Toward Measuring Instructional Interactions “At-Scale”

Abstract

This study explores two approaches to directly measuring the quality of instruction: teachers' assignments with student work and focused lesson observations. The technical quality and potential feasibility of these approaches for measuring instruction in large numbers of classrooms are compared within two different content areas (reading comprehension and mathematics). Generalizability and decision studies determined the optimal number of observations and assignments needed to obtain a reliable measure of a teacher's practice, and the association of these direct measures of instructional quality with student achievement was estimated. For both content areas, four assignments assessed by two raters yielded a reliable estimate of quality and as few as two observations yielded a reliable estimate of quality when teachers complied with the requirements of the research. The quality of observed instruction and teachers' assignments differentially predicted gains in students' achievement on the Stanford Achievement Test within each content area. The implications for measuring instruction “at-scale” in different content areas are discussed.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 109 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.