Modelling Ordinal Responses with Uncertainty: a Hierarchical Marginal Model with Latent Uncertainty components
Preprint
Abstract
In responding to rating questions, an individual may give answers according to his/her knowledge (feeling) or to his/her level of indecision (uncertainty). In this paper, latent binary variables are assumed to rule the answers to rating questions. In this way the joint distribution of the ordinal variables, describing the responses, is modelled by a mixture of components corresponding to the cases of uncertainty in a subset of the answers and feeling in the remaining ones.Keywords
All Related Versions
This publication has 16 references indexed in Scilit:
- A class of mixture models for multidimensional ordinal dataStatistical Modelling, 2016
- Mixture models for ordinal responses to account for uncertainty of choiceAdvances in Data Analysis and Classification, 2016
- Varying uncertainty in CUB modelsAdvances in Data Analysis and Classification, 2016
- Testing order restrictions in contingency tablesMetrika, 2015
- Analyzing bivariate ordinal data with CUB marginsStatistical Modelling, 2014
- Middle category endorsement in odd-numbered Likert response scales: Associated item characteristics, cognitive demands, and preferred meaningsJournal of Research in Personality, 2009
- Identifiability of extended latent class models with individual covariatesComputational Statistics & Data Analysis, 2008
- A mixture model for preferences data analysisComputational Statistics & Data Analysis, 2005
- Multivariate Logistic ModelsJournal of the Royal Statistical Society Series B: Statistical Methodology, 1995
- The Optimal Number of Response Alternatives for a Scale: A ReviewJournal of Marketing Research, 1980