A New Strategy for Reducing Selection Bias in Nonexperimental Evaluations, and the Case of How Public Assistance Receipt Affects Charitable Giving

Abstract

Prior research considers the extent to which public assistance recipients' charitable activity differs from the habits of the general population. Although receiving public assistance is negatively associated with donating money, the relationship to volunteering is unclear. In response to challenges overcoming selection bias, we conducted a multivariate cluster-based subgroup analysis to reduce bias in our claims about the ways in which public assistance receipt affects charitable activity. This innovative approach to dealing with the problem of selection bias has implications and applications across the social sciences.

Evaluating the effect of public R&D subsidies on Italian firms' innovation activities: a subgroup analysis of 2008 CIS data

“…The subgroup analysis here applied is an atheoretical approach that minimises bias from selection on observables and analytically clusters cases as similar (Camillo and D'Attoma, 2012;Peck et al, 2012;D'Attoma and Camillo, 2011). As reported in Peck et al (2012), first it uses cluster analysis to increase the possibility of finding local spaces in which covariates are more likely balanced.…”

“…The subgroup analysis here applied is an atheoretical approach that minimises bias from selection on observables and analytically clusters cases as similar (Camillo and D'Attoma, 2012;Peck et al, 2012;D'Attoma and Camillo, 2011). As reported in Peck et al (2012), first it uses cluster analysis to increase the possibility of finding local spaces in which covariates are more likely balanced. Second, it uses a measure of global imbalance (GI) in the data and its related imbalance test, based on the concept of inertia 18 as a measure of dependence between a set of pre-treatment covariates and the treatment assignment indicator, that allows an overall assessment of how well matched cases are.…”

“…But, ATEs often obscure the changing effects within heterogeneous populations. Particularly if subgroup effects operate in opposite directions, the overall ATE will be estimated as zero (Peck et al 2012). To the best of our knowledge, only few recent studies on quantitative evaluation of R&D policies consider the possibility of differential impacts.…”

How Does Public Assistance Use Affect Charitable Activity? A Tale of Two Methods

“…This arena has been little studied, with our recent work (Guo & Peck, 2009;Peck, D'Attoma, Camillo, & Guo, 2012) and that of Brooks (2004Brooks ( , 2002 being the most direct tests of these policy effects. This arena has been little studied, with our recent work (Guo & Peck, 2009;Peck, D'Attoma, Camillo, & Guo, 2012) and that of Brooks (2004Brooks ( , 2002 being the most direct tests of these policy effects.…”

“…This arena has been little studied, with our recent work (Guo & Peck, 2009;Peck, D'Attoma, Camillo, & Guo, 2012) and that of Brooks (2004Brooks ( , 2002 being the most direct tests of these policy effects. While Brooks' research considers charitable giving only in terms of money, our work considers the donation of both money and time and concludes that receiving public assistance, all else equal, may suppress money donations but may actually increase one's donations of time for charitable purposes (Guo & Peck, 2009), at least among some subsets of this heterogeneous population (Peck et al, 2012). While Brooks' research considers charitable giving only in terms of money, our work considers the donation of both money and time and concludes that receiving public assistance, all else equal, may suppress money donations but may actually increase one's donations of time for charitable purposes (Guo & Peck, 2009), at least among some subsets of this heterogeneous population (Peck et al, 2012).…”