By Justin Hill, Johns Hopkins University
A recent meta-analysis by Vanderbilt’s Lam D. Pham and colleagues focused on the connections between teacher merit pay systems and student test scores, teacher motivation, and teacher turnover. The search criteria utilized by the authors yielded an overall sample of 37 studies, with 26 of those studies focusing on education in the United States. The researchers then used a random-effects model to analyze the general effect of teacher merit pay systems and investigate specific aspects of merit pay systems, such as a comparison of effects in different subject areas, student age levels, amounts of rewards offered, and the inclusion of professional development opportunities.
The primary finding of the meta-analysis indicated that teacher participation in a merit pay system was associated with a 0.043 standard deviation (SD) increase in student test scores (p < 0.001) when compared with teachers not participating in a merit pay system. Most of the studies included within the meta-analysis focused on test score improvements in either math or English and a comparison of those outcomes indicated a greater effect for math (+0.05 SD; p < 0.001) than for English (+0.03 SD; p < 0.001). In comparing outcomes based on student age, the researchers found a greater effect for elementary school students (+0.10 SD; p < 0.05) than for middle school (+0.01 SD; p < 0.1) or high school (+0.06 SD; p < 0.001) students. A comparison of programs based on the amount of money awarded to teachers indicated that awards above the median amount produced better results (+0.06 SD; p < 0.05) than systems offering awards below the median amount (+0.03 SD; p < 0.001). Finally, merit pay systems that also included professional development opportunities produced better student test improvement (+0.08 SD; p < 0.05) than those that did not include professional development (+0.04 SD; p < 0.001).
The researchers also found evidence that merit pay systems are associated with improved teacher motivation as well as decreased teacher turnover. Given that this meta-analysis suggested promise for merit pay systems, the authors concluded the article with suggestions for future research into merit pay. These suggestions include research into specific incentive amounts that would balance teacher motivation with cost efficiency, the effects of public recognition, and the combination of incentives for improvement alongside threats of dismissal for low performance.