Assessing Instructional Content and Interactions ’At-Scale’

LPC Faculty: Lindsay Clare Matsumura & Richard Correnti (Co-PI’s)
Associated Individuals & Organizations: Laura Hamilton (Co-PI), RAND
Funding Agency: Spencer Foundation and the W.T. Grant Foundation

In this research project we examine measures of the instructional core in order to understand: a) their relationship with student achievement and b) how learning opportunities are distributed to students. We have selected two instructional measures as the focus of our research - content measures (collected through daily logs) and ratings of assignment quality. In the past, both measures have been shown to be reliable and valid and both have the ability to be used at-scale. While either measure alone has potential for explaining variability in classroom achievement, the focus of our investigation is how these measures combined predict student learning. Toward that end we examine student learning as measured through state standardized tests (PSSA) and biannual ratings of curriculum embedded student essays. We predict: a) that measures of instruction will differentially predict learning outcomes on these assessments, and b) that combined measures of instruction will outperform other predictors. Annual surveys also provide a) a third measure of practice for comparison, b) information about feasibility in terms of cost and respondent burden, and c) the potential for instructional measures to be used as a professional learning tool. Finally, an additional focus of this research capitalizes on the longitudinal data being collected. We examine how stable teacher practice is within and across years, and also what factors predict instruction. We also investigate how students accumulate learning opportunities, if inequities exist between groups of students and whether variation in students’ opportunities predicts differences in student learning.