« Previous | Contents | Next »
Listen
WELL WHAT DO YOU THINK (2004): THE SECOND NATIONAL SCOTTISH SURVEY OF PUBLIC ATTITUDES TO MENTAL HEALTH, MENTAL WELL-BEING AND MENTAL HEALTH PROBLEMS
ANNEX G: MULTIVARIATE ANALYSIS
G.1 In addition to the crossbreak analysis conducted on the weighted data it was agreed that some form of multivariate analysis would be performed. Multivariate analysis is a technique used to handle multiple variables at the same time in order to explore the relationships between them. More specifically, it provides a means of identifying the strength of different determinants of attitudes and behaviours, and the extent to which these are correlated with other attitudes, behaviours and socio-demographic characteristics. The particular form of multivariate analysis used in the present study was CHAID analysis.
chaid analysis
G.2 CHAID analysis (Chi-squared Automatic Interaction Detector) segments a population into two or more groups according to their tendency to exhibit a particular characteristic. It then continues to split these groups until no more statistically significant differences are found. CHAID uses statistical techniques (based on CHI-square analysis to select the key drivers (such as gender, age or particular attitudes)of the characteristic and, for these key driver, identifies which groups of respondents are the most likely to exhibit the characteristic. It is important to note that while CHAID analysis highlights relationships between variables, it cannot be used to identify the direction of causality in those relationships. Rather it simply enables us to identify statistically significant differences in terms of the characteristic under study, by the variables included in the analysis. In calculating significant differences, CHAID takes into account the size of the sample. Obviously, the larger a sample is, the more statistically significant relationships CHAID is likely to find. In this study CHAID analysis was used to test the strength of relationships between attitudes towards, and experience of mental ill-health and a range of socio-demographic, behavioural and attitudinal variables.
G.3 The following questions were selected for inclusion in the CHAID for attitudes towards mental ill-health:
QB3: Control over factors affecting mental health;
QB4-15: GHQ12;
C6/7: Experience of mental ill-health;
QD2: Sources of information on mental ill-health;
QD3: Portrayal of mental health in the media; and
QD6: Knowledge of campaigns.
The questions included in the CHAID for the experience of mental ill-health were:
Both sets of CHAID also included the following demographic variables:
Age
Gender
Income
Managing on income
Educational qualifications
Employment status
The full computer output from the CHIAD analysis is presented overleaf.
Figure G.1: CHAID analysis - 'Choose Life' as a discriminator of Tolerance
Figure G.2: CHAID analysis - 'see me' as a discriminator of Tolerance
Figure G.3: CHAID analysis - Experience of a mental health problem as a discriminator of Tolerance
« Previous | Contents | Next »