The article reviews various methods for assessing the unidimensionality of tests and items, evaluating their rationale and effectiveness. It highlights that many indices lack a clear rationale and are often adjustments of previous indices to address criticisms. The author suggests that indices based on the size of residuals after fitting a two- or three-parameter latent trait model may be the most useful for detecting unidimensionality. The paper clarifies the term "unidimensional" and distinguishes it from related concepts such as reliability, internal consistency, and homogeneity. It defines unidimensionality as the existence of one latent trait underlying the data. The article discusses the critical assumption of measurement theory that a set of items measures a single common trait and emphasizes the importance of unidimensionality in psychological testing. It reviews indices based on answer patterns, reliability, principal components, factor analysis, and latent trait models, critiquing their limitations and proposing improvements. The author argues that most indices lack a clear rationale and are not effectively tested, and suggests that the latent trait model, particularly with its focus on local independence, may provide a more robust approach to assessing unidimensionality.The article reviews various methods for assessing the unidimensionality of tests and items, evaluating their rationale and effectiveness. It highlights that many indices lack a clear rationale and are often adjustments of previous indices to address criticisms. The author suggests that indices based on the size of residuals after fitting a two- or three-parameter latent trait model may be the most useful for detecting unidimensionality. The paper clarifies the term "unidimensional" and distinguishes it from related concepts such as reliability, internal consistency, and homogeneity. It defines unidimensionality as the existence of one latent trait underlying the data. The article discusses the critical assumption of measurement theory that a set of items measures a single common trait and emphasizes the importance of unidimensionality in psychological testing. It reviews indices based on answer patterns, reliability, principal components, factor analysis, and latent trait models, critiquing their limitations and proposing improvements. The author argues that most indices lack a clear rationale and are not effectively tested, and suggests that the latent trait model, particularly with its focus on local independence, may provide a more robust approach to assessing unidimensionality.