WE PAY FOR RESULTS, LITERACY, LEARNING,

The core intervention model of KiuFunza is simple (see diagram):

Baseline: At the start of the school year, program staff visit a school, explain the bonus offer, sign-up teachers in grades 1, 2 and 3 and ask for their bank details;

Endline: At the end of the school year they assess students on a basic skills test (reading, writing and arithmetic);

Payment and feedback: Then, based on test scores, payments are calculated and incentives are paid at the start of the next school year. Schools and administrators receive a report on student performance by grade-subject to help them identify gaps and improve their practice.

These steps are based on the practices of the most effective organizations around the world. Research shows that highly effective organizations (1) set clear targets and incentives for their employees; (2) measure progress towards these targets; and (3) provide feedback and make sure the employee’s effort and progress toward the target has personalized consequences.

The KiuFunza incentive scheme has three levels of incentives: incentives for subject teachers in grades 1–3, for head teachers, and a school level infrastructure bonus. The first two incentive types are paid to teachers’ own bank accounts, the infrastructure bonus to the school bank account.

The subject teacher incentive scheme rewards teachers for a few proficiency levels per subject. For example, there are three levels for grade 2 reading: words, sentences and short story. When a student passes one of the levels, the responsible teacher receives a small bonus. The more skills a student passes, the more the teacher earns. Harder skills (with lower overall pass rates) earn the teachers more money. 

Our research has shown that a range of thresholds works better than one high threshold since it can provide incentives to teachers even if their students perform poorly at the start of the intervention. The assessments are sample based. Teachers earn money based on the performance of a random sample of students in a grade. The sample results are used to pay teachers for all students in the grade. 

There are two other incentive types: the head teacher receives 20 percent of the combined subject teacher payments in her or his school. In addition, there is a school level infrastructure bonus payable to the school account. The infrastructure bonus is paid each year in each of the ten regions to the school with the highest pass rate from among the lowest performing 25 percent of schools.  

The explanation of the incentive offer is done by a small team visiting the school at the start of the school year (baseline). The team brings an information booklet where teachers can read details of the offer. Teachers are also shown two instruction videos about the KiuFunza intervention: see below for these videos (Kiswahili spoken with English subtitles). 

Here is an overview of the curriculum skills used in the most recent KiuFunza assessments: 

The student mastery of these skills is tested in the “high stakes” endline assessments. These assessments are used to calculate the performance rewards paid to the teachers: the better students perform, the more their teachers earn. To guarantee integrity of the testing procedure, two types of random selection are used. First, students are selected randomly from the student population in each grade. Second, for each student assessment, a test set is randomly selected from a set of ten.  

The assessments are administered in a one-on-one setting, by experienced Twaweza volunteers. They sit down with each of the randomly selected students and guide them through the randomly selected test set that is printed in a large font test booklet.