Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/253574 
Year of Publication: 
2021
Citation: 
[Journal:] Quantitative Economics [ISSN:] 1759-7331 [Volume:] 12 [Issue:] 3 [Publisher:] The Econometric Society [Place:] New Haven, CT [Year:] 2021 [Pages:] 945-980
Publisher: 
The Econometric Society, New Haven, CT
Abstract: 
Bullying cannot be tolerated as a normal social behavior portraying a child's life. This paper quantifies its negative consequences allowing for the possibility that victims and nonvictims differ in unobservable characteristics. To this end, we introduce a factor analytic model for identifying treatment effects of bullying in which latent cognitive and noncognitive skills determine victimization and multiple outcomes. We use early test scores to identify the distribution of these skills. Individual-, classroom- and district-level variables are also accounted for. Applying our method to longitudinal data from South Korea, we first show that while noncognitive skills reduce the chances of being bullied during middle school, the probability of being victimized is greater in classrooms with relatively high concentration of boys, previously self-assessed bullies and students that come from violent families. We report bullying at age 15 has negative effects on physical and mental health outcomes at age 18. We also uncover heterogeneous effects by latent skills, from which we document positive effects on the take-up of risky behaviors and negative effects on schooling attainment. Our findings suggest that investing in noncognitive development should guide policy efforts intended to deter this problematic behavior.
Subjects: 
Bullying
cognitive and noncognitive skills
unobserved heterogeneity
JEL: 
C34
C38
I21
J24
Persistent Identifier of the first edition: 
Creative Commons License: 
cc-by-nc Logo
Document Type: 
Article

Files in This Item:
File
Size
720.75 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.