Socio-Demographic Clustering
Depression risk did not follow a simple linear socioeconomic gradient across clusters. A five-cluster solution was selected to balance model fit and interpretability. K-modes clustering was chosen because all clustering variables were cate…
1 sources - 5 claims
Depression risk did not follow a simple linear socioeconomic gradient across clusters. A five-cluster solution was selected to balance model fit and interpretability. K-modes clustering was chosen because all clustering variables were categorical. The study used five categorical variables to form socio-demographic clusters. The five clusters combined demographic, socioeconomic, geographic, and behavioural characteristics.