LPA2: a Fortran V Computer Program for Green's Solution of Latent Class Analysis Applied to Latent Profile Analysis

Abstract

A computer program for solving latent profile analysis (Gibson, 1959) is presented. Green's (1951) solution for latent class analysis was used. N observations on p variables constitute the data matrix. The program estimates latent mean vectors and classifies the observation vectors to them. The result is a number of clusters of the observation vectors. A χ²-test of local independence is defined. Latent and observed discriminabilities are defined as the proportion of a variable's variance which is explained by between-variance. Discriminability constitutes the lower bound for estimated variable reliability. The program performs the analysis for a specified series of numbers of latent profiles and specified values on the diagonal of the product moment correlation matrix.

Get full access to this article

View all access options for this article.

References

Anderson, T. W. An introduction to multivariate statistical analysis. New York: Wiley, 1958.

Gibson, W. A. Three multivariate models: Factor analysis, latent structure analysis and latent profile analysis. Psychometrika, 1959, 24, 229-252.

Green, B. F., Jr. A general solution for the latent class model of latent structure analysis. Psychometrika , 1951, 16, 151-166.

Lazarsfeld, P. F. Latent structure analysis . In S. Koch (Ed.), Psychology: A study of a science. Vol. 3. New York: McGraw-Hill, 1959, Pp. 476-543.

Lazarsfeld, P. F. and Henry, N. W. Latent structure analysis. Boston : Houghton Mifflin, 1968.

Meredith, W. Some results based on a general stochastic model for mental tests. Psychometrika, 1965, 30, 419-440.