Multidimensional Validity Revisited: A Multidimensional Approach to Achievement Validation

Richard E. Snow, working from his new aptitude theory, advocated a multidimensional approach to validating the construct of academic achievement. This report briefly describes Snow’s recommended approach and then summarizes new evidence from the present studies in terms of three related themes: (a) multidimensionality of science achievement, (b) transaction between achievement and situations, and (c) multidimensional approach to construct validity. Overall, our studies established the predictive validity of several key motivational constructs for science achievement, demonstrated how the relations between these constructs and achievement varied as a function of reasoning dimensions and assessment, and suggested how alternative assessment methods (constructed response and performance assessments) shed light on the notion of multidimensional validity. On the Multidimensional Structure of Science Achievement Snow and colleagues (Hamilton, Nussbaum, & Snow, 1997; Nussbaum, Hamilton, & Snow, 1997) found evidence of a multidimensional structure for science achievement in the National Education Longitudinal Study of 1988 (NELS:88). They established three underlying dimensions, which they called basic knowledge and reasoning, quantitative science reasoning, and spatial-mechanical reasoning. Just as importantly, student demographic characteristics, prior science experience (course taking, extracurricular activities), and motivation correlated with these three dimensions in different ways. For example, they found a strong gender effect on the spatial-mechanical dimension subscore, but not on the basic knowledge and reasoning or quantitative science reasoning subscores. Curious, Snow set out to test these findings in a new study, writing that “the primary objective of this study is to determine if knowledge and ability distinctions previously found important in high school math and science achievement tests occur also in other multiple-choice and constructed response assessments. . . . A second objective is to examine the cognitive and affective correlates of these distinctions. And a third objective is to examine alternative assessment designs that would sharpen and elaborate such knowledge