A role for information theory in personality modeling, assessment, and judgment

Abstract

Claude Shannon's groundbreaking work on information theory (Shannon, 1948a) was published just as the field of psychological testing was reaching its potential. Many of the fundamental underpinnings of psychological testing theory were proposed in this same era and the application of these ideas over the ensuing decades has greatly informed our understanding across numerous psychological domains, including individual differences in personality, cognitive abilities, interests, and much more. While the prospect of integrating information theory into psychological assessment was initially received with great enthusiasm, it was quickly and definitively dismissed by Lee Cronbach (1955a), mainly on the grounds that it was incompatible with the practical demands of psychological testing (e.g., limited testing time, relatively small samples). In this chapter, we reconsider Cronbach's rationale in light of recent technological advancements brought about by the "information age," and propose that re-introducing information-theoretic approaches to psychological assessment can advance our knowledge of personality, person-perception, and personality assessment. We conclude by providing several examples of research applications that have already invoked an information theory approach to assessment.

Development of the Hogan Personality Content Single-Items Inventory

“…In addition, responses to inventories consisting entirely of single-item scales may be expected to communicate more total information about the respondent relative to inventories of similar length consisting of multi-item scales (Condon & Mõttus, 2021; Condon et al, 2020). This is because standard scale development practices—such as selecting items to cross benchmarks for internal consistency within multi-item scales (e.g., coefficient α should exceed .70 or .80), or eliminating items with negligible loadings of cross-loadings in factor analyses (DeVellis & Thorpe, 2021; Hinkin, 1998)—should be expected to introduce a greater level of redundancy.…”

The Pathology of Imagination: Picturing the Worst

“…It allows for the analysis of structural relations between the constructs of imagination and personality pathology without excessive burden on participants. While planned missingness designs are sub-optimal for many research questions, they are well-suited for the current work as they allow for a comprehensive and meaningful analysis of constructs that are challenging to measure with complete data (Condon & Mõttus, 2021).…”

Development of the Hogan Personality Content Single-Items (HPCS) Inventory

“…Additionally, responses to inventories consisting entirely of single-item scales may be expected to communicate more total information about the respondent relative to inventories of similar length consisting of multi-item scales (Condon et al, 2020;Condon & Mõttus, 2021). This is because standard scale development practices -such as selecting items to cross benchmarks for internal consistency within multi-item scales (e.g., coefficient 𝛼𝛼 should exceed .70 or .80), or eliminating items with negligible loadings of cross-loadings in factor analyses (DeVellis & Thorpe, 2021;Hinkin, 1998) -should be expected to introduce a greater level of redundancy.…”