Mental Models of High School Success -- Causal Evidence

How to cite

Hübsch, Theresa; Willadsen, Helene; Mahlstedt, Robert; Pinger, Pia; Settele, Sonja. 2025. "Mental Models of High School Success -- Causal Evidence." AEA RCT Registry. AEARCTR-0015963. https://doi.org/10.1257/rct.15963-1.1

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

Theresa Hübsch* , Helene Willadsen , Robert Mahlstedt , Pia Pinger , Sonja Settele

* Primary investigator

Registry ID
AEARCTR-0015963
DOI
https://doi.org/10.1257/rct.15963-1.1
First registered
Last updated
Status
Completed
Trial dates
2025-05-07 – 2025-08-01
Countries
Denmark ()
JEL codes
D83, D84, I24
Registry record
http://www.socialscienceregistry.org/trials/15963

Abstract

We conduct an experiment with Danish adolescents who approach the end of elementary school to study the causal effect of exposure to information about gender differences in school performance in the context of upper secondary education. We invite 20 percent of the Danish female population born in 2009 to participate in an online survey. In the survey, everyone receives information about the performance of boys in upper secondary education. Subsequently, we randomly assign one half of the respondents to a treatment group that also receives information on the performance of girls in upper secondary school, while the remaining half forms a pure control group that does not receive information. Post-treatment, we elicit respondents’ own success expectations if they were to start upper secondary education – our main outcome of interest. We also measure posterior beliefs about the performance of girls relative to boys in upper secondary education, as well as a set of secondary outcomes, such as planned study effort and hypothetical school track recommendation decisions to male vs. female peers.

Experimental design

A main and a follow-up survey with adolescent girls currently enrolled in Danish schools, mainly in grade 9.
In our main survey, participants will be randomly assigned (50-50) into treatment and control group. Participants in the treatment group receive true information about the success rate of girls and boys with similar grades, while the control group receives no information treatment. Subsequently, we measure various beliefs about respondents' own success chances and the success chances of other adolescents.
In the follow-up survey, there is no new information provided, nor is the originally provided information repeated. We simply measure the main outcomes: posterior beliefs (success probabilities beliefs about one's own cohort), expected probability of own success, and expected final GPA .

Design details

Randomization method
Randomization done in office by a computer
Randomization unit
Individual level
Clusters
700 - 800 adolescent girls
Observations
700 - 800 adolescent girls
Arms
350-400 individuals treated, 350-400 individuals control
Primary outcomes
- Expected probability of own success (graduating high school with a GPA ≥ 7) - Expected final GPA - Posterior beliefs: Updated success probability beliefs about one's own cohort
Secondary outcomes
- Effort investment (expected study hours per week) - Expected enjoyment of high school - Expected probability of ever enrolling at a university - Track recommendation for hypothetical peers (boy/girl with similar grades) - Enrollment probability beliefs for girls vs. boys born in 2002 with similar elementary school grades as the respondent

Cited by

No citation data available for this registration yet.

Keywords