Dimensionality

Abstract

This chapter offers a discussion of multidimensionality in health outcome scales and describes methods that can help indicate if there is multidimensionality in a data set. Two different situations are considered: confirmatory analysis where an a priori hypothesis is tested regarding which items measure what latent construct, and exploratory analysis where no such hypotheses exist. The Hospital Anxiety and Depression Scale (HADS) data is used for illustration throughout the chapter. One way of testing unidimensionality is to formulate multidimensional models and test the unidimensional model against them. Item parameters are estimated using either a marginal maximum likelihood (MML) approach or a conditional maximum likelihood (CML) approach. Multidimensional Rasch models have been proposed in the MML framework as well as in the CML framework. Further, the chapter discusses diagnostics for detection of multidimensionality and estimation of the magnitude of multidimensionality.

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 — 30 citations exist upstream.

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