Theories of Effective Teaching: What Do Adolescent Perceptions of a Classroom Experience Tell Us?

• In Event: Excellence in Education Research: Early Career Scholars and Their Work

Abstract

The majority of U.S. states have developed statewide teacher evaluation systems. Common reform features include the use of multiple measures of teacher performance (e.g. observational instruments, students’ standardized test scores, teacher portfolios) as well as changing teacher rating scales from dichotomous (e.g., satisfactory versus unsatisfactory) to a range of distinct ordinal performance levels. While variability exists in the details of what data is gathered and how those data are used to evaluate teaching, a noteworthy and recent addition to the assessment of effective K – 12 teaching is the inclusion of student perception surveys. Student perceptions of the learning environment provide unique and important information for the assessment of teaching quality. Prior empirical work suggests that students differentiate among three dimensions of teaching quality, generally described as classroom management, instructional challenge, and emotional support.
Using the Measures of Effective Teaching (MET) longitudinal data, we investigate the factorial structure and predictive validity of student ratings of teacher quality using multilevel factor analyses. Our investigation focuses on the use of the Tripod survey instrument in sixth, seventh, and eighth grade math class sections in the first year of the MET study (N = 52,419, students, 2,190 classrooms). The Tripod student perceptions survey has become, in part because of the publicity garnered by the MET project, among the most popular student perceptions survey on the U.S. market; during the 2012-2013 school year the Tripod student perception survey was used in approximately 1400 schools. Tripod survey developers have stated that the 36-item survey measures seven dimensions of effective teaching, or “the Seven Cs.” These dimensions include the following: (a) Care, (b) Confer, (c) Captivate, (d) Clarify, (e) Consolidate, (f) Challenge, and (g) Control.
We tested five alternative factor structures of the Tripod survey instrument that have been used in prior empirical studies that employed the Tripod survey instrument. Our preliminary results indicate that a simple one-dimensional factor structure fit the data well and outperformed alternative and more complex factor structures.

Author

• Tanner LeBaron Wallace, University of Pittsburgh