Can social recognition for teachers and principals improve student performance? Evidence from India

ABSTRACT

Empirical evidence on the effectiveness of performance-based rewards for teachers is primarily based on the evaluation of monetary reward schemes. We present results from a randomized evaluation of a teacher and principal incentive programme in India that offered a non-pecuniary recognition reward based on students’ test scores on standardized assessments. We find a positive (0.16 SD) yet statistically insignificant effect on student performance when both teachers and principals are incentivized. In schools where only teachers are incentivized, the estimates remain statistically insignificant but are also much smaller in magnitude (0.012 SD). Our findings provide suggestive evidence that recognition rewards may have the potential as a low-cost tool to improve student achievement when both teachers and principals are incentivized. However, further research is required to substantiate the findings and investigate the mechanisms at play.

KEYWORDS:

• Student achievement
• teachers
• principals
• Non-Monetary incentives
• RCT

JEL CLASSIFICATIONS:

• I21
• O15
• D91
• C93

Acknowledgement

Puneet Arora is funded by the Excellent Teams Project of the Faculty of Business Administration, Prague University of Economics and Business (IP310031). Tareena Musaddiq’s postdoc fellowship is supported by training grant R305B170015 from the U.S. Department of Education's Institute of Education Sciences.

Disclosure statement

Abhinav Vats was formerly Chief Minister’s Good Governance Associate for the Palwal District in India. He is currently pursuing his M.B.A at Ross School of Business and no longer works for the Indian government in any capacity.

Ethics approval

This study received the ethics approval from Georgia State University: IRB Protocol ID# H18636. Consent requirement was waived by IRB.

Data availability statement

The experiment for this study was conducted with schools of Palwal School District. The study uses administrative data on student, teachers, principals and schools accessed from the district’s education department. The data shared by the school district is restricted and was shared with the study team under an MOU for this study. The study team cannot share this data without a formal data sharing request approved from the administrators of Palwal School District.

Notes

¹ See Barrera-Osorio et al. (2022) for review of the recognition reward to teachers’ literature. It is one of the very few studies from developing countries studying the impact of recognition incentive for teachers on student performance, but they do not involve principals in their incentive design.

² Prior interventions in education directly incentivizing both teachers and principals are rare. Muralidharan and Sundararaman (2011), one notable study, evaluated a group-level monetary incentive program for the school as a whole. However, unlike our study, it did not offer direct encouragement for teachers and principals to coordinate.

³ The Palwal school district in Haryana overall has 251 public schools spread over 4 blocks (Hassanpur, Hathin, Hodal and Palwal). We partnered with the Palwal block within the larger Palwal school district for our experiment. Palwal block houses 101 public schools across 119 villages. 84 of these 101 schools offer Grade 7.

⁴ The experimental sample had 84 schools across 75 villages. The distribution across the treatment arms was as follows: (i) Treatment 1 had 26 schools from 24 villages, (ii) Treatment 2 had 29 schools from 25 villages and, (iii) Control group had 29 schools across 26 villages.

⁵ Our analytic sample consists of students who appear for both the May and September MAT. We check for systematic attrition at the student level. Results are shown in Table A4. We find an overall attrition rate of 5.5%. Attrition rates for T1, T2, and the control group are 6%, 5.79%, and 4.6%, respectively.

⁶ For robustness, Table 2 also reports randomization inference p-values (with 1000 repetitions; Young 2019) and the wild bootstrapped p-values (with 999 repetitions; Cameron, Gelbach, and Miller 2008) below each estimate. The results do not change qualitatively. We also do not find any clear distributional effects by student baseline performance or meaningful difference in estimated treatment effects by teacher gender for either treatments. While it would have been interesting to study other heterogeneous effects and the potential mechanisms at play, we are limited by the administrative data restrictions with respect to what information could be availed for the study..