School Inspections, Bureaucrats, and School Quality: Experimental Evidence from Tanzania

How to cite

CILLIERS, Jacobus; Habyarimana, James. 2020. "School Inspections, Bureaucrats, and School Quality: Experimental Evidence from Tanzania." AEA RCT Registry. AEARCTR-0005714. https://doi.org/10.1257/rct.5714-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Jacobus CILLIERS* , James Habyarimana

* Primary investigator

Registry ID
AEARCTR-0005714
DOI
https://doi.org/10.1257/rct.5714-1.0
First registered
Last updated
Status
On going
Trial dates
2019-01-01 – 2021-03-31
Countries
Tanzania, United Republic of ()
Registry record
http://www.socialscienceregistry.org/trials/5714

Abstract

The government of Tanzania has recently reformed its school inspection program, reframing the formerly punitive function as a friendly source of support to schools, now called the School Quality Assurance (SQA) program. A key part of the reform is to replace the traditional school inspections with Whole School Visits (WSVs), conducted by $3-4$ School Quality Assurance Officers (SQAOs). Each SQA visit produces a set of diagnostics and recommendations across a range of domains that are shared with all stakeholders. The Ward Education Officers (WEOs) are then responsible for conducting follow-up visits to the schools to monitor compliance to these recommendations. We propose a randomized evaluation of the SQA program, as well as an additional treatment to increase monitoring and oversight by the WEOs. Out of a nationally representative sample of $397$ schools we randomly assign $198$ schools to receive the WSVs. Moreover, in half of these schools, also randomly assigned, WEOs are encouraged through text-based reminders to engage head teachers on how they are addressing the recommendations of the SQAOs. We collected baseline data in February-March 2019, midline data in Februray-March 2020, and plan to conduct endline data in November 2020. We will estimate the impact of these interventions on the beliefs and behaviors of the Ward Education Officers, head teachers, teachers, and the school community, as well as the learning outcomes of a cohort of students who were in grade 2 and 3 at baseline.

Intervention

1. School Quality Assurance Officers, previously known as school inspectors, visit schools for 2-3 days, and provide an assessment of overall school quality as well as recommendations for improvement. They share this report with the school, and also government.
2. Ward Education Officers receive a text message summarizing the main recommendations and encouraging to conduct follow-up visits at the schools.

Experimental design

The program is evaluated using a cluster randomized control trial, with a random phased-in design and randomization taking place at the ward level. Out of a nationally representative sample of 397 wards, 198 wards are randomly assigned to treatment, and 199 wards are assigned to control. We then randomly sampled one school in each ward to participate in the study. The treatment schools are assigned to receive a Whole School Visit at some point between April and November 2019. In (randomly assigned) half of the treatment schools Ward Education Officers receive regular text messages informing them of the recommendations made at the WSV. The remaining 199 schools are assigned to only receive WSVs after the completion of the planned endline data collection in November 2020.

Design details

Randomization method
Randomization done in office by a computer.
Randomization unit
Ward level (one school sampled per ward)
Clusters
397 schools.
Observations
397 schools, 3,970 teachers, and 7,940 students.
Arms
1. Whole School Visits only: 99 schools 2. Whole School Visits and text messages sent to the Ward Education Officers: 99 schools. 3. Control: 199 schools.
Primary outcomes
Beliefs of school quality, quality of school management, teaching quality, and student learning.

Cited by

No citation data available for this registration yet.

Keywords