Persistence of Teacher Behavior Change

Last registered on May 06, 2025
View Trial History

Pre-Trial

Trial Information

General Information

Title
Persistence of Teacher Behavior Change
RCT ID
AEARCTR-0015906
Initial registration date
May 01, 2025

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
May 06, 2025, 5:01 AM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Country
Pakistan
Region
Islamabad Capital Territory

Primary Investigator

Name
Christina Brown
Affiliation
University of Chicago
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Tahir Andrabi
PI Affiliation
Pomona College
PI Name
Asim Khwaja
PI Affiliation
Harvard Kennedy School
PI Name
Isabel Macdonald
PI Affiliation
Irrational Labs
PI Name
Maroof Syed
PI Affiliation
Centre for Economic Research in Pakistan

Additional Trial Information

Status
In development
Start date
2025-05-01
End date
2025-12-31
Keywords
Education
Additional Keywords
education technology
JEL code(s)
Secondary IDs
Prior work
This trial is based on or builds upon one or more prior RCTs.
Abstract
This trial is a follow up study of a previously conducted RCT titled “Technology to Empower Actors Across the Learning Ecosystem”, implemented in over 500 public schools in Islamabad Capital Territory, Pakistan. Teachers were randomly assigned to (i) a control group (business as usual), (ii) a high-discretion treatment, where they received information on student learning levels and materials to tailor instruction, or (iii) a low-discretion treatment, which added structured daily guidance on student-level use of the materials. Preliminary results show that the intervention led to gains of 0.14–0.17 standard deviations in student learning outcomes. Building on these results, the current phase of the study focuses exclusively on workbook provision, the most cost-effective component of the original interventions. It evaluates the persistence and scalability of the program by (i) examining whether initial gains in student learning sustain after direct support ends, (ii) testing whether teachers continue to adopt improved instructional practices with new student cohorts, and (iii) assessing whether repeated use of the intervention improves, maintains, or diminishes its effectiveness. Findings from this study will offer evidence on the durability of low-cost, information-driven interventions for improving foundational learning and highlight strategies for institutionalizing such practices in public education systems.
External Link(s)

Registration Citation

Citation
Andrabi, Tahir et al. 2025. "Persistence of Teacher Behavior Change." AEA RCT Registry. May 06. https://doi.org/10.1257/rct.15906-1.0
Sponsors & Partners
Experimental Details

Interventions

Intervention(s)
We will study the effects of three sets of interventions.

Phase 1 interventions (conducted in Fall 2022-Spring 2023):

Teachers are assigned to one of four arms:
Control: Basic training
Workbook: Basic training + workbook with differentiated material
Information: Workbook components + Information about student learning levels
Guidance: Information components + Guidance on which students should receive which practice problems from the workbook

Parents are assigned one of the three arms:
Control: Status quo
Information: Information about student learning levels
Guidance: Guidance on which students should receive which practice problems from the workbook

Phase 2 interventions (Spring 2025-Spring 2026):

Schools are assigned to one of the following:
Control: Status quo
Workbook: Provision of workbook with differentiated material
Reminders: Reminders about effectiveness of phase 1 interventions
Workbook + Reminders
Intervention (Hidden)
We will provide workbooks to students and teaching guides to teachers at the beginning of the new academic year and continue monitoring them for the rest of the academic year through classroom observations and teacher surveys. Students will be tested on their skills in math once at the beginning of the academic year and for the second time near the end of the academic year to measure any difference in their learning levels. Teachers will be observed to identify whether they continue to use the materials (workbooks) and teaching practices introduced to them in the prior intervention. It will build upon the trial listed as AEARCTR-0009905.
Intervention Start Date
2025-05-15
Intervention End Date
2025-12-01

Primary Outcomes

Primary Outcomes (end points)
1. Math test scores
2. Teacher usage of the workbooks
3. Teacher pedagogy as measured by the CLASS observation tool
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
English and Urdu test scores
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
Phase 2 will be randomized at the school level. Randomization is stratified by school baseline test scores.
Experimental Design Details
This study builds upon the prior intervention by randomly assigning half of the participating schools to receive a second round of math workbooks and teacher guides (Workbook group), while the other half continue with standard practices (Control group). In control schools, teachers conduct "business as usual" without additional materials, allowing us to isolate the impact of continued workbook access.

The study measures outcomes at both student and teacher levels. For students, we assess whether learning gains from the original intervention persist even without further intervention support, and whether newly exposed students (those previously too young or not enrolled) benefit when their teachers receive the workbooks for the first time. For teachers, we evaluate whether instructional practices developed during the initial intervention endure.

Three key subgroups enable these comparisons: (1) Originally treated students with control teachers (to test persistence of learning gains); (2) New students with treated teachers (to determine if teachers sustain effective practices for new cohorts); and (3) Pure control students and teachers. This design addresses critical questions about the longevity of educational interventions—whether student improvements fade without reinforcement, and whether teacher training effects spill over to future classrooms.

The study measures outcomes at both student and teacher levels. For students, we assess whether learning gains from the original intervention persist even without further intervention support, and whether newly exposed students (those previously too young or not enrolled) benefit when their teachers receive the workbooks for the first time. For teachers, we evaluate whether instructional practices developed during the initial intervention endure.

Three key subgroups enable these comparisons: (1) Originally treated students with control teachers (to test persistence of learning gains); (2) New students with treated teachers (to determine if teachers sustain effective practices for new cohorts); and (3) Pure control students and teachers. This design addresses critical questions about the longevity of educational interventions—whether student improvements fade without reinforcement, and whether teacher training effects spill over to future classrooms.
Randomization Method
Randomization done in Stata
Randomization Unit
Workbook: school
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
326 public schools
Sample size: planned number of observations
80,000 students
Sample size (or number of clusters) by treatment arms
Schools receiving the workbooks: 163 schools
Schools receiving reminder messages: 163 schools
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
CERP Institutional Review Board
IRB Approval Date
2024-12-22
IRB Approval Number
Protocol ID 2024-003
Analysis Plan

Post-Trial

Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials