In lieu of an abstract, here is a brief excerpt of the content:

field experiments for studies of discrimination devah pager Experimental methods provide a powerful means of isolating causal mechanisms . Traditional experiments typically begin with clearly de‹ned “treatment ” and “control” conditions, to which subjects are randomly assigned.1 All other environmental in›uences are carefully controlled. A speci‹c outcome variable is then recorded to test for differences between groups. Frequently , in an effort to ensure a naive or “natural” reaction to the experimental condition, subjects are not told the purpose of the experiment. Experiments are often considered to be the “gold standard” for studies of causal effects precisely because of their ability to carefully isolate the effect of one variable on another. Because so much social science research is dedicated to developing causal arguments, experimental techniques clearly offer a valuable approach. At the same time, experimental methods have been criticized for failing to provide realistic conditions in which to study important social relationships . Many experiments are conducted in laboratory settings, far removed from the regular interactions and environments of everyday life. Experiments also typically rely on undergraduate psychology students as their subject pool, thus further limiting the generalizability of the ‹ndings to a broader population. To the extent that experiments compromise external validity (generalizability to the wider world) for enhanced internal validity (precise estimation of causal effects), the trade-offs entailed by using experimental methods must be carefully considered. 38 This chapter on ‹eld experiments offers one solution to the problems of generalizability in experimental methods. Field experiments blend experimental methods with ‹eld-based research, relaxing certain controls over environmental in›uences in order to better simulate real-world interactions . While retaining the key experimental features of matching and random assignment that are important for inferences of causality, this approach relies on real contexts (actual employment searches, real estate markets, consumer transactions, etc.) for its staged measurement techniques . For example, rather than asking undergraduate subjects to rate hypothetical job applicants in a lab experiment, a ‹eld experiment would present two equally quali‹ed job applicants to real employers in the context of real job searches. In what follows, I consider the use of ‹eld experiments in one speci‹c domain: the study of labor market discrimination. This case makes it possible to highlight the unique features of ‹eld experiments while also considering the potential complexities and pitfalls of this particular methodological approach. Field experiments offer unique advantages in isolating the causal effect of discrimination within the context of real-life settings. At the same time, the logistical requirements of ‹eld experiments—particularly in studies of hiring behavior—are often quite taxing, and, for that reason, it remains an uncommon method of social science research. In the following discussion, I describe the methodology of ‹eld experiments, discuss their strengths and limitations, and relate several ‹rsthand experiences that illustrate the complexities and frustrations involved in original data collection of this kind. the use of field experiments for studies of discrimination Discrimination has long been a fascinating and frustrating subject for social scientists. It is fascinating because it is thought to be a powerful mechanism that underlies many historical and contemporary patterns of inequality ; it is frustrating because it is elusive and dif‹cult to measure. Over a century of social science interest in the question of discrimination has resulted in numerous techniques intended to isolate and identify its presence, and to document its effects. Field experiments offer a unique approach to the study of discrimination because they provide carefully controlled comparisons of treatment on the basis of selected characteristics (e.g., race, gender, age) within the context of real-life decision making. Field experiments designed speci‹cally for the measurement of disField Experiments for Studies of Discrimination • 39 crimination are typically referred to as audit studies. The audit methodology was ‹rst pioneered in the 1970s with a series of audits conducted by the Department of Housing and Urban Development to test for racial discrimination in real estate markets (Yinger 1995; Wienk et al. 1979; Hakken 1979). The approach has since been applied to numerous settings, including mortgage applications, negotiations at a car dealership, and hailing a taxi (Turner and Skidmore 1999; Ayres and Siegelman 1995; Ridley, Bayton , and Outtz 1989; Yinger 1995; Massey and Lundy 2001; Cross et al. 1990; Turner, Fix, and Struyk 1991; Bendick, Jackson, and Reinoso 1994; Neumark 1996).2 In the case of employment discrimination, two main types of audit studies offer useful approaches: correspondence tests and inperson audits. correspondence tests The correspondence test approach, so named for...


Additional Information

Related ISBN
MARC Record
Launched on MUSE
Open Access
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.