Essex Summer School in
Social Science Data Analysis

Cristiano Vezzoni

Cristiano Vezzoni is assistant professor in Research Methodology, Department of Sociology and Social Research, University of Trento (Italy). His substantive research interests are in political behaviour, relations between religion and politics, contextual effects on vote. His methodological interests concern the development and application of statistical methods for comparative survey research, with a particular focus on problems of measurement.

1M Measurement Instruments in Survey Research

Cristiano Vezzoni, University of Trento
9 - 20 July (two week course / 35 hrs)

Detailed Course Outline [PDF]

Course Content

Measurement in survey research often involves combining answers to a number of questions, such that the result is a measurement of a latent variable (e.g., degree of political participation, religious belief, self esteem, political knowledge). After a general introduction to the problem of measurement in sociology and psychology, traditional measurement models (reliability analysis and factor analysis) will be presented, applied, and compared with more recently developed measurement methods, known as Item Response Theory (IRT). Two types of IRT-models will be distinguished: Mokken models for non-parametric measurement and Rasch models for parametric measurement. A model for categorical data (latent class analysis) will also be discussed. Finally, some developments in the direction of unfolding analysis will be presented. Specific attention will be paid to the formulation of survey questions that allow for the application of different measurement models, depending on the kind of data produced (continuous, ordinal, dichotomous), as well as to the problem of measurement comparison in cross-cultural research. Participants are encouraged to bring their own data.

Course Objectives

First, that participants understand different measurement models, and know how to select among them. Secondly, that participants have sufficient practical experience to use these models in their own research. The course is intended for both students that are preparing their own research instruments (namely, questionnaires), and students that already have data (collected or derived from larger surveys, eg. ESS) and are developing their own analysis.

Course Prerequisites

Participants need a basic familiarity with data analysis and a moderate grasp of basic statistical reasoning. Prior experience with SPSS and with stand-alone packages is recommended.

Representative Background Reading

Saris W., and I.N. Gallhofer. 2007. Design, Evaluation, and Analysis of Questionnaires for Survey Research. Wiley.

Borsboom, D. 2005. Measuring the Mind. Cambridge.

Embretson, S.E., and S. Reise. 2000. Item Response Theory for Psychologists. Erlbaum.

Sijtsma, K., and I.W. Molenaar. 2002. Introduction to Nonparametric Item Response Theory. Sage.

[top of page]