Measurement Equivalence in Emotion Perception Across Racial Groups

• In Event: Race-Conscious Practice and Representation in Contemporary Social and Emotional Learning

Abstract

The perception of emotions in others is a fundamental human ability (Ekman, 2003; Ekman et al., 1987; Ekman, Sorenson & Friesen, 1969; cf. Matsumoto, 1992) and is the underpinning of interpersonal communication (Mehrabian & Ferris, 1967; Mehrabian, 1971). Some have even gone as far to say that most information is communicated nonverbally through emotion (Mehrabian, 1971) and thus accurately perceiving emotions is critical for effective communication. However, while emotion perception is a universal phenomenon, the accurate perception of emotion can be biased by subtle differences in expression and processing across groups. In particular, research has demonstrated considerable evidence of ‘like-me’ biases in facial perception across racial in-groups and out-groups (Elfenbein, Beaupre, Levesque, & Hess, 2007; Johnson & Fredrickson; Kelly et al., 2007; Meissner & Brighanm, 2001; Sporer, Trinkl, & Guberova, 2007). These biases arise for numerous reasons including the accuracy and speed at which we are able to process faces of racially different others. The substantive consequence for SEL research is that these biases may lead to the inaccurate perception of emotion across racially diverse individuals. To date, most assessments of emotion perception have largely used racially homogeneous expressors which may lead to the inaccurate or even unfair assessment of emotion perception ability.
In this session, we discuss preliminary evidence in favor of using racially diverse expressions of emotion to accurately and fairly assess emotion perception ability across racial groups. We discuss the design and creation of a new test of emotion perception that includes racially diverse expressers (i.e. White, Black, Asian, and Hispanic expressers, see Figure 1). Furthermore, using robust measurement equivalence analysis, we show that by including racially diverse expressors we can fairly assess emotional intelligence across racial groups. In a large sample of working adults in the US, we find small average differences in emotion perception ability based on race of perceiver and also find evidence of measurement equivalence across racial groups (see Table 1). We will discuss our findings and how this assessment tool and race-conscious approach may be used in SEL research.

Authors

• Matthew LaPalme, Yale University

• James Floman, Yale University

• Sigal Barsade, Wharton School of Business

• Marc A. Brackett, Yale University