Teacher Value-Added at the High-School Level: Different Models, Different Answers?

Issue/Topic: Assessment--Value Added; Teaching Quality--Evaluation and Effectiveness
Author(s): Goldhaber, Dan; Goldschmidt, Pete; Tseng, Fannie
Publication: Educational Evaluation and Policy Analysis
Published On: 1/16/2013

Teacher quality is one of the most important factors in student achievement. Using value-added models (VAMs), which use growth in students’ performance on standardized achievement tests to produce a measure of a teacher’s contribution to student learning, is increasingly a component of teacher evaluation. High school students typically do not take annual achievement tests and their courses are not always contiguous, making it difficult to produce value-added measures.

To test whether an alternative VAM would provide more accurate measures of high school teachers’ performance

  • The estimated effects of teacher quality on student performance vary according to the type of VAM being used.

  • The alternate VAM tested in this study has some advantages for evaluation of high school teachers because the model requires fewer data. The model also produced much smaller estimates of teacher effect on student achievement than did traditional models.

Policy Implications/Recommendations:
  • While VAMs are not perfect, they are better predictors of teachers' impact on student learning than traditionally used indicators, such as teacher credentials, coursework, and years experience.

  • Different VAMs result in different measures of teacher effectiveness. Thus, consideration of each VAMs characteristics is necessary in determining how value-added measures will be used to make decisions in areas such as compensation, tenure, and professional development.

  • The authors did not conclude which VAM provided more accurate results for high school teachers.

Research Design:
Stratified random-sampling design of data collected by ACT. Researchers conducted different VAM (across-time models and across-subject models) on one data set.

Data from students in grades 10 through 12 in public and private schools in the Midwest. The students were tested in one, two, or three subjects. The data set also included the size, type, and geographical location of each school, as well as student, teacher, and school characteristics (e.g., student gender, school average class size, and teacher years of experience).

Year data is from:
Not specified


Data Collection and Analysis:
ACT collected the data as part of their QualityCore end-of-course assessments pilot. The data set included student pretest and posttest results across many subjects, as well as information on student achievement in a variety of subjects, links between teachers and students, and information on student achievement at the beginning and end of each school year.


Reference in this Web site to any specific commercial products, processes or services, or the use of any trade, firm or corporation name is for the information and convenience of the public, and does not constitute endorsement or recommendation by the Education Commission of the States. Please contact Kathy Christie at 303-299-3613 or kchristie@ecs.org for further information regarding the information posting standards and user policies of the Education Commission of the States.