Categorizing identities: Race, gender, disability, and sexual orientation

Abstract

In survey research, the treatment of demographic information often ignores the complexity of identity. Instead of treating aspects of identity as clear-cut, simple questions, self-categorization theory illustrates the fluid and contextual nature of self-identification (Abdelal, 2009; Onorato & Turner, 2004; Turner, Oakes, & Haslam, 1994). Accounting for the fluidity and multiplicity of individual identity in survey research can be difficult. For example, the U.S. Census in 2000 was the first time that respondents could select more than one racial category. Certainly, this change in the survey could lead to more accurate representation of individual identity, but there may be disadvantages to augmenting demographic questions, such as difficulty comparing to data from previous years or unintended consequences like multiracial individuals being 'lumped' together despite having very different experiences. For institutional researchers, collecting demographic data can create competing priorities between uniform data reporting for national agencies, specialized information gathering for institutional partners, and inclusive question wording that represent the complexity of identity.