The effect of relative performance feedback on academic outcomes
How to cite
Schwarz, Anna; Sausgruber, Rupert; Lackner, Mario; Winter-Ebmer, Rudolf. 2022. "The effect of relative performance feedback on academic outcomes." AEA RCT Registry. AEARCTR-0008977. https://doi.org/10.1257/rct.8977-1.0
Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.
* Primary investigator
Abstract
Despite extensive research on the effect of relative performance feedback, a consensus on the direction and mechanics of the effect, especially in the higher education context, is far. Moreover, the role of students' prior beliefs about their performance is poorly understood. Using a large-scale survey at a big European university, we can cleanly elicit these beliefs before providing relative performance feedback to a randomly selected treatment group. By linking the survey data to administrative data, we can subsequently measure the effect on academic performance in the short- and long-run and by prior beliefs. Additionally, the treatment group has to actively decide to see the information, which allows us to analyze selection into receiving relative performance feedback. The survey context further enables us to estimate the effect of relative performance feedback on several secondary outcomes, such as competitiveness, stress, academic self-concept. Our clean experimental design using survey as well as administrative data lends itself to comprehensively analyze the effect of relative performance feedback in higher education and provide evidence on the role of prior performance beliefs.
Intervention
The intervention is placed within a university-survey, that is sent to all bachelor beginners at the end of their first semester. The treatment consists of providing one half of students with the opportunity to be informed about their position in the GPA (Grade Point Average) as well as the ECTS (European Credit Transfer and Accumulation System) distribution. It is thus an intention-to-treat design. We implement this by telling students the quartile of the distribution in which their performance is located. The reference group consists of all students who started their studies in the same semester and for the same field of study. The decision whether students choose to see their relative performance will be recorded. This enables us to analyze, who selects into receiving information. Allowing students to opt into receiving performance feedback may also attenuate negative effects, that are documented in the literature.
Experimental design
The students will be randomized into treatment and control group and see the following items in the official university-survey:
Survey items to be answered before the intervention:
- Competitive preferences and locus of control
- Questions on how much students interact with fellow students
- Prior beliefs about relative performance, separated by ECTS and GPA
Treatment group: Question, whether they want to see their actual quartile for ECTS and GPA, respectively. If yes, they see the treatment. If no, they don't.
Control group: directly continues to next block of questions.
Survey items to be answered after the intervention:
- Satisfaction with the studies
- Study effort in the past and upcoming semester
- Stress and perceived competitiveness in the studies
- Academic self-concept
- Own and perceived prosociality in the studies
Design details
Cited by
No citation data available for this registration yet.