Perceived Signaling Effects in Higher Education: Evidence from France

Last registered on March 21, 2025

Pre-Trial

Trial Information

General Information

Title
Perceived Signaling Effects in Higher Education: Evidence from France
RCT ID
AEARCTR-0015557
Initial registration date
March 17, 2025

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
March 21, 2025, 11:04 AM EDT

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

Locations

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Primary Investigator

Affiliation
VATT Institute for Economic Research

Other Primary Investigator(s)

PI Affiliation
DIW Berlin

Additional Trial Information

Status
In development
Start date
2025-03-17
End date
2026-12-31
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
We designed a field experiment to test predictions of a signaling model in the context of academic distinctions at France's largest public university. Specifically, we examine students' misperceptions about how employers reward academic honors in the first year after graduation. Our experimental design informs students about the differential first-year earnings caused by academic distinctions, based on findings from Astruc-Le Souder, Bargain, and Locks (2024). Our key outcomes include, in the short run, students' intentions such as willingness to exert effort (e.g., study hours), aspirations for academic distinctions, and intentions to highlight honors on their CVs or public profiles like LinkedIn. In the medium and long run, we track actual academic performance indicators from academic records (e.g., final GPA, honors attainment) as well as job-seeking behavior and early labor market outcomes through follow-up surveys conducted post-graduation.

Our model predicts that students will respond more strongly the closer they are to the honors threshold, the more efficiently they convert study hours into GPA gains, and the more highly their prospective employers value academic distinctions. We will test our empirical findings against these predictions.



External Link(s)

Registration Citation

Citation
Giaccobasso, Matias and Gedeao Locks. 2025. "Perceived Signaling Effects in Higher Education: Evidence from France." AEA RCT Registry. March 21. https://doi.org/10.1257/rct.15557-1.0
Experimental Details

Interventions

Intervention(s)
We study first- and second-year master's students at France's largest public university as they form expectations about their academic and early career outcomes. We invite 14,595 students to participate in a survey on study habits and perceptions of the labor market. Within the survey, we conduct an information-provision experiment. We elicit beliefs on, and provide information about, the differential first-year earnings caused by academic distinctions (Astruc-Le Souder, Bargain, and Locks; 2024).

The goal is to measure how the information provided affects students' perceptions of the returns to academic distinctions, their study habits, such as effort and time allocation, and their short-run academic and career intentions. In the medium and long run, we also examine their academic performance and job market behavior.

Intervention Start Date
2025-03-17
Intervention End Date
2025-03-22

Primary Outcomes

Primary Outcomes (end points)
Our main outcomes of interest are:

1) Academic performance indicators: GPA (continuous) and honors attainment (binary, and for each threshold). These will be based on academic records provided by the University

2) Job seeking behavior: number of applications sent, type of jobs to which someone applied

3) Early labor market outcomes: employment status (binary), start date (continuous), entry wage (continuous), current wage (continuous), type of job (discrete).

#2 and #3 will be based on follow-up surveys.

Some additional notes:

1) We will test for selective attrition to the follow-up survey
2) We will do bounding exercises
3) We will do weighting exercises
4) Our baseline specification will include winsoring for heavy-tailed variables, such as wages. We will report results with non-winsored variables, as well as for different levels of winsoring for robustness and transparency purposes.
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Right after the information-provision, the survey includes a number of questions designed to be used as outcome. This data allows for a series of secondary outcome that indicate students' intentions:

- Intention to exert effort (i.e. "How much effort do you plan to put into increasing your average this semester?")
- Aspirations for academic "Bien" distinctions (i.e. "Do you aim to achieve a distinction at the end of your degree? If so, which one?")
- Intentions to report honors on CVs or social media profiles (i.e., "If you receive a "Bien" honors in your Master's degree, how likely are you to include it on your CV or LinkedIn profile for job applications?")

Notes:

1) Our baseline specification will include a binary definition of these variables, but we will report figures with the full distribution by group, for transparency purposes.
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
We emailed 14,595 students at France's largest public university to participate in an online survey on study habits and perceptions of the labor market. Importantly, the survey was conducted in collaboration with the university, and therefore we are able to link students' survey responses with their academic records.

Within the survey, we conducted an information-provision experiment. We elicited beliefs on, and provide information about, the differential first-year earnings caused by academic distinctions, based on some of our previous work:

- Bargain, O., Astruc Le Souder, M., & Locks, G. (2024). A Question of Honor? The Labor Market Advantage of Academic Signaling. The Labor Market Advantage of Academic Signaling.

We want to measure how the information provided inside the survey affected the subsequent attitudes toward study effort, self-promotion, academic performance, and early career labor market outcomes. The main focus of our information-provision experiment is not to investigate if providing information (relative to not providing information) has an effect on the average effort exerted, and the probability of obtaining honors, just to name a few. When provided with the information about returns to academic mentions, some individuals may update their beliefs down and others may update the beliefs up, so those effects may cancel each other out. Instead, our main focus is to measure the causal effects on beliefs, by exploiting how individuals update relative to their prior beliefs. With that goal in mind, we will use econometric models used in some of our previous work such as

- Giaccobasso, M., Nathan, B., Perez-Truglia, R., & Zentner, A. Where Do My Tax Dollars Go? Tax Morale Effects of Perceived Government Spending. American Economic Journal: Applied Economics.

In any case, for transparency, we will also report the average effects on receiving the information and we will compare it the the causal effects on beliefs.

In the survey, we elicit beliefs on, and provide information about labor market benefits (wage premium) associated with achieving high academic honors. In particular, some individuals were selected to receive the following piece of information:

"A recent academic study of former master's students at [university_name] found that earning a "Bien" (Highest Honors) is associated with a 12.5% ​​increase in cumulative earnings during the first year after graduation. This equates to approximately 1.5 additional salaries during that first year."

The hypotheses are the following:

1) Students are more willing to put additional effort when they find out that returns to obtaining academic honors are larger.

2) However, effects are heterogeneous depending on expected returns to additional effort. In particular, a signaling model predicts that students will respond more strongly the closer they are to the honors threshold, the more efficiently they convert study hours into GPA gains, and the more highly their prospective employers value academic distinctions.

Hence, our analysis will dig in into three key sources of heterogeneity:

a) GPA distance to "Bien" threshold
b) Students' perceived marginal returns to effort (i.e., marginal gains in GPA per additional hour of study)
c) How likely is that their expected prospective employers reward honors. More specifically, we will test for differential effects depending on the expected margin for negotiation in wages from prospective employers, and by major, e.g., econ/biz/mgmt. versus other majors (see Bargain, Astruc Le Souder, & Locks; 2024 for more details).

We are able to explore this heterogeneity based on a series of questions asked pre-treatment in the information-experiment survey. In particular, we ask students how likely they think they will be able to negotiate their wages in the job-seeking process and how their own GPA would change if they dedicate additional hours to study.

In addition, we also plan to explore some secondary forms of heterogeneity:

a) By gender
b) By risk-preference and self-confidence
c) Depending on how strong students value financial compensation when deciding to accept a job offer
d) For students in 2-year masters, we will explore if there are differences depending on whether they are in the 1st or 2nd year
e) We will also explore whether the information treatment "backfires" for those who are above the threshold, similar to what's documented in Robinson et al (2021)

Robinson, Carly D., et al. "The demotivating effect (and unintended message) of awards." Organizational Behavior and Human Decision Processes 163 (2021): 51-64.

Implementation details:

- The invitations were sent on 03-17-2025, 11:20am
- The pre-registration was closed a few minutes after, but researchers did not access the survey data until the survey was closed, on 03-22-2025, 1:59pm.
Experimental Design Details
Not available
Randomization Method
Randomization done in office by a computer: Conditional on accessing the survey, the survey platform (Limesurvey) will select respondents into treatment and control groups with 50% probability.
Randomization Unit
Student
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
14595 students
Sample size: planned number of observations
We conducted a pilot survey to test response rate with Undergraduate students. Out of 600 invitations, 137 clicked the link, but only 99 answered the key questions. Hence, based on an actual response rate of 1/6, we expect about 2432 responses.
Sample size (or number of clusters) by treatment arms
We assigned students to treatment and control groups based on a 50% probability. Hence, we expect 1216 students in the treatment group and 1216 students in the control group.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Since we do not know the response rate yet, we cannot estimate the minimum detectable effect size in advance. We will provide ex-post estimates on this, once the data collection is finished.
IRB

Institutional Review Boards (IRBs)

IRB Name
IRB Approval Date
IRB Approval Number