Analysing a Mental Health Survey by Chi-Squared Automatic Interaction

• Main Authors: Huang, H. C., LIN, Ting Kwong, Ngui, P. W.
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 1993
• Subjects: Econometrics; Medicine and Health Sciences
• Online Access: https://ink.library.smu.edu.sg/soe_research/289
• Institution: Singapore Management University

Based on a mental health survey conducted in 1987 by the Singapore Association for Mental Health, an exploratory statistical technique CHAID (Chi-Squared Automatic Interaction Detection), which is an offshoot of AID (Automatic Interaction Detection) designed for a categorical dependent variable, was employed to establish the characteristics of people who were vulnerable to mental problems in Singapore. Of the sixteen predictors considered, five were found to play significant roles in identifying the various vulnerable groups. They were the respondent's perception of his or her own health, the number of problems encountered during the four weeks before the survey, whether experiencing a depression or not during the four weeks prior to the survey, the respondent's perception of family support and the number of life events in the past six months before the survey, summarised into a weighted score. The dendrogram obtained from CHAID was very useful in displaying the structure of the relationship of the predictors with mental health. Examining the dendrogram, one can easily classify the various vulnerable groups by tracing each of the terminal groups to the root of the tree.