Guyll, M. & Madon, S. (2000). Ethnicity research and theoretical conservatism [Comment in response to Sue, 1999], American Psychologist, 55, 1509-1510.

 This paper is copywrited by APA. http://www.apa.org/journals/

Ethnicity Research and Theoretical Conservatism

Sue (December, 1999) levels a serious charge against the scientific community of psychology. He asserts that the enforcement of scientific criteria is selective and biased against ethnicity research. To remedy this bias Sue proposes that the field adopt an exceptionally strong formulation of the principle of empirical skepticism that would preclude theory generalization across ethnicity and require theory testing in all populations.

Though we agree with many of the points Sue makes, we feel compelled to delineate an alternative process whereby the evenhanded implementation of scientific standards could produce the appearance of bias against ethnicity research. We contend that an apparent bias against ethnicity research could stem from four factors: (1) the scientific community is theoretically conservative, (2) existing theory is often based on data from White Americans, (3) existing theory is often mute regarding ethnicity, causing its predictions to generalize across ethnicity, and (4) ethnicity research often hypothesizes ethnic differences, requiring a multi-ethnic sample to adequately test predictions. We discuss these factors in turn to describe how an apparent bias against ethnicity research could emerge from an impartial review process. We then consider negative consequences of adopting Sue’s extreme position regarding empirical skepticism. We conclude with an alternative to barricading theory along ethnic lines.

An Unbiased Process

The scientific community is theoretically conservative. Because novel theoretical claims based on new research often lack a strong empirical foundation, they are especially vulnerable to subsequent refutation by contradictory evidence, or by the discovery of an internal flaw. Consequently, there are risks associated with publishing research that proposes new theory. Research that introduces new theory, or challenges existing theory, must pass a highly critical review. Though this process may cause some sound research to be rejected, the ultimately beneficial effect is that the field wastes less time working with inferior theories because poor research is less likely to survive rigorous evaluation. Thus, theoretical conservatism produces a bias in favor of existing theory. New theory only gains acceptance if existing theory is shown to be inadequate and alternate explanations have been satisfactorily addressed.

Existing theory is often based on data from White Americans. Because most psychological research is conducted in America, most theory is based on data from White Americans (Sue, 1999). For this reason, theoretical conservatism, which favors existing theory, will usually favor a theory developed in predominately White samples. Should findings from ethnicity research challenge existing theory, theoretical conservatism will produce a bias against the ethnicity research and in favor of the existing theory. However, the bias against the ethnicity research occurs because its findings are novel, and not because of the ethnic composition of its sample.

Existing theory is often mute regarding ethnicity. When existing theory is mute regarding ethnicity, its scope implicitly includes people of all ethnicities. Hence, a subsequent study based on participants of another ethnicity that supports existing theory will be considered a replication, judged less important, and evaluated less favorably. Conversely, a subsequent study based on participants of another ethnicity that contradicts existing theory will be considered neither a falsification of existing theory, nor a clear demonstration of ethnic differences, because alternative explanations exist for obtaining different findings across studies (Campbell & Stanley, 1963). Thus, ethnicity research that challenges the generalization of existing theory to a particular ethnicity will face an uphill battle because theoretical conservatism will be biased in favor of existing theory over its entire scope, which typically includes all ethnic groups.

Ethnicity research often hypothesizes ethnic differences. When the goal of ethnicity research is to investigate or conclude ethnic differences, then the sample cannot consist of participants of only a single ethnicity because the research design will not permit a test of the study’s hypothesis. One might counter that an investigation using a single ethnicity sample does provide evidence for ethnic differences if its findings differ from those obtained in other studies that utilized participants of another ethnicity. However, this response is insufficient because, as noted above, other explanations exist for obtaining different findings across studies. Research that hypothesizes ethnic differences, but presents data from only a single ethnic group, suffers from a serious limitation and will be evaluated less favorably. Because of the limitation, theoretical conservatism will produce a bias against such research.

Negative Consequences of Extreme Empirical Skepticism

To reduce bias against ethnicity research Sue recommends that the field adopt an extreme position of empirical skepticism that would preclude theory generalization across ethnicity. We strongly disagree with Sue’s proposal for several reasons. First, his solution is unworkable because, as Sue himself acknowledges, it is an "impossible task" to test a theory in all populations (p. 1075). Second, precluding theory generalization will isolate researchers who seek to make general theoretical statements, but who only have access to large numbers of participants who are of a single ethnicity. Because many researchers face this situation, Sue’s proposal would encourage a Balkanization of psychological science. Third, Sue’s recommendation would require researchers to accept as a fundamental premise the idea that individuals from different ethnic groups differ from each other with respect to all psychological processes. This is less parsimonious and egalitarian than the assumption of theory generalization because the latter connotes an essential sameness among all human beings.

A Recommendation

The ultimate challenge of psychological science is the development of universal theory that can explain individual differences both within and across culture. Ethnicity research contributes most when it advances broad theories that apply to multiple ethnic groups. To this end, theories should include constructs that characterize multiple ethnic groups to varying degrees, and should not present the ethnic group itself as an explanation for differences. The goal ought to be broader, cross-cultural theory and unification of empirical findings—not ad hoc theory and fragmentation along ethnic lines.

References

Campbell, D. T., & Stanley, J. C. (1963). Experimental and quasi-experimental designs for research. Boston: Houghton Mifflin.

Sue, S. (1999). Science, ethnicity, and bias: Where have we gone wrong? American Psychologist, 54, 1070-1077.

Author Note

Max Guyll, Institute for Social and Behavioral Research; Stephanie Madon, Department of Psychology.

Correspondence concerning this comment should be addressed to Stephanie Madon, W112 Lagomarcino, Iowa State University, Ames, IA 50011-3180. Electronic mail may be sent via internet to madon@iastate.edu