Ordinal Data Models for No-Opinion Responses in Attitude Survey

Abstract

In analyzing data from attitude surveys, it is common to consider the “don’t know” responses as missing values. In this article, we present a statistical model commonly used for the analysis of responses/evaluations expressed on Likert scales and extended to take into account the presence of don’t know responses. The main objective is to offer an alternative to the usual custom to treat them as missing values by considering them as a source of uncertainty. The original proposal in this article is the introduction of the relevant covariates in order to discriminate subpopulations that can show different behaviors in choosing between a substantive response and the don’t know option.

How to Interpret the Effect of Covariates on the Extreme Categories in Ordinal Data Models

“…An alternative approach may be the analysis of ME measures for ordinal data models that take into account the “don’t know” option as in Iannario et al (2020) or ME interpretation in case where the data generating process results into too many zeroes as for zero-inflated ordinal data models (see Harris and Zhao 2007). Finally, although this article is developed in a frequentist framework, it could be of interest to study the corresponding counterpart measures in a Bayesian setting.…”

“…Tutz and Schneider's proposal deals with a more flexible distribution that allows to generalize our approach on the extreme categories distinguishing between a tendency to middle categories and a tendency to extreme categories. An alternative approach may be the analysis of ME measures for ordinal data models that take into account the "don't know" option as in Iannario et al (2020) or ME interpretation in case where the data generating process results into too many zeroes as for zero-inflated ordinal data models (see Harris and Zhao 2007). Finally, although this article is developed in a frequentist framework, it could be of interest to study the corresponding counterpart measures in a Bayesian setting.…”

Revisiting China’s Social Volcano: Attitudes toward Inequality and Political Trust in China

“…In attitudinal surveys, it is common for there to be a considerable proportion of such responses. 4 Although some studies treat them as missing values or “neutral” between agreement and disagreement on the Likert-type scale, more and more studies now question such practices and show that respondents really mean “don’t know” when they answer “don’t know” (Denman et al 2018; Iannario et al 2018; Shen et al 2019). On the basis of my substantive interests and methodological considerations, I decided to focus on the comparison between people who explicitly express their discontent or distrust and those who do not.…”

Uncertain Choices: The Heterogeneous Multinomial Logit Model

“…Explanatory variables can determine the probabilities of the mixture. D’Elia and Piccolo (2005), Iannario and Piccolo (2010), Iannario (2012), Tutz et al (2017), and Iannario et al (2020) all considered models of this type, and Piccolo and Simone (2019) provided an extensive overview.…”