Value-added Teacher Estimates as Part of Teacher Evaluations: Exploring the Effects of Data and Model Specifications on the Stability of Teacher Value-added Scores


  • Nicole B Kersting University of Arizona
  • Mei-Kuang Chen University of Arizona
  • James W. Stigler University of California, Los Angeles



Value-added Analysis, Value-added Models, Value-added Estimates, Stability, Teacher Evaluations, Accountability.


In this study we explored the effects of statistical controls, single versus multiple cohort models, and student sample size on the stability of teacher value-added estimates (VAEs). We estimated VAEs for all 5th grade mathematics teachers in a large urban district by fitting two level mixed models using four cohorts of student data. We found that student sample size had the largest effect on changes in teachers’ relative standing and designation into performance groups, while control variables affected VAEs only minimally. However, we also found that teacher VAEs showed a fair degree of stability; year-to-year correlations ranged between .62 and .66, and changes in teacher effectiveness systematically varied by teacher experience, with beginning teachers showing the largest improvements over the four years under study.  Our results suggest that some model specifications are likely to produce teacher value-added scores that can reflect meaningful differences in teachers while we also found that other models might produce VAEs that might be unreliable. 




Download data is not yet available.

Author Biographies

Nicole B Kersting, University of Arizona

Nicole B. Kersting is an Assistant Professor in the Department of Teaching, Learning and Socio-cultural Studies and a faculty member of the Interdisciplinary Graduate Program in Statistics at the University of Arizona. Her research is focused on the measurement of different aspects of teacher
quality: Teacher knowledge, instructional quality, and student learning, and value-added approaches to estimate teacher performance. She developed and validated a novel approach to measure teacher knowledge in mathematics that is based on teachers' analyses of classroom video clips.


Mei-Kuang Chen, University of Arizona

Mei-kuang Chen is a postdoctoral researcher in the College of Education, University of Arizona. She graduated from the Department of psychology, with a focus on Program Evaluation and Research Methodology. Her research interest is in applying critical thinking, research methods, and statistical tools in various areas to gain a better understanding of human phenomena.

James W. Stigler, University of California, Los Angeles

James W. Stigler is Professor of Psychology at UCLA. His research focuses on understanding processes of teaching and learning, especially of mathematics and science, from kindergarten through college. James was Director of the TIMSS video studies and CEO of Lessonlab.  Stigler is best known for his observational studies of mathematics and science teaching, and has pioneered the use of multimedia technology for the study of classroom instruction.




How to Cite

Kersting, N. B., Chen, M.-K., & Stigler, J. W. (2013). Value-added Teacher Estimates as Part of Teacher Evaluations: Exploring the Effects of Data and Model Specifications on the Stability of Teacher Value-added Scores. Education Policy Analysis Archives, 21, 7.