Fabian T. Pfeffer

Effects of Social Mobility on Individual Well-Being, Attitudes, and Behavior: Full and Partial Identification

Research Project Description

Overview:

Social and behavioral scientists have long sought to estimate the effects of intergenerational socioeconomic mobility (the difference between an individual's social origin and destination) on a range of individual outcomes. More recently, there has also been widespread public speculation that the increasing rate of downward social mobility among working-class Americans is the primary cause of ongoing social and political upheaval. However, the empirical study of mobility effects faces a fundamental methodological challenge: Due to the linear dependency among social origins (O), destinations (D), and social mobility (M = D - O), researchers cannot use conventional statistical methods to estimate the unique contributions of the three variables to any given outcome. We propose a set of innovative approaches to estimate mobility effects that will also be readily applicable to other substantive applications in which linear dependency problems exist. First, we clarify what can be known from the data alone without further assumptions. Second, we introduce a new non-parametric bounding approach that partially identifies what we call the pure mobility effect. We apply these new approaches to investigate the effects of intergenerational social mobility on a range of measures of individuals' sociopsychological well-being, political attitudes, demographic outcomes, and health using data from the General Social Survey (GSS), the Panel Study of Income Dynamics (PSID), and its new Well-Being module (PSID-WB).



Intellectual Merit:

Unlike prior work, this project does not introduce implicit or ad hoc parametric assumptions to resolve the methodological challenge arising from linear dependency. Instead, we clarify what conclusions analysts can draw from their data alone and how theoretically derived and clearly stated assumptions can provide informative bounds on a central parameter of interest, here, the pure mobility effect. This project will not only provide new empirical evidence about the effects of downward mobility at a time when it has become more widespread, but will also result in new tools that can be employed by policy-makers to address these concerns, and by other researchers who can benefit from the provision of bounds on a central parameter of interest for a wide range of applications (from studies of social deviance to economic risk assessment).

Broader Impacts:

This project addresses fundamental concerns about the nation's prospect for social cohesion and economic health. By disentangling the impact of social mobility on individuals' well-being, attitudes, and behavior from that of individuals' current economic condition, the results of this project will allow policymakers to better target social policy aimed at alleviating social suffering and dysfunction, e.g. among racial minorities, such as African-Americans, who have traditionally been more heavily impacted by downward social mobility. As part of a broad dissemination strategy, this project will also develop a userfriendly R package as well as an open-access online tool that increases access and use of the proposed methodological tools by non-expert analysts and decision-makers. This tool will also make the existing data infrastructure on social mobility and individual well-being based on two large NSF-funded studies (GSS and PSID) more accessible for a wide audience across different academic disciplines, industry, and government. The empirical application of the approaches developed here will also offer important training opportunities for students.

Funding:
National Science Foundation
(1948310)

Funding Period: 8/1/2020 to 7/31/2023

PSC In The News

RSS Feed icon

Mehta makes it clear why young people are leading the rise of COVID cases in Michigan: Socializing

More News

Highlights

Frey's Social Science Data Analysis Network, SSDAN wins 2020 MERLOT Sociology Classics Award

Doing COVID-19 research? These data tools can help!

More Highlights


Connect with PSC follow PSC on Twitter Like PSC on Facebook