3.4 Reducing Sex/Gender Bias in Research

Self-Reflection

Feminist researchers reject the myth of the impartial researcher and believe that all researchers approach their projects with assumptions, beliefs, and biases. One source of bias may be a person’s positionality. In other words, the person’s specific position and power in society, influenced by multiple aspects of social identity, will influence what they deem important to study and how they approach their research. For example, a young unmarried psychologist who has just moved in with their heterosexual partner may become interested in studying the division of labor in the home. Later, if that psychologist becomes a parent, they may become interested in studying work policies and their influence on work/life balance. A researcher who has a socially marginalized identity may be particularly interested in investigating the lived experience of people who are often neglected in research (Settles et al., 2020).

More information

A photo shows a woman at a table reflecting. She is shown seated with her chin resting on her hands and looking up and off to the side. Many thin squiggly arrows are depicted above her head in a tangled mess and pointing in all directions.

Everyone is influenced by their positionality, but feminist psychologists are generally more open about it. They often engage in a process of reflexivity, in which they examine and report their own perspectives and beliefs, asking “What is the research process, and how am I influencing it?” (Lazard & McAvoy, 2020, p. 12). Reflexivity can also help researchers acknowledge how their theoretical perspectives influence their execution of the research as well as the analysis of their data (Figure 3.7).

Reflexivity is relatively standard in qualitative research and is considered a core feature of feminist qualitative research (Wilkinson, 1988). However, it’s much less common in quantitative research and almost unheard of in nonfeminist quantitative research, which is more likely to reflect the myth of the impartial scientific observer who isn’t influenced by personal perspectives. As a result, feminist research is sometimes considered more subjective and personal, even though all researchers have assumptions and perspectives that influence their work. However, as we saw in Chapter 2, people from dominant social groups or who hold a dominant perspective may be unaware of their biases or see their own perspectives as the norm and therefore not worthy of mention. Ultimately, all research is influenced by people’s beliefs and perspectives; feminists are just more honest about it! In Table 3.1, we provide some guidelines for how researchers and consumers of research can consider the influence of bias throughout different phases of the research process.

Generalizability and Context

Why is it important to pay attention to the context of a study?

As we’ve discussed, the way variables are operationalized, how questions are asked, and what situations participants are in during research studies can all influence the data and make sex/gender differences appear greater than they actually are. For example, research that takes place in one setting may not generalize to other settings. Generalizability assumes that humans share core psychological processes, a concept known as universality (Matsick, Kruk, Oswald, & Palmer, 2021). However, feminist intersectional researchers reject assumptions of universality and instead consider historical, social, and political factors (Matsick, Kruk, Oswald, & Palmer, 2021); as a result, they often report contextual factors such as demographics and political climate. For example, they might acknowledge that they collected data during a spike in activity related to #MeToo or that they wrote their research article during a contentious presidential election campaign (Gervais et al., 2021). They also ask reflective questions to understand how context uniquely influences data and outcomes—for example, “How are participants situated in structural contexts, and what role may inequality and power differentials play in their lives that can inform the interaction with this research?” (Matsick, Kruk, Oswald, & Palmer, 2021, p. 415). By asking such questions, they avoid oversimplification and misinterpretation, which also reduces the likelihood that data based on one particular group at one particular time might be used as justification for policies and legislation that affect all people (Matsick, Kruk, Oswald, & Palmer, 2021).

When you ask a question in one way, you may get different results than if you asked it in another way. As a critical consumer of psychological research, you need to consider whether the methods or setting of a study may have influenced the results. Doing so will help you avoid making universal or overly simplistic generalizations about research outcomes and about any sex/gender differences in particular. As we talk about research on sex/gender differences in the next section, you’ll see how the context of a study (e.g., the setting, how the questions are asked) can profoundly affect the results. Further, it’s important to remember that reporting limitations or weaknesses doesn’t discredit science but instead makes it more accurate (Matsik et al., 2021). Allowing for more truthful reporting of results reduces the chance of overgeneralizing findings and helps increase transparency for consumers of scientific research (Matsik et al., 2021).

Representation and Intersectionality

Why is it important to pay attention to who the participants are and how research can be generalized?

As noted above, an overreliance on WEIRD samples can affect the choices made throughout the research process (Rad et al., 2018). Furthermore, even if the research question is targeting Western women, researchers may assume that the experiences of cisgender, white, middle-class women are representative of all Western women (Syed & Kathawalla, 2020). Such assumptions are likely to occur in university settings, where researchers probably have demographic characteristics similar to those of their study participants. A simple tenet that feminist researchers can follow is to “look for what (or who) is left out” (Matsick, Kruk, Oswald, & Palmer, 2021, p. 418).

Furthermore, people (and contexts) are layered and dynamic. Focusing on one variable, such as sex/gender, while trying to control for all other variables, won’t capture this complexity. Throughout this text, we’ve emphasized the importance of an intersectional approach. However, it’s very hard to do truly intersectional research because that would require measuring and analyzing a large number of variables (e.g., race, class, gender, age), and if quantitative analysis is done, it would take an enormous sample to find people in all the categories. For this reason, feminist researchers advocate for interdisciplinary collaboration, particularly with disciplines such as the humanities, where an understanding of subjective factors could enrich quantitative modes of knowing and lead to a more accurate reflection of lived realities (Matsick, Kruk, Oswald, & Palmer, 2021).

High-quality intersectional research is extremely rare. It not only has to be representative of people with a variety of social identities, it must take into consideration how power dynamics and social structures influence participants’ lives. A content analysis of articles published in two prominent counseling psychology journals found that a minority of published articles (less than 1%) could be considered intersectional and only a small subset of those were classified as strong or transformative (Shin et al., 2017). Notably, research that incorporates intersectional perspectives tends to be devalued in mainstream academic settings and seen as unimportant or lacking in rigor (Settles et al., 2020). People with socially marginalized identities (often across multiple dimensions of identity) are often the ones who do intersectional research, and multiple factors contribute to the devaluation of their work. These include their socially marginalized identities, the fact that the topics they research are often not mainstream (e.g., poverty, victimization), and their frequent use of qualitative methods. The fact that such research often acknowledges researchers’ positionality contributes to the perception that it is less scientific. However, as discussed above, all research is done from a social position—acknowledging it is simply being honest!