Do team-based input incentives boost skill development? Evidence from India

How to cite

Jain, Tarun; Debnath, Sisir. 2018. "Do team-based input incentives boost skill development? Evidence from India." AEA RCT Registry. AEARCTR-0003645. https://doi.org/10.1257/rct.3645-1.0

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

Tarun Jain* , Sisir Debnath

* Primary investigator

Registry ID
AEARCTR-0003645
DOI
https://doi.org/10.1257/rct.3645-1.0
First registered
Last updated
Status
On going
Trial dates
2017-09-01 – 2019-09-01
Countries
Private
JEL codes
I21, M52
Registry record
http://www.socialscienceregistry.org/trials/3645

Abstract

Skill development programs are sub-optimally effective because participants do not attend frequently enough, or because they drop out of the program before completion. Especially while learning skills with non-linear rewards to attendance, such as English language and soft skills, motivating students to attend more frequently, and to stick with the program till completion might yield significantly better outcomes.

Drawing on the case of a one year English and non-cognitive skills development program in India, we analyze whether group incentives for attendance can be effective in increasing attendance and reducing dropout. To investigate this question, we divide the program’s sessions (at the instructor level) into four treatment arms: (1) A treatment where students are combined into self-chosen two-person teams, with rewards for the team with the maximum joint attendance in a term, (2) A treatment where students are combined into exogenously chosen two-person teams, with rewards for the team with the maximum joint attendance in a term, (3) A treatment where students individually rewarded for maximum attendance in a term, and (4) A control group where no incentives are implemented. We anticipate that the team incentives (treatments 1 and 2) will increase attendance and decrease dropout compared to the individual rewards. We further analyze the impact of the incentives on the development of English and non-cognitive skills.

Intervention

The experiment was conducted with the universe of the partner organization’s students. The organization consists of multiple facilitators, each running 4 sessions per day with 20 students each. We randomize the facilitators into the treatment arms described above. The baseline data was collected starting September 2017, as new sessions were beginning. The one year program lasted till December 2018, when we started collecting the endline data.

Experimental design

The experiment was conducted with the universe of the partner organization’s students. The organization consists of multiple facilitators, each running 4 sessions per day with 20 students each. We randomize the facilitators into the treatment arms described above. The baseline data was collected starting September 2017, as new sessions were beginning. The one year program lasted till December 2018, when we started collecting the endline data.

Design details

Randomization method
We randomize the facilitators who run each session into a treatment arm. The randomization was conducted on a computer using Stata.
Randomization unit
The unit of randomization is a facilitator who runs four sessions each.
Clusters
There are 253 facilitators in the experiment.
Observations
The anticipated number of students in the experiment is 20240.
Arms
The number of facilitators in each treatment arm is 63 for treatment with self-chosen teams, 61 for treatment with exogenously chosen team, 63 in the individual reward treatment, and 66 in the control group.
Primary outcomes
The primary outcomes are (1) daily attendance as measured by the program's administrative records, (2) dropout, (3) Scores on tests administered by the partner, (4) Scores on cognitive skills test administered through a survey module, (5) Scores on non-cognitive skills test administered through a survey module.

Cited by

No citation data available for this registration yet.

Keywords