Existing Initiatives to Reduce Discrimination


A variety of initiatives have been proposed to reduce discrimination and prejudice (and thus racism, sexism, homophobia and negative affect toward other marginalized groups). For a summary see the 2017 chapter by Murrar, Gavac, and Brauer (see also Schellhaas & Dovidio, 2016; chapters 30-35 of Dovidio, Hewstone, Glick, & Esses, 2010).

  • Intergroup contact: The goal is to foster contact between members of different social groups, preferably around a common goal or an activity that both groups enjoy (Pettygrew and Tropp, 2006)
  • Indirect intergroup contact: see Murrar and Brauer (2019) paper
  • Social categorization: The goal is to make salient that members of different groups belong to the same superordinate category or that they share a common group membership on some other demographic dimension. (Gaertner & Dovidio, 2000)
  • Perceived heterogeneity: The goal is to make salient that members of a particular outgroup are relatively variable with regard to their opinions, character traits, and hobbies.
  • Group malleability: The goal is to make salient that groups do not have a fixed mentality, but have characteristics that are malleable and can change over time (Halperin et al., 2011).
  • Consciousness raising: The goal is to make people aware of implicit biases and inconsistencies in their reasoning. Often, people are taught self-regulatory strategies to overcome their implicit biases. (Devine et al., 2012; Devine et al., 2017; Bruneau, Kteilly, & Urbana, 2019).
  • Social norms: The goal is to make salient that behaving in a non-discriminatory and inclusive way is not only “prescriptively normative” (it’s morally the right thing to do), but also “descriptively normative” (most peers do it). (Murrar, Campbell, & Brauer, 2019; Tankard & Paluck, 2018).
  • Emotions: The goal is to promote perspective-taking and empathy, which can be achieved through specific instruction or role-playing (Vescio, Galinski, Penner).
  • Cooperative learning: The goal is to create groups of individuals belonging to different social groups and to make them mutually interdependent on each other in a learning context. (Aronson, 1978).
  • Entertainment education: The goal is to use entertainment media – TV shows, movies, books, radio soap operas, songs, computer games – to communicate pro-social messages about inclusion and diversity (Murrar & Brauer, 2018).

Although these initiatives have been tested in individual studies, many of them have not undergone the rigorous scientific testing that is required to be able to conclude that they are effective in real-world settings (for a review see Paluck & Green, 2009). Here is a list of the most common weaknesses:

  • Many initiatives have only been tested in laboratory settings, usually with college students as participants.
  • In many studies, the researchers limited themselves to measuring self-reported attitudes toward outgroups.
  • Generally participants knew that they were part of a study, and one may wonder whether they simply reported what they thought the experimenter expected them to report (a phenomenon called “experimental demand”).
  • Most studies do not include delayed outcome measures, and we thus do not know if the observed effects last longer than a few minutes.
  • A large number of studies use a correlational design or a pretest-posttest design, which usually implies that plausible alternative explanations for the observed effects cannot be excluded.
  • Certain initiatives have been evaluated with certain groups (e.g., children’s attitudes toward children with disabilities) but not with other groups (e.g., adult individuals’ attitudes toward ethnic minorities)

We searched the scientific literature for studies that satisfy the following criteria: (1) random assignment to experimental conditions, (2) a delay of at least one month before the measurement of the outcomes, and (3) the use of “hard” outcomes (e.g., grades, turn-over; see here for more information on hard outcomes). We found only one study: Devine and colleagues’ (2017) test of the gender bias habit-breaking workshop, in which participants are taught several mental techniques to overcome judgmental biases. Even for this article, though, the analyses were exploratory, only one of the outcome variables showed a positive effect, and this effect did not meet conventional levels of statistical significance.

Most initiatives to reduce discrimination and prejudice have not been evaluated. Among the few that have been, many turn out to be ineffective, and some even appear to be counterproductive. For example, there is no evidence that fostering intergroup contact improves attitudes toward ethnic outgroups in adult populations (Paluck, Green, & Green, 2018). Fostering social categorization and perspective-taking in group discussions seems to be ineffective as well (Brauer, Judd, & Jacquelin, 2001; Cramer-Walsh, 2007).

Diversity training – which usually involves consciousness raising and perspective-taking – has been evaluated in many studies. The evidence for its effectiveness is mixed at best. Chang and colleagues (2019) showed that diversity training often does not work the way it is supposed to. Whereas most studies find null effects (Forscher et al., 2017a; Kulik & Roberson, 2008), some studies find positive effects (Forscher et al., 2017b; Carnes, Devine, Manwell, Byars-Winston, Fine, Ford, et al., 2015), and yet other studies report negative effects (Anand & Winters, 2008; Dobbin & Kalev, 2013).

Dobbin and Kalev (2016) showed that mandatory diversity training leads to a decrease in ethnic minority employees in management positions. Trainers report that employees often respond to mandatory diversity training with anger and resistance—and many participants report more animosity toward other groups afterward (Kulik, Pepper, Roberson, & Parker, 2007). Employees do not appreciate being blamed for poor race relations, and the references to unconscious processes and the negative messages that are frequently used in diversity training do not increase employees’ motivation to behave inclusively (Dobbin and Kalev, 2018).

Other initiatives that are often implemented in organizational settings also have failed to show the desired effects. Standardized hiring tests, objective performance ratings, and the implementation of grievance procedures do more harm than good (Dobbin & Kalev, 2016). Many initiatives on college campuses are ineffective (Haidt & Jussim, 2016, Wall Street Journal). Recently, Forscher et al. (2019) have shown procedures to change people’s implicit biases are ineffective because changes in implicit bias do not translate into changes in behavior (see this page for a discussion of the scientific literature on implicit bias).


Return to “How to Promote Inclusion in 750 Words” here