Working Paper: NBER ID: w11463
Authors: Brian A. Jacob; Lars Lefgren
Abstract: In this paper, we compare subjective principal assessments of teachers to the traditional determinants of teacher compensation ¡V education and experience ¡V and another potential compensation mechanism -- value-added measures of teacher effectiveness based on student achievement gains. We find that subjective principal assessments of teachers predict future student achievement significantly better than teacher experience, education or actual compensation, though not as well as value-added teacher quality measures. In particular, principals appear quite good at identifying those teachers who produce the largest and smallest standardized achievement gains in their schools, but have far less ability to distinguish between teachers in the middle of this distribution and systematically discriminate against male and untenured faculty. Moreover, we find that a principal¡'s overall rating of a teacher is a substantially better predictor of future parent requests for that teacher than either the teacher¡'s experience, education and current compensation or the teacher¡'s value-added achievement measure. These findings not only inform education policy, but also shed light on subjective performance assessment more generally.
Keywords: education; teacher evaluation; principal assessments; student achievement
JEL Codes: I20; I28; J30; J70
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Subjective principal assessments of teachers (A14) | Future student achievement (I24) |
| Principal assessments (C12) | Identification of top and bottom 20% of teachers based on ability to raise standardized test scores (I24) |
| Principal's overall rating of a teacher (A14) | Future parent requests for that teacher (J13) |
| Principal's overall rating of a teacher (A14) | Traditional measures of teacher quality (A21) |