Pecuniary & Non Pecuniary Incentives for Teachers : Evidence from Nigeria

Methods

Summary

To investigate the effectiveness of monetary and non-monetary incentives on teachers’ effort and effectiveness, we’ll run an experiment in which different groups of teachers will be differently incentivized. We will then observe our outcomes of interest: presence and nature of after school lessons, time spent working outside the classroom in school and students’ survey. We will assign different treatments to different schools to prevent data contamination and spillover effects of the incentives across treatment and control groups.  The treatment and control schools will be chosen randomly from the population of interest and in each school, these teachers will be randomly grouped into small groups of 4 or 5 individuals.
The incentives will be implemented as an award given to the 3 groups of teachers with the highest average improvement in effort/productivity (as measured by our key indicators). The monetary incentives will be either NGN 20,000 (US $50) for a group of 5 teachers; the near money incentive will be a similar value gift card for a popular shopping mall in Lagos; the non-monetary incentive will be an award of excellence given at a school assembly. Baseline and end-line data will be collected through the teacher effectiveness surveys and general knowledge tests. 

Challenges

We do not anticipate any significant challenges over the course of this project, and there are minimal risks associated with the project. Some minor challenges that may be encountered include:
• Printing, transportation and administration of over 3000 teacher effectiveness surveys and general knowledge tests for the students.
• Possible resistance from the teachers
• Lack of co-operation from school principals etc.

Pre Analysis Plan

By estimating a simple OLS regression of the outcomes of interest on the independent treatment variables we can see the differential impact of the various incentives on the outcome of interest.

The basic empirical model we will have incentives as the independent variable and the outcomes of teacher effort and students academic performance as dependent variables.