regional campus. Five discipline scores (two suibscores per discipline) were similarli generated. From the discriminant analyses it was found that students at the parent campus performed significantly better on items written by faculty at their campus, while students at the regional campus performed significantly better on items written by faculty at the regionally located campus. It was concluded that curriculum bias did exist and that the test construction procedures and exchange and review of items by both faculties were inadequate to prevent the occurrence of a curriculum dependence effect.