Estimating SEM with Ordinal Data and Scale Usage Heterogeneity

32 Pages Posted: 11 Nov 2010

See all articles by Nino Hardt

Nino Hardt

Ohio State University (OSU) - Fisher College of Business

Joachim Bueschken

Catholic University of Eichstätt-Ingolstadt

Date Written: November 8, 2010

Abstract

In the social sciences, it is often useful to introduce latent variables and use structural equation modeling to quantify relations among observable and latent variables. Data gathered through surveys is frequently ordinal in nature, as Likert-scale type questions are used in most questionnaires. Although a large number of publications shows that treating ordinal data as metric can lead to wrong results, this practice is still common. Moreover, respondents treat the rating scales differently. There may be "yea"-sayers whose responses are biased in the positive direction (or the opposite). Some respondents may also use some part of the scale only. This problem can be explicitly addressed by modeling heterogeneity scale usage. This note shows how to estimate SEM with ordinal data and scale usage heterogeneity.

Keywords: Structural Equation Model, Scale Usage, Ordinal Data, Bayesian Estimation

JEL Classification: C11, C13

Suggested Citation

Hardt, Nino and Bueschken, Joachim, Estimating SEM with Ordinal Data and Scale Usage Heterogeneity (November 8, 2010). Available at SSRN: https://ssrn.com/abstract=1705385 or http://dx.doi.org/10.2139/ssrn.1705385

Nino Hardt (Contact Author)

Ohio State University (OSU) - Fisher College of Business ( email )

2100 Neil Avenue
Columbus, OH 43210-1144
United States

Joachim Bueschken

Catholic University of Eichstätt-Ingolstadt ( email )

Auf der Schanz 49
Ingolstadt, D-85049
Germany
+498419371976 (Phone)
+498419372976 (Fax)

HOME PAGE: http://www.wfi.edu/mkt

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
558
Abstract Views
2,471
rank
54,452
PlumX Metrics