« Back to Results

Beliefs and Education

Paper Session

Tuesday, Jan. 5, 2021 10:00 AM - 12:00 PM (EST)

Hosted By: American Economic Association
  • Chair: Orazio Attanasio, Yale University

A (Dynamic) Investigation of Stereotypes, Belief-Updating, and Behavior

Katherine Coffman
,
Harvard Business School
Maria Paola Ugalde Araya
,
Arizona State University
Basit Zafar
,
University of Michigan

Abstract

To understand gender differences in educational and labor market outcomes, we study how the gender stereotype associated with the domain affects belief updating and subsequent behavior in a dynamic setting. Our hypothesis is that the stereotype associated with the task/field interacts with how individuals perceive noisy signals about performance in that domain (as well as in others), and this leads to systematic impacts on how beliefs are updated and subsequently what choices are made. For this purpose, we design an experiment where students first perform quizzes in two domains (one that is male-stereotypical, and one that is female-stereotypical), and report beliefs about their absolute and relative performance on both tests. They then receive noisy signals of their performance on the tests, and have to decide the compensation scheme for second-round tests in both domains. Importantly, the design generates quasi-random variation in the kind of feedback one receives conditional on actual ability. We also vary the gap between the two rounds, to understand the dynamics of belief updating.

Effects of Perceived Productivity on Study Effort: Evidence from a Field Experiment

Fulya Ersoy
,
Loyola Marymount University

Abstract

How does effort respond to the perceived relationship between effort and performance? To answer this question, I conduct a field experiment with a widely used online learning platform. I exogenously manipulate students' beliefs about the effort-performance relationship by assigning them to different information treatments, each of which provides factual information. I find that both information about an average individual and anecdotal information move students' beliefs towards the information provided. Furthermore, students change their study effort in the same direction with the shifts in their beliefs. This result suggests that altering students' beliefs about the effort-performance relationship influence their human capital accumulation.

How Do Expectations about Government Benefits Affect Human Capital Investment?

Manasi Deshpande
,
University of Chicago
Rebecca Dizon-Ross
,
University of Chicago

Abstract

Supplemental Security Income (SSI), administered by the Social Security Administration (SSA), pays $10 billion annually to the families of 1.3 million low-income, disabled children. SSI youth have poor life outcomes, including high dropout rates and arrest rates. When SSI children turn 18, they must re-qualify for the program as adults. Since the definition of disability is more stringent for adults than children, 70% of SSI children with certain mental and behavioral conditions are removed at age 18. However, qualitative evidence indicates that most households vastly underestimate the probability of removal. The hypothesis underlying this project is that a lack of information about the likelihood of removal contributes to the poor adult outcomes of SSI children. Families may believe that their children's benefits will continue in adulthood and underinvest in education. This project conducts an RCT that provides the families of SSI children with information on the likelihood of removal from SSI at age 18. We will investigate the effect of the information on short-term investments in human capital and long-term child outcomes -- including education, employment, and crime -- through a combination of survey and linked administrative data. In addition to investigating a promising policy, this project will speak to the broader question of how expectations about the availability of future government benefits affect current human capital investments, which is relevant to the design of all tax-and-transfer programs.

Beliefs, Preferences, and Student Effort

William Murdock III
,
Harvard University
John J. Conlon
,
Harvard University
Spencer Y. Kwon
,
Harvard University
Dev Patel
,
Harvard University

Abstract

Academic achievement gaps across race, gender, and socio-economic status are strongly correlated with corresponding differences in student studying behavior. We examine the underlying determinants of this studying effort by eliciting the relevant preferences and beliefs for several thousand U.S. high school students in a novel survey. Equipped with these individual-level utility curves and studying-to-grade production functions, we flexibly solve for each pupil’s expected study behavior. The predictions based on students’ beliefs and preferences are strongly linked to their reported study habits. Counterfactual simulations show that differential preferences for receiving higher grades are the primary driver of the achievement gaps rather than variation in willingness to study or beliefs about the efficacy of studying. We also find that pay-for-grades incentives have muted predicted effects, as have been found experimentally, because students have low beliefs about the efficacy of studying in producing better grades.
Discussant(s)
Michael Thaler
,
Princeton University
Arnaud Maurel
,
Duke University
Heather Sarsons
,
University of Chicago
Joseph Altonji
,
Yale University
JEL Classifications
  • I2 - Education and Research Institutions
  • D8 - Information, Knowledge, and Uncertainty