5.4: Explicit vs. Implicit Stereotypes

The cognitive processes described in the previous section operate largely outside of conscious awareness. We can even hold gender egalitarian beliefs and still process social information in biased ways. This division, and possible contradiction, between what happens automatically and what we consciously endorse is central to the study of implicit versus explicit stereotypes.

Explicit stereotypes are beliefs that a person is aware of and can report. If someone says, "women are too emotional to be leaders," that is an explicit gender stereotype. Explicit stereotypes can be measured through surveys and self-report, and people can choose to reject or endorse them. However, self-report is susceptible to social desirability bias, so these measures alone may not accurately assess inner belief systems. Implicit stereotypes are associations between groups and traits that operate outside of conscious awareness (Greenwald & Banaji, 1995). The most widely used tool for measuring implicit stereotypes is the Implicit Association Test (IAT), developed by Anthony Greenwald and colleagues (1998). The IAT measures the speed with which people pair concepts. For example, a gender-career IAT might ask participants to sort words related to "male" or "female" alongside words related to "career" or "family." If someone pairs "male" with "career" faster than they pair "female" with "career," the test registers an implicit association between men and professional roles. Large-scale data from Project Implicit (you can be a research participant here!) has collected millions of IAT responses. Their findings indicate that most participants, regardless of their own gender, implicitly associate men with career and science terms and women with family and arts terms (Nosek et al., 2002).

Critics of the IAT methodology point to variation in the test-retest reliability, meaning that a person's score can shift from one sitting to the next (Gawronski et al., 2017). Others question how well IAT scores can predict actual discriminatory behavior. The relationship between implicit bias as measured by the IAT and real-world outcomes is statistically significant but small (Oswald et al., 2013). This means that the IAT’s power to predict individual outcomes is limited. Those who support use of the IAT argue that implicit associations are best understood as reflecting cultural exposure to stereotypes rather than personal beliefs, and that the test is best for demonstrating patterns at the population level (Greenwald et al., 2009). While it is important to acknowledge criticisms, the IAT has been important in establishing that stereotyped associations can be held and may affect behavior outside of conscious awareness.

If implicit associations shape decisions (who gets called on in STEM classes, whether a trans woman is recognized as a woman, who seems "naturally" suited for leadership when making hiring decisions), they can produce discriminatory outcomes even in the absence of discriminatory intent (Moss-Racusin et al., 2012). The next section examines what those outcomes look like in more detail.

---