Research

Working Papers

How Exceptional Should Role Models Appear? A Field Experiment on Success Framing (submitted)

Abstract: Motivational interventions often feature exemplary individuals as role models, but the effects may depend not only on who is featured but also on how they are introduced. I examine whether different framings of the same role models' success influence their effects on academic performance and psychological strain. Using a field experiment with 1,920 Chinese students, I compare the same role models introduced using exceptional versus broader, more moderate descriptions of the same underlying academic achievements. The exceptional framing raises total scores by 0.091-0.096 standard deviations relative to the moderate framing. By contrast, relative to the exceptional framing, the moderate framing lowers the probabilities of reporting feeling depressed and experiencing academic pressure by 12.3 and 9.3 percentage points, respectively. Evidence on potential mechanisms suggests shifts in academic self-perceptions rather than study time, academic aspirations, or goal attainment. Findings suggest that role model interventions should align the portrayal of success with target outcomes.

Pdf; Pre-analysis plan; SSRN * Earlier versions of this paper circulated under the title "Inspirational or Demotivational: Experimental Evidence on the Impacts of Role Models."

The Power of Words: How Role Models Influence Belief and Behavior

with Sherry Wu

Abstract: Role models inspire and model behavior, yet prior interventions typically bundle the messages that serve these functions, making it difficult to identify which components drive their effects. I examine whether practical learning-strategy messages and anti-stereotype messages work separately or jointly in a field experiment with 2,719 middle school students in China. Treated classrooms were randomly assigned to see role models who discuss learning strategies, the same role models sharing anti-stereotyping perspectives, or them combining both. I find that combining strategies and perspectives improves first-year students' total scores by 0.04-0.06 s.d. and math scores by 0.07–0.10 s.d. Adding learning-strategy information to anti-stereotype messages significantly improves total exam scores relative to anti-stereotype messages alone. However, strategy information alone does not improve academic outcomes and increases reports of feeling depressed, indicating that omitting stereotype-overcoming information can be not just less effective but psychologically costly.

Pdf (new draft coming soon); Pre-analysis plan; SSRN

Peer personality, perceptions of gender inequality, and academic performance

Abstract: This paper studies how classroom peer personality shapes students’ perceptions of gender inequality in mathematics and academic performance. I exploit within-school-by-grade variation in peer openness and peer persistence. I find that peer openness reduces students’ belief that boys are better at mathematics than girls, whereas peer persistence raises total test scores. When both personality dimensions are considered jointly, peer openness lowers gender-biased beliefs but also reduces short-run total test performance, whereas peer persistence improves test performance and increases gender-biased beliefs. Mechanism analyses suggest that peer openness increases social safety and belonging, whereas peer persistence leads to higher perceived parental expectations, stronger educational aspirations, and greater relative-rank pressure. These findings indicate an equality-achievement trade-off arising from differences in classroom personality climates.

Pdf available upon request.

Selected Work in Progress

A large-scale field experiment on how group-targeted role models shape aspirations and actions

with Sherry Wu and Mike Mai

Human role models, AI, and ideology: A field experiment with Chinese factory workers [working title]

with Mike Mai and Sherry Wu

Managers’ use and disclosure of GenAI in employee performance feedback [working title]

with Chelsea Che and Hengchen Dai

Screen use, academic performance, and well-being: A large-scale field experiment in rural Chinese schools [working title]

with Ye Xu, Clair Yang, and Tony Ke

Permanent Working Paper

Mobile money and subjective well-being: Cross-country evidence

Page updated

Google Sites

Report abuse