Designing and Testing a Scalable Teacher Incentive Programme in Tanzania | Learning Note
School participation in Tanzania has increased dramatically over the past two decades: primary school enrolment increased from 4.9 million in 2001 to 10.9 million in 2020. While 81 per cent of primary-school-age children are currently enrolled, over the last ten years, the primary completion rate has dropped and remains below 70 per cent since 2015 (data from UNESCO Institute for Statistics).1
Despite improvements in enrolment, indicators of foundational learning remain low. According to the 2020 report of the Standard Two National Assessment (STNA), conducted by the National Examinations Council of Tanzania (NECTA), in 2019 five per cent of Grade 2 students pass the benchmark for reading proficiency (“Can correctly read exactly 50 words of the passage in one minute and with 80 per cent or higher comprehension”). The report finds that 17 per cent of students pass the benchmark (80 per cent correct) of the addition and subtraction sub-tasks.
These outcomes are not the result of students’ lack of academic aspiration: according to the RISE Tanzania baseline survey, 73 per cent of Grade 2 and 3 students say they would like to complete secondary school or university.
In a recent report, the Global Education Evidence Advisory Panel (World Bank, 2020) asked what programmes and policies are the most cost-effective instruments for addressing the learning crisis and improving learning for all children. The report creates three categories: the “great buys” category includes programmes that provide very low-cost but salient information on the benefits, costs, and quality of education. The “good buys” category includes programmes that provide structured pedagogy, instruction targeted by learning level, merit-based scholarships and pre-school interventions. Finally, the category “promising but low-evidence” includes teacher accountability and incentive reforms.
KiuFunza, a teacher performance pay programme in Tanzania, fits this last category. KiuFunza (shorthand for Kiu ya Kujifunza or Thirst to Learn) provides test-score linked cash incentives to teachers in Grades 1, 2, and 3 to increase foundational literacy and numeracy outcomes for students. The programme is managed by Twaweza East Africa, a Civil Society Organization, and was set up to provide evidence on the impact of teacher incentives in a series of experimental evaluations. This note discusses the rationale for teacher incentives in Tanzania, the design elements of KiuFunza and preliminary results for the most recent phase of KiuFunza (this phase was implemented in 2019-2021 and the impact evaluation is part of the RISE Tanzania research agenda).
- Five percent of primary Grade 2 students in Tanzania pass the oral reading fluency benchmark on the Standard Two National Assessment and 17 percent pass the benchmark for addition and subtraction sub-tasks (NECTA, 2020).
- Low teacher morale and lack of independent assessments in the early grades help explain these low foundational pass rates. A large majority of teachers prefer teaching in upper rather than lower primary grades. Teachers in Grades 2 and 3 over estimate the foundational reading and numeracy skills of their students by more than 100 percent, when compared with results from an independent assessment.
- We describe results of KiuFunza 3, a teacher performance pay programme that offers individual cash rewards for core foundational skills to teachers in lower primary. The programme uses a simple incentive design based on results from an earlier study. To make the programme easier to scale, we further simplified implementation and sharply reduced the costs of high-stakes student assessments. Despite the lower cost, we find the incentives cause substantial improvements in foundational reading and numeracy.
- Despite potential concerns that bonus pay may lead teachers to focus on well-performing students, we do not find systematic differences in the treatment effects by baseline scores in any year.
Read the full publication by Youdi Schipper, Isaac Mbiti (both RISE Tanzania Country Research Team), and Mauricio Romero (RISE India Country Research Team) in PDF here.
This post was originally published on the RISE Programme website.