Test Item Bias (Quantitative Applications in the Social Sciences)

Steven J. Osterlind discusses five strategies for detecting bias: analysis of variance, transformed item difficulties, chi square, item characteristic curve, and distractor response.

Dr. Steven J. Osterlind is Professor of Measurement and Statistics and Director of Educational Psychology program, University of Missouri-Columbia. Dr. Osterlind’s expertise is in psychological assessment, including tests and measurement, statistics, psychometric methods and test development. He received his doctoral degree in 1976 in Educational Psychology (Measurement & Statistics) from the University of Southern California. In 1979, he was an American Scholar’s Fellow at Yale University. At the University of Missouri he teaches graduate-level courses in multivariate statistics, analysis of variance, regression, general linear modeling, and psychometric methods. Additionally, he teaches seminar courses on specialized topics, including Item Response Theory and Computer Applications of Testing. He has worked on numerous national testing programs, including serving as statistician for NAEP (National Assessment of Educational Progress). In licensing and certification, he has worked with dozens of professional associations and organizations on their assessment programs. Dr. Osterlind has authored five books, the most recent of which is a major textbook titled Modern Measurement: Theory, Principles, and Application of Mental Appraisal; and, he has written more than 60 articles, book chapters, and other research reports in assessment. Additionally, he has authored more than 20 tests. He is principle author of College Basic Academic Subjects Examination (C-BASE), a test of collegiate achievement test currently adopted by more than 100 universities across the nation.