Experimental Evidence on the Determinants of Peer Effects during Learning

Last registered on January 09, 2020

View Trial History

Pre-Trial

Trial Information

General Information

Title

Experimental Evidence on the Determinants of Peer Effects during Learning

RCT ID

AEARCTR-0005163

Initial registration date

January 06, 2020

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

January 09, 2020, 3:39 PM EST

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

Locations

Country

Belgium

Region

Flanders

Primary Investigator

Name

Kaat Iterbeke

Affiliation

KU Leuven

Contact Primary Investigator

Other Primary Investigator(s)

PI Name

Wouter Schelfhout

PI Affiliation

UAntwerpen

Contact Investigator

PI Name

Koen Declercq

PI Affiliation

KU Leuven

Contact Investigator

PI Name

Kristof De Witte

PI Affiliation

KU Leuven

Contact Investigator

Additional Trial Information

Status

In development

Start date

2019-12-01

End date

2020-12-31

Keywords

Education

Additional Keywords

Peer effects, Randomised control trial, Entrepreneurship education, Non-cognitive skills

JEL code(s)

C93, I21, L26

Secondary IDs

Abstract

Using a randomised control trial in Flemish secondary education, our study aims to investigate the determinants of peer effects in a computer-assisted entrepreneurship education programme. To establish our results, schools are randomly assigned to a control condition, an experimental condition in which students work individually, and an experimental condition in which students work in pairs. The comparison of the three conditions allows us to separately identify the effect of the programme itself and the effect of cooperative learning on various student outcomes. Moreover, using the second experimental condition, we are able to identify the causal impact of differences in peer characteristics by randomly assigning students to pairs. We consider a large set of individual and peer characteristics (such as the socioeconomic background, cognitive ability, learning preferences, and personality traits). In addition, we aim to identify the underlying channel of cooperative learning using the quality of cooperation (such as the degree of discussion and helpfulness of the peer) and students’ motivational state as instruments.

External Link(s)

Registration Citation

Citation

De Witte, Kristof et al. 2020. "Experimental Evidence on the Determinants of Peer Effects during Learning." AEA RCT Registry. January 09. https://doi.org/10.1257/rct.5163-1.0

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)

Schools will be randomly assigned to either the control condition or one of the experimental conditions after the completion of the pre-test. We randomise on three different levels to ensure proper identification of (heterogeneous) peer effects and to obtain suitable control groups:
1. Assignment of schools to the control or experimental condition,
2. Within the experimental condition, assignment of schools to the individual or pair treatment,
3. Within the pair treatment, assignment of students to random pairs.
Students in the control schools complete the pre- and first post-test at the same time as students in the treatment schools, and receive no entrepreneurship education meanwhile. Students in the pair treatment share a computer and complete the programme in pairs of two. The student pairs are randomly formed by the teacher at the start of the lectures using a matching game. Students in the individual treatment complete the computer-assisted programme individually.

Intervention Start Date

2020-01-15

Intervention End Date

2020-06-30

Primary Outcomes

Primary Outcomes (end points)

Students’ knowledge of entrepreneurship

Primary Outcomes (explanation)

Students’ knowledge of entrepreneurship is measured via ten questions that refer directly to the didactic material.

Secondary Outcomes

Secondary Outcomes (end points)

In addition to the performance in a knowledge test and given the nature of the programme, i.e. entrepreneurship education, changes in entrepreneurial intentions and attitudes toward entrepreneurship of students will be examined using nine statements. Moreover, given we have information on the overall profit or loss students turn during the business simulation, we are able to measure whether cooperative learning influences students’ performance during the lectures.

Secondary Outcomes (explanation)

Experimental Design

An open call for schools to participate was launched at the beginning of December 2019. Participating students are in grade 9, 10, 11, and 12. Schools that register for participation will be randomised to the aforementioned three conditions.

To evaluate the effectiveness of the programme and the impact of peers, students have to take three tests, i.e., a first test prior to the lectures to measure students’ baseline knowledge and background characteristics (such as the socioeconomic background, cognitive ability, learning preferences, and personality traits), a second test during the last lecture to capture the short-term effects of the programme (with respect to students' knowledge of entrepreneurship, entrepreneurial intentions and attitudes), and a third test administered approximately four to six weeks after the lectures to measure longer-term effects.

Experimental Design Details

Randomization Method

Schools will be randomly assigned to the conditions using a random number generator in Stata.

Randomization Unit

We randomise on three different levels to ensure proper identification of (heterogeneous) peer effects and to obtain suitable control groups:
1. Assignment of schools to the control or experimental condition,
2. Within the experimental condition, assignment of schools to the individual or pair treatment,
3. Within the pair treatment, assignment of students to random pairs.

Was the treatment clustered?

Yes

Experiment Characteristics

Sample size: planned number of clusters

60 schools

Sample size: planned number of observations

1800 students

Sample size (or number of clusters) by treatment arms

Control condition: 480 students, 16 schools
Experimental condition I (Individual treatment): 480 students, 16 schools
Experimental condition II (Pair treatment): 840 students, 28 schools

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

See pre-analysis plan.

Supporting Documents and Materials

IRB