Are Connections the Way to Get Ahead? - Social Capital, Student Achievement, Friendships, and Social Mobility - Program on Education Policy and Governance Working Papers Series - PEPG 23-01

Oct 15, 2026

Uploaded by Sarah

Chetty et al. (2022) argue that county density of cross-class friendships—adult-bridging capital—causally shapes social mobility in the United States. This paper finds mobility depends on county density of family capital, community capital, and mean student achievement in grades 3–8. Using comparable multiple-regression models, it adds state fixed effects and broader bridging capitals, showing adult-bridging matters: mobility remains linked to achievement, family, and community but not school-bridging or political capital. Policy implications emphasize strengthening disadvantaged lives at home, school, and in communities.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages61

Reading Time2 h 34 min

AccessFree to download

Abstract

Social capital, friendships, and mobility (background)

Research approach and results

Bridging capitals and model comparisons

Policy implications

Program on Education Policy and Governance Working Papers Series Are Connections the Way to Get Ahead? Social Capital, Student Achievement, Friendships, and Social Mobility Paul E. Peterson, Angela K. Dills and M. Danish Shakeel 1 PEPG 23-01 Harvard Kennedy School 79 JFK Street, Taubman 304 Cambridge, MA 02138 Tel: 617-495-7976 Fax: 617-496-4428 [www.hks.harvard.edu/pepg/](www.hks.harvard.edu/pepg/) 1 Peterson: Harvard University; Dills: Western Carolina University; Shakeel: The University of Buckingham, UK SOCIAL CAPITAL AND SOCIAL MOBILITY Are Connections the Way to Get Ahead? Social Capital, Student Achievement, Friendships, and Social Mobility Paul E. Peterson ([ppeterso@gov.harvard.edu](ppeterso@gov.harvard.edu))* Program on Education Policy and Governance, Department of Government Harvard University, USA [https://orcid.org/0000-0001-8073-5186](https://orcid.org/0000-0001-8073-5186) Angela K. Dills ([akdills@email.wcu.edu](akdills@email.wcu.edu)) School of Economics, Management, and Project Management, Western Carolina University, USA [https://orcid.org/0000-0001-9527-9902](https://orcid.org/0000-0001-9527-9902) & M. Danish Shakeel ([danish.shakeel@buckingham.ac.uk](danish.shakeel@buckingham.ac.uk)) E. G. West Centre for Education Policy, The University of Buckingham, UK [https://orcid.org/0000-0002-7700-8540](https://orcid.org/0000-0002-7700-8540) May 19, 2023 Word count: 13,424 *Corresponding Author (Phone: +1-617-495-7976) . A prior version of this paper was presented before the 80th Annual Midwest Political Science Association Conference, April 13-16, 2023, Chicago. David Campbell, David Sikkink, Martin West, Raj Chetty, Scott Winship, and William Howell commented on earlier drafts ofthis paper. Michael Poor and Antonio Wendland provided administrative assistance. The Program on Education Policy and Governance at Harvard University supported the project through Walton Family Foundation grant \#00107915 . We own any remaining errors. SOCIAL CAPITAL AND SOCIAL MOBILITY Disclosure statement No potential conflict of interest was reported by the authors. Author Bios Paul E. Peterson is the Henry Lee Shattuck Professor of Government and Director of the Program on Education Policy and Governance at Harvard University, and a Senior Fellow at the Hoover Institution at Stanford University. Angela K. Dills is Professor of Economics and the Gimelstob-Landry Distinguished Professor of Regional Economic Development at Western Carolina University. M. Danish Shakeel is a professor and director of the E. G. West Centre for Education Policy atthe University of Buckingham, UK. SOCIAL CAPITAL AND SOCIAL MOBILITY Abstract Chetty et al. (2022) say county density of cross-class friendships (referred to here as“adult-bridging capital”) has causal impacts on social mobility within the United States. We instead find that social mobility rates are a function of county density of family capital (higher marriage rates and two-person households), community capital (community organizations, religious congregations, and volunteering), and mean student achievement in grades 3-8. Our models use similar multiple regression equations and the same variables employed by Chetty et al. but also include state fixed effects, student achievement, and family, community, schoolbridging (cross-class high school friendships), and political (participation and institutional trust) capital. School-bridging capital is weakly correlated with mobility if adult-bridging is excluded from the model. R-squared barely changes when adult-bridging is incorporated into the model. When it is included, mobility continues to be significantly correlated with the achievement, family, and community variables but not with school-bridging and political ones. We infer that county mobility rates are largely shaped by parental presence, community life, and student achievement. To enhance mobility, public policy need

FAQs

What does the paper say about the role of adult-bridging capital in social mobility?

The study finds mobility remains significantly correlated with achievement, family, and community variables when adult-bridging is included. It also reports that school-bridging is weakly correlated with mobility if adult-bridging is excluded and that R-squared barely changes when adult-bridging is incorporated.

Which factors does the paper identify as shaping county mobility rates?

County mobility rates are inferred to be largely shaped by parental presence, community life, and student achievement.

What does the paper recommend for improving social mobility through public policy?

Public policy should enhance the lives of disadvantaged people at home, in school, and in communities, rather than focusing only on the social class of their friendships.

Oct 15, 2026

Uploaded by Sarah