A School-Level Experiment on the House System in India

Last registered on February 04, 2026

Pre-Trial

Trial Information

General Information

Title
A School-Level Experiment on the House System in India
RCT ID
AEARCTR-0017811
Initial registration date
February 01, 2026

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
February 04, 2026, 10:06 AM EST

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

Locations

Region

Primary Investigator

Affiliation
UBC

Other Primary Investigator(s)

PI Affiliation
Duke University
PI Affiliation
University of Pennsylvania
PI Affiliation
University of British Columbia
PI Affiliation
Indian Administrative Service

Additional Trial Information

Status
On going
Start date
2024-07-21
End date
2027-12-31
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
Segregation along group lines can perpetuate discrimination and inequality. One social psychological approach to reshaping relations between groups is "recategorization" -- introducing a new group identity that cuts across pre-existing identities. We study this approach through a partnership with government officials in India. We randomized secondary schools to introduce a House System or to operate as business-as-usual. The House System randomly assigns students to four Houses, with school activities organized in these House groups. We will evaluate impacts on social networks, group identity, student well-being, and school behaviors.
External Link(s)

Registration Citation

Citation
Ghosh, Arkadev et al. 2026. "A School-Level Experiment on the House System in India." AEA RCT Registry. February 04. https://doi.org/10.1257/rct.17811-1.0
Sponsors & Partners

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

Request Information
Experimental Details

Interventions

Intervention(s)
Intervention Start Date
2024-07-21
Intervention End Date
2027-08-31

Primary Outcomes

Primary Outcomes (end points)
*** Student Endline Survey ***

FIRST STAGE
Whether student reports that their school has a House System and if they know the name of their House (or don’t know).

If reports that their school has a House System: (i) which House has the most points currently (or don’t know), (ii) what House activities did they take part in this academic year, and (iii) how often have Houses been used to group people.

If reports yes, school has a House System, AND in treated school: (i) the names of the other Houses (or don’t know), (ii) which House are random 2 of 5 closest friends in (or don’t know).

All students are asked: (i) how often teachers offer rewards for good school behavior, (ii) how often they participated in extra-curricular activities, (iii) how often they had been put into groups for school activities, and (iv) how do teachers typically form these groups.

SOCIAL LIFE
Friendship networks: five closest friends from a pre-populated list of classmates. (We can match these friends to caste, gender, grade, section, and religion predicted by name, using student-level administrative data. For treated schools, we can also match to the House a student was randomly assigned to.)

Social isolation: (i) how many friends met up with outside school over the past week, (ii) how many days spent lunch break alone over the past 7 days at school, (iii) how many schoolmates would feel comfortable going to with a personal problem.

Loneliness: how often do you feel left out at school?

WELL-BEING AT SCHOOL
(i) How often the respondent has been bullied or teased in the past 2 months, and how often they have seen classmates bullied or teased, (ii) happiness at school over past 2 weeks (0 to 10), (iii) PHQ-4 scale for last 2 weeks at school, and (iv) how much enjoy coming to school (0 to 10).

GROUP IDENTITY
Identity knowledge: whether the student knows their own, and their closest friends’, caste category and jati; and knowledge of the caste category of 3 randomly selected classmates, along with accuracy of guesses.

Identity markers: Whether the surveyor observes that the student is wearing a caste-colored wrist thread, the respondent’s report of whether classmates wear these wrist threads outside of school, and knowledge of which caste wears a red and yellow wrist thread.

Feeling thermometer towards each of the four Houses (only for students in treated schools that report that their school has a House System).

Identity fusion: overlapping circles question for caste/jati, school, religion, and being Indian. Same question for “your House” asked to those that report that their school has a House System.

Feeling thermometer score toward three randomly chosen classmates.

PERCEIVED NORMS
Students are asked to guess how many students out of 10 in their grade would say that they “strongly agree” with the six caste- and gender-related statements listed below in Secondary Outcomes (i.e. we are pre-registering the first-order beliefs as secondary, and the second-order beliefs as secondary, since we think the intervention is more likely to affect perceived norms than actual attitudes).

*** Administrative Data ***

Student-level attendance. (We are also hoping to get student-level test scores for 2026, but this is not guaranteed).
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
*** Student Endline Survey ***

ATTITUDES
Agreement with: (i) caste identity is important for a person’s chances of success in India today, (ii) it is ok to ask someone else what their caste is, (iii) it is ok to joke about someone’s caste or religion, (iv) boys tend to be better than girls at math and science, (v) it is just as important for women to have a career as it is for men, and (vi) girls can be just as good leaders as boys.

ASPIRATIONS AND EXPECTATIONS
Highest level of education desired, and highest level of education expected.

OTHER
Whether the student knows their religion (we expect this to be close to 100%, so we will likely have ceiling effects), and that of their 5 closest friends. How do they know the caste category of one randomly selected friend (mainly to get basic descriptives for how people learn the caste of others). Knowledge of whether random classmates have siblings or not (to measure non-identity-related learning about others).

Stated preference over school with House system or school without; same choice but school without has 2 hours less homework per week; same choice but school without has 5 hours less homework (these House preference questions are only asked to those that report having a House System in their school).
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
[Note: due to a tight timeline before school exams, we did not have time to pilot our endline survey on out-of-sample schools. In addition, we wanted to confirm the presence of a large “first stage” before deciding to roll out the full endline (since without a first stage, the endline would be a waste of money). As a result, we posted our pre-registration after 3 days of surveying were complete (~10% of schools), at which point we were confident in a reasonable first stage.]

We worked with a local NGO to run a school-level RCT in Tirunelveli district, Tamil Nadu. We worked with a sample of 176 secondary schools, randomizing half of those schools to implement a “House System,” with children randomized to one of four Houses, and school activities using the Houses for grouping students. We think of this intervention as a re-categorization-based approach (from social psychology), with the new group identity marker of “Houses” potentially reshaping the nature of pre-existing group identities, and with possible effects on broader student behaviors. The intervention began during the 2024/25 school year, and to save on cost, we did not administer a baseline survey.

We are now launching an endline survey of students. We are aiming to survey a random sample of 20 students per school, evenly spread across genders and school grades.

Some of the survey questions will help us establish a “first stage” – checking that the House System is both more common, and more intensively implemented, in treatment schools than in control schools. From some implementation monitoring and our initial endline surveys, we already know that our first stage is not one. As a result, we will focus on intent-to-treat (ITT) estimates, with some discussion of possible implied IV coefficients given different assumptions on what is the relevant “first stage.”

For our core analysis, we will use student-level data to run intent-to-treat OLS regressions with randomization strata and school grade fixed effects, and standard errors clustered at the school-level. For natural families of outcomes, we will sometimes create an index by standardizing each outcome and taking the simple mean. We will explore heterogeneity along four dimensions: caste, grade, gender, and school size (many schools in the sample are small, and it is not clear that a House System would have meaningful effects in schools that are small enough for everyone to already know everyone).

Using our friendship network data, we will also run dyadic regressions to explore the role of group identity (e.g. same-House, same-caste, same-religion, same-gender) on friendship formation.

In the treated schools, we also had a classroom-level sub-randomization. To be eligible for the sub-randomization, the treated classroom had to have all four Houses represented, and with at least one SC/ST and one non-SC/ST student in each House. We randomized half of these eligible classrooms to be treated. Treatment entailed randomly picking two of the Houses in that classroom, and moving SC/ST students from one House to the other, increasing the variation in SC/ST share across the Houses within that classroom.

The sub-treatment has two implications for our analysis. First, we can use the within-treatment variation to look at the effects of a student having more versus fewer SC/ST students in their House in their class (this analysis will be somewhat under-powered, though we can also combine the natural variation that already comes from random assignment to Houses). Second, when looking at the overall effects of the House System (comparing treatment versus control schools), we can explore excluding the students treated by the sub-randomization (and reweighting accordingly), since this sub-randomization creates more caste homogeneity within a House, which is not what an actual House System policy would do.
Experimental Design Details
Not available
Randomization Method
STATA.
Randomization Unit
School (for House System), Student (for assigned House).
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
176 schools were randomised, though 6 schools were subsequently excluded from the study sample (prior to any endline data collection), since they were classified as special schools or had closed due to under-enrollment. This leaves us with 170 schools.
Sample size: planned number of observations
Roughly 3,030 (170 schools, aiming for 20 students per school, but some schools are small enough that the eligible sample is fewer than 20).
Sample size (or number of clusters) by treatment arms
84 treatment schools, 86 control schools
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB

Institutional Review Boards (IRBs)

IRB Name
University of British Columbia
IRB Approval Date
2026-01-22
IRB Approval Number
H25-04214