Conceptualizing identification: A comment on Downs, Bowman, and Banks (2017).

Abstract

As Downs, Bowman, and Banks (2017) noted, researchers use the term “identification” to refer to a variety of different types of measures that actually seem to assess quite disparate subconstructs. These authors proposed that identification should be seen as a polythetic construct and reported data assessing the validity of 6 different subconstructs making up a comprehensive model of identification. However, we propose that this conceptualization of identification is premature. First, given the diversity of the subconstructs, it would likely be difficult to give a precise definition of identification in this model, and we question whether identification is a useful term if it cannot be defined in a specific way. Second, we argue that it is not clear what “family resemblance” there is between these subconstructs that justifies their inclusion in a polythetic model of identification, and note that identification lacks the qualities of other more commonly accepted polythetic constructs, such as self-esteem and experiential avoidance. Finally, we argue that the data reported does not demonstrate that these subconstructs relate to a broader construct in a uniform way. We propose that rather than including these diverse approaches to measuring identification as subconstructs in a polythetic model, researchers should label their measures more specifically, allowing us to test important hypotheses about whether and how these different types of measures relate to each other and to outcomes of interest. Further research is needed to evaluate whether identification is a truly polythetic term before encouraging further confusion of terms in this subfield. (PsycInfo Database Record (c) 2020 APA, all rights reserved) Public Policy Relevance Statement—Successful communication requires that members of a community share common understandings of the meaning of the terms they use. Downs et al. (2017) set forth a polythetic model of identification, but we argue that this line of thinking may further muddle how media researchers use the term “identification.” Instead, we suggest that researchers label constructs more specifically so that all members of the community have a shared understanding of what is being measured. (PsycInfo Database Record (c) 2020 APA, all rights reserved)