Peer Perceptions and Students' Investments in Schoolwork --- Second Randomized Controlled Trial and Additional Data Collection

This project aims (i) at providing incentivized evidence on students' beliefs about their peers' study time, (ii) at evaluating the effects of an information treatment recalibrating students' beliefs about their peers' study time.

The project aims at investigating the effects of the information treatment on the short and longer term (i.e., right after the information treatment, and about a month later), and it also aims at investigating the direct and indirect effects of the treatment (i.e., its direct effects on the students receiving the information treatment, and its indirect effects on the students who did not receive the information treatment but were indirectly treated through their classmates).

2023-02-24

2023-06-30

Firstly, this project aims at providing incentivized evidence on students’ beliefs about their peers’ study time.

Secondly, this project investigates the effects of an information treatment which recalibrates students’ beliefs about their peers’ study time. We study the effects of recalibrating students’ beliefs about their peers’ schoolwork investments on four main outcomes (two short-term outcomes, and two longer-term outcomes):

(i) intended study time measured right after the information treatment (only for students below the median study time of their class).
(ii) students’ willingness to pay for a text-messages coaching program aimed at motivating them to remember and achieve their schoolwork goal, measured right after the information treatment (only for students below the median study time of their class).
(iii) students’ beliefs about their peers’ study time, measured about a month after the information treatment.
(iv) study time during the weeks following the information treatment.

See the pre analysis plan for the secondary outcomes.

To shed light on the mechanisms, the second randomized controlled trial and data collection further investigates how recalibrating students’ beliefs about their peers’ study time affects:

(i) students’ beliefs about the role of study effort vs. ability in explaining schooling performances.
(ii) students’ beliefs about their relative ability compared to their classmates’.
(iii) students’ emotional health.
(iv) students’ educational aspirations.

These secondary outcomes are all measured about a month after the information treatment on peers’ study time.

To better understand the long-term consequences of students’ misperceptions about their peers’ study time, the second randomized controlled trial and data collection also aims at investigating the effects of the information treatment on students’ middle school grades and on their high school enrollment (conditional on available administrative data). However, we acknowledge that we have low expectations that a one-time information treatment could have significant impacts on these longer-run educational outcomes.

As in the first randomized controlled trial, we aim (i) at providing incentivized evidence on students’ beliefs about their peers’ study time, (ii) at evaluating the effects of an information treatment recalibrating students’ beliefs about their peers’ study time.
We aim at investigating the effects of the information treatment on the short and longer terms (i.e., right after the information treatment, and about a month later), and we also aim at investigating its direct and indirect effects (i.e., its direct effects on the students receiving the information treatment, and its indirect effects on the students who did not receive the information treatment but were indirectly treated through their classmates).

The project targets 10th graders in Norwegian middle schools. In the second randomized controlled trial, we plan on recruiting at least 20 schools. We expect about 3 classes per school and about 15 consenting students per class, resulting in a sample size of at least 900 students. We will add this second round of data collection to the first round which comprises 23 schools, 78 classes, and 998 students.

The randomization will be implemented at the class level and stratified within schools. For each school participating in the study, before the intervention, we will randomly select treated and control classes. We will first form strata of two or three classes within each volunteer school; the strata will be based on the consent rate within each class. Within each strata, classes will then be assigned to the treatment or control condition on a random basis.

In each class assigned to the treatment condition, all the students whose reported study time at baseline is strictly below the median of their class will receive an information treatment on their classmates’ study time. Students in classes assigned to the control condition and students whose reported study time at baseline is above the median of their class will not receive the information treatment.

Not available

Computerized randomization using Stata.

Classes.

Yes

At least 60 classes.

At least 900 students.

At least 30 classes randomly assigned to the treatment condition, and at least 30 classes randomly assigned to the control condition.

Within treated classes, about 50% of students will receive the information treatment.

-Intended study time right after the information treatment above the baseline class median: +0.08. -High willingness to pay for a coaching sms-program measured right after the information treatment: +0.10. -Wedge in students’ belief about their peers’ study time one month after the information treatment: -0.06. -Study time one month after the information treatment above the baseline class median: +0.06. The minimum detectable effects are computed with a sample size of 140 classes (minimum expected sample size from the first and second randomized controlled trials) with 7.5 or 15 students per class (for short-term and longer-term outcomes respectively), and using information from the first randomized controlled trial on the mean, standard deviation and intra-class correlations of the main outcomes.