Contextualized Science Assessments: Addressing the Use of Information and Generalization of Inferences of Students' Performance

• In Event: 77.027 - Authentic, Embedded, and Large-Scale Science Assessment

Abstract

Information provided by assessments (summative, interim, or formative) has had a limited use by teachers. Research (Goertz et al, 2009; Means et al, 2011; Shepard et al., 2011) has shown that teachers can hardly interpret and use the information provided in reports, they hardly use the information to learn about students’ conceptual understanding, and they hardly look for the sources of students’ problems. The paper describes an empirically tested approach to develop assessments that addresses two critical validity claims: use of data by the intended users (Moss, 2013) and quality of inferences supported by small samples of items (Koretz, 2005). The paper contributes to better understanding assessment development that is more aligned with the use of information for formative purposes.

Authors

• Maria Araceli Ruiz-Primo, Stanford University

• Min Li, University of Washington

• Jim Minstrell, FACET Innovations

• Xiaoming Zhai, University of Georgia

• Klint Kanopka, Stanford University

• Philip Hernandez, Stanford University

• Dongsheng Dong, University of Washington - Seattle