A binomial latent trait model for the study of Likert‐style attitude questionnaires

A binomial logistic latent trait model for attitude items constructed and scored in the Likert tradition is proposed. Unconditional maximum‐likelihood estimates for subject and item parameters are examined and the model is applied to both simulated and real data. Advantages for data analyses which accrue from having an explicit model for Likert questionnaires are considered with one particular feature being a ready disclosure as to whether or not the neutral category functions as is expected. 1978 The British Psychological Society