Multi-level covariance structure analysis of intra-familial substance use

Conventional covariance structure analysis, such as factor analysis, is often applied to data that are obtained in a hierarchical fashion, such as parents and siblings observed within families. However, multivariate modeling of such data are most frequently done as if the data were obtained as a simple random sample from a single population. An alternative specification is presented which explicitly models the within-level and between-level covariance matrices in familial substance use. Results demonstrate homogeneity in substance use within families but heterogeneity across families which could be accounted for by family-level variables of marital status, economic status, and biological relationships. It is shown that conventional covariance structure software can be easily adapted to handle hierarchical models, providing a large set of new analysis possibilities for multi-level data.