Complex sample design effects and inference for mental health survey data

Abstract

Mental health researchers world-wide are using large-scale sample survey methods to study mental health epidemiology and services utilization in general, non-clinical populations (Alegria et al. in press). This article reviews important statistical methods and software that apply to descriptive and multivariate analysis of data collected in sample surveys. A comparative analysis of mental health surveys in international locations is used to illustrate analysis procedures and ‘design effects’ for survey estimates of population statistics, model parameters and test statistics. This article addresses the following questions. How should a research analyst approach the analysis of sample survey data? Are there software tools available to perform this analysis? Is the use of ‘correct’ survey analysis methods important to interpretation of survey data? It addresses the question of approaches to the analysis of complex sample survey data. The latest developments in software tools for the analysis of complex sample survey data are covered, and empirical examples are presented that illustrate the impact of survey sample design features on the interpretation of confidence intervals and test statistics for univariate and multivariate analyses. Copyright © 1998 Whurr Publishers Ltd.