The MTMM approach to coping with measurement errors in survey research

Abstract

Designing a survey involves many more decisions than most researchers realize. Some survey specialists therefore talk about the art of designing survey questions (Payne 1951). Designing a survey in a scientific way requires knowledge of how the many decisions that researchers take in survey design affect the quality of questions. Many studies have been done in this area. Inspired by the work of Frank Andrews (1984), we elaborated his Multitrait Multimethod (MTMM) approach to evaluating the quality of questions. On the basis of a meta-analysis of a large number of MTMM experiments, we developed a program (SQP) that can predict the quality of questions before they are used in the field (Saris and Gallhofer, 2007). In this paper we will briefly explain the approach we have chosen, illustrate the method and discuss its advantages and disadvantages.