An “Unfolding” Latent Variable Model for Likert Attitude Data

Abstract

Likert attitude data consist of responses to favorable and unfavorable statements about an entity, where responses fall into ordered categories ranging from disagreement to agreement. Social science and marketing researchers frequently use data of this type to measure attitudes toward an entity such as a policy or product. We focus on data on American and British attitudes toward their respective nations (“national pride”). We introduce a multidimensional unfolding model (MUM) to describe the relationship between the data and the attitudes underlying them. Unlike most existing models, the MUM allows the data to reflect not just attitudes, but also response style, which is defined as a consistent and content-independent pattern of response category selection such as a tendency to agree with all statements. The MUM can be used to model multiple attitudes, which allows researchers to expand their analysis of the data of interest to include all available Likert data so as to increase information on response style. For example, we include additional data on immigration attitudes to help distinguish the effects of response style and national pride on our data. The MUM can be used to fit linear models for the effects of background variables on attitudes. Resulting inferences about attitudes are adjusted for response style and should be less biased. Simulation results strongly suggest that, unlike Likert's popular scoring model, the MUM yields unbiased inferences even when there are unequal proportions of favorable and unfavorable statements.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 80 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.