Item Response Theory

Item response theory (IRT) is an approach to measurement assessment that is severely underutilized in communication science in general, and media psychology in particular. Communication scholars have more fully embraced a direct alternative to IRT, confirmatory factor analysis (CFA), for the development of valid and reliable operationalizations of its referentially real constructs. An IRT approach to measurement is a clear divergence from the underlying assumptions and principles of classical test theory (CTT). The adoption of IRT brings with it a focus on the item characteristic curve (ICC). IRT originated with a focus on dichotomous responses, but has long since expanded to analyses of polytomous measures (e.g., Likert‐type items). There are strengths and weaknesses to tackling measurement evaluation from a CTT‐based versus an IRT‐based perspective. While analytical procedures like CFA may be more intuitive than IRT and have been made more readily available to communication researchers for decades through a variety of software packages (e.g., AMOS, LISREL, EQS), the rise of the R open‐source statistical analysis environment allows a variety of IRT applications to be within the field's reach. As a result, media psychology researchers should begin to explore IRT as it continues to build its psychometric acumen. It is an analytical approach that may prove valuable for specific types of persistent measurement issues evident in media psychology.