Research, Gender Bias in

Gender bias in research is as old as science itself. Although gender bias is less pervasive in research in the 21st century than it was in the past, it persists in some scientific fields. In the context of social and medical research, the term bias refers to a systematic error that can occur at any level or stage of empirical investigation, which is caused and goes undetected by the researcher. Gender bias in research occurs because researchers' stereotypes and prejudices about gender become implicitly, and hence unknowingly, but systematically infused with the research process. Such biases typically support the unfair preferential treatment of masculine characteristics (enhances men) and unfair negative treatment of feminine characteristics (derogates women). Gender bias in research is therefore undesirable and to be avoided.

The kind of prejudice that has existed in research since the beginning of science is known as androcen-trism. The andocentric view holds men and masculine characteristics as the norm and as the standard against which all people are compared. From an androcentric view of the world, women who do not act in accord with the beliefs, attitudes, values, and behaviors predicted from androcentric theories and researchers are judged to be incorrect, aberrant, inferior, and so on. Androcentrism in science and research is in part a byproduct of the male majority that exists in some realms of science. With mostly men occupying nearly every domain of science until relatively recently, the androcentric bias permeated theories and interpretation of data, and hence gender bias in research was a result of mostly men and too few women conducting research.

A brief review of the basic elements of the research method and the scientific process helps to clarify the creation and persistence of gender bias in research. The scientific method is by itself a useful tool that helps scientists to study nature and reality more closely than the human senses and imagination allow on their own. While the scientific method itself aims to be bias-free, it is not impervious to errors in human judgment. At each stage of the scientific method, the [Page 714]scientist is personally challenged with making decisions that are ultimately educated guesses about reality and, more specifically, about human life.

The scientific method and research process typically proceeds through the following six stages: (1) formulation of a research hypothesis that is informed by a theory; (2) development of a research design suitable to address to test the hypothesis; (3) selection of tests, measures, and procedures that are relevant to testing the hypothesis; (4) sampling and data collection; (5) selection and implementation of statistical procedures to test the hypothesis; and (6) interpretation and reporting of the results. At each of these stages, the scientist's own subjectivity, personal biases, and prejudices may become entangled with the research. Thus, without careful scrutiny and peer review, any research product that involves human data is vulnerable to gender bias.

For example, at the first stage of research, the conceptual level before data collection begins, the researcher is responsible for reviewing literature relevant to the research question. While reviewing the literature, the researcher must choose which theory or theories will inform the hypotheses to be tested. When the predictions to be tested involve individuals, the researcher must consider the extent to which gender differences are relevant to the scope and aims of the study. If gender differences could be an issue, the researcher must decide which theories will accommodate hypotheses that involve gender. Regardless of the researcher's personal opinions or beliefs about gender, the theory or theories will guide how the researcher formulates a hypothesis. However, the researcher is free to choose any theory that he or she deems relevant. There are no rules for how to review literature or how to select a theory; this is by itself not necessarily a problem. The problem related to gender bias is such that some theories are more prone to gender bias than others. For instance, at least two different types of gender-related biases, known as alpha bias and beta bias, may result at the first stage of research.

...