Bulletin Spring‧Summer 1996

New Deve l opment s in t he App l i ca t i on of St ructural Equa t i on Models Statistics today has many important applications in various fields. Amongst the most common of these applications is the establishment of a statistical model to reveal the underlying phenomena in a particular situation, and the subsequent use of this model to arrive at correct decisions and conclusions. Two Kinds of Variables in Substantive Theo r y In behavioural, educational, medical, and social science, substantive theory (which is concerned with real issues or real effects) usually involves two kinds of variables, namely, the observed and latent variables. Observed variables are those that can be measured directly, like income, test scores, weight, etc. Variables that cannot be measured directly are called latent variables. Examples of these can be intelligence, quantitative ability, health condition, buying behaviour , etc. In practice, a system of equations can be used to denote the characteristics of a latent variable which are reflected by a number of observed variables. Usually, the causations and correlations among the observed and latent variables form the integral part of the theory underlying such a system of equations, and structural equation modelling is the most important and useful statistical technique in assessing these relations. Dr. W.Y. Poon obtained her B.Sc. in mathematics (1983) and M.Phil in statistics (1985)from CUHK, and her Ph.D. inbiostatistic sfrom the University of California at Los Angeles in 1987. She returned to the Department of Statistics at CUHK as a lecturer in 1987 and was promoted to senior lecturer in 1995. Dr. S.Y. Lee obtained his B.Sc. in mathematics from CUHK in 1972, and his Ph.D. inbiostatistic sfrom the University of California at Los Angeles in 1977. He returned to the Department of Mathematics at CUHK as a lecturer in 1977, transferred to the Department of Statistics in 1982, and was promoted to reader rank in 1990.