Intellectual Merit: The Pennsylvania State University, University Park proposes to apply item response theory to analyze tests given in undergraduate chemistry, physics and biology classes with specific attention to gender as an explanatory factor. The researchers will evaluate the psychometric properties of the assessments using industry standard practices of differential item functioning analysis, longitudinal analysis, and multidimensional analysis. Gender differences in approaches to science learning, and ultimately in achievement, will be explored and interpreted in collaboration with content area specialists and instructors. The findings will be connected on the psychometric properties of the class tests with data on implicit attitudes derived from separate student assessments. The intellectual merit of the project is to provide an exploration of the ongoing dynamics of gender and assessment in undergraduate science. The PIs expect to analyze over 50,000 student test events, and to provide a rigorous examination of the character and basis of gender differences in achievement in early undergraduate classes.
Broader Impacts: The broader impact of the project is to foster a more extensive understanding of the subtle (and not so subtle) associations between gender, testing and achievement in undergraduate science. Instructors will be more aware of gender differences in achievement across specific item?]types, or content areas, and can consider whether adjustments to classroom assessments might be desirable. At the same time, the project will raise awareness more generally about the strengths and weaknesses of the status quo for testing and evaluation in large scale science classes. The PI team will train researchers and instructors to be aware of the psychometric properties of their multiple choice tests, and to appreciate the consequences of their use for assessment.
