Pygmalion Effect and Classroom Networks: Evidence from Schools in Pakistan

How to cite

Jain, Ronak; Asim, Minahil; Khandelwal, Vatsal. 2021. "Pygmalion Effect and Classroom Networks: Evidence from Schools in Pakistan." AEA RCT Registry. AEARCTR-0007846. https://doi.org/10.1257/rct.7846-2.0

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

Ronak Jain* , Minahil Asim , Vatsal Khandelwal

* Primary investigator

Registry ID
AEARCTR-0007846
DOI
https://doi.org/10.1257/rct.7846-2.0
First registered
Last updated
Status
On going
Trial dates
2021-02-16 – 2022-01-31
Countries
Private
JEL codes
D91, D84, I24, I25, C93, D83
Registry record
http://www.socialscienceregistry.org/trials/7846

Abstract

Teacher expectations, as literature in psychology has shown, can matter for a student’s performance by creating self-fulfilling prophecies, which is known as the Pygmalion Effect (Rosenthal & Jacobson, 1968, 1992). Using an experimental design, we aim to shed light on answering whether this hypothesized effect of teacher expectations can be harnessed to improve student effort, achievement and engagement, in a low-resource context, like Pakistan. As part of our treatment, teachers deliver statements conveying high expectations of effort to students. Using an experimental design, our study aims to shed light on whether: (a) expectations of high effort and achievement affect student performance and engagement; (b) the treatment effect differs according to whether these expectations are conveyed to a student individually or in a pair and (c) the treatment effect differs by gender. We will also analyze how the intervention affects student peer interactions.

Intervention

Using an experimental design, we aim to answer whether high teacher expectations can be used to improve student effort, cognitive and non-cognitive outcomes, parental engagement, and teacher engagement.

Experimental design

We divided ~320 classrooms in our study into three groups: (1) Control Group (status quo) (2) Individual Treatment Arm, where student-specific high teacher expectations and encouragement were conveyed individually and (3) Peer Treatment Arm, where student-specific high teacher expectations and encouragement were conveyed to a student and in addition, they were randomly matched with another classmate (of same gender) and were encouraged to work together.

Design details

Randomization method
Cluster Randomization (using STATA)
Randomization unit
Classrooms at the first stage of randomization (i.e. dividing our sample into control and treatment group). In our treatment arms, we further use within classroom randomization at the second stage.
Clusters
~320 Classrooms
Observations
~2000 Students and Parents and ~130 Teachers.
Arms
~106 classrooms per treatment
Primary outcomes
Our main student-level outcomes include self-efficacy, self-esteem, motivation and effort, student stress levels, growth mindset attitudes, and academic performance. Our main class-level outcomes include attendance, engagement, punishments/warnings, classroom behavior.
Secondary outcomes
We intend to additionally measure the following teacher-level outcomes: expectations, engagement, and perceived interest in child's education. Our parental-level outcomes include engagement, investment in child's education and beliefs.

Cited by

No citation data available for this registration yet.

Keywords