Society for Research on Educational Effectiveness (SREE), Merritt, Eileen G., Rimm-Kaufman, Sara E., Berry, Robert Q., Walkowiak, Temple A., and Larsen, Ross A. A.
Classroom observational measures can provide information about high quality student-teacher interactions, allowing researchers to consider the impacts of these practices on student outcomes. Such measures can take a "process-oriented approach" that considers the nature of interactions between teachers and students, such as the sensitivity of teachers' interactions with students, teachers' effective management of the classroom, and the depth of instruction and quality of feedback given to students--all processes that have been linked to achievement gains (Pianta, Belsky, Houts & Morrison, 2007; Pianta & Hamre, 2009; Ponitz, Rimm-Kaufman, Brock & Nathanson, 2009). Alternatively, observational measures can take a "domain-specific approach" that considers teachers' practices that support information processing in specific subject-areas, such as mathematical problem solving. Seidel and Shavelson conducted a recent meta-analysis that examined the effects of teaching on student learning, considering both domain-specific and process-oriented approaches to teaching in K-12 classrooms. They found that domain-specific processes had larger effects (d = 0.41) on cognitive outcomes in elementary classrooms than all other factors (Seidel & Shavelson, 2007). The present study combines process-oriented and domain specific approaches, using a newly developed measure: the Mathematics Scan (M-Scan) Measure of Mathematics Instructional Quality (Berry, Rimm-Kaufman, Ottmar, Walkowiak & Merritt, 2011). Three primary research questions guided the authors' analyses: (1) What is the contribution of mathematics instructional quality to achievement for low-income students? The authors hypothesized that mathematics instructional quality is a strong predictor of mathematics instructional quality for low-income students, even after controlling for prior achievement, class-size, peer ability level, teachers' experience, and teachers' content knowledge, (2) What is the relative contribution of classroom ability level beyond classroom quality and class size in predicting achievement for low-income third grade students? The authors hypothesized that classroom ability level would contribute to achievement above and beyond mathematical instructional quality and other classroom factors, and (3) Does class size make a difference above and beyond instructional quality and classroom ability level in predicting achievement for low-income third grade students? Their hypothesis was that class size was significantly related to achievement above and beyond mathematics instructional quality and classroom ability level. This research reports the following findings: (1) The authors found that mathematical instructional quality was significantly related to third grade achievement for students from low-income families. For every one point higher a teacher scored on the M-Scan, students scored approximately 12 points higher on a third grade achievement test. Mathematical instructional quality explains 8% of the classroom level variance and 0.4% of the total variance in student achievement after controlling for student prior achievement, teachers' content knowledge, class size, classroom ability level, and teachers' experience level; (2) Results showed the classroom ability level was not a significant predictor of student achievement with all of the other variables in the model. This suggests that when students are offered the same level of instructional quality, teacher content knowledge, teacher experience, and have similar class sizes, the ability level of peers is not significant; and (3) Class size was a significant predictor of achievement for students from low-income families. For every 3 fewer students in a classroom, students scored 11 points higher on the third grade achievement test. Class size explained 28% of the classroom level variance and 3% of the total variance in achievement above and beyond the other variables in the model. Findings suggest the importance of mathematics instructional quality and smaller class size among students from families with low income. Interventions that support teacher improvement in standards-based mathematics may hold promise in efforts to reduce the achievement gap. Appended are: (1) References, and (2) Tables and figures.