Send us your feedback

Thank you for your feedback. An email has been sent to the ESRC support team.

An error occured whilst sending your feedback. Please review the problems below.

The research catalogue is an archive of ESRC-funded grants and outputs. Links, files and other content will no longer be maintained or updated after April 2014.

Latent variable modelling of categorical data: Tools of analysis for cross-national surveys

In a cross-national social survey, the same questions are asked of respondents in several different countries, often tens of thousands of individuals in dozens of countries.

This project will develop statistical methods for analysing the extensive data from such surveys. These methods can be used to compare levels and patterns of public values, opinions and attitudes in different countries. For example, the methodological work of the project will be tested and illustrated with cross-national comparisons in three areas:

  • confidence in criminal justice systems
  • public attitudes towards science and technology
  • characteristics of civil society.

The same techniques are also used to examine cross-national comparability of survey questions, ie whether the questions work the same way in all of the countries in a survey.

The aim of the project is to determine how such methods can be best used to analyse cross-national survey data. This will be examined using a combination of statistical theory, computer simulations and analysis of data from existing cross-national surveys. The findings will provide practical advice and examples to all survey researchers, and will be communicated through workshops, on-line training materials, and academic publications.