To evaluate student learning in a computer-supported environment known as "GenScope," a system was developed for assessing students' understanding and learning of introductory genetics material presented in two developed GenScope instruments. Both quantitative and qualitative methods were used to address traditional evidential validity concerns as well as more contemporary concerns with consequential and systemic validity. Findings from three GenScope implementation classrooms and interviews with two teachers and five secondary school students show strong evidential validity, but only limited consequential validity. In response to these findings, a set of curricular activities was developed to scaffold student assessment performance without compromising the evidential validity of the assessment system. The study shows the usefulness of newer interpretive models of validity inquiry and the value of multifaceted Rasch measurement tools for conducting such inquiry. Two appendixes contain sample items from one assessment and a sample GenScope investigation. (Contains 3 tables, 5 figures, and 23 references.) (SLD)