ABSTRACT
Though, in principle, the concept of latent factors seems to have been suggested by Galton (1888), the formulation and early development of factor analysis have their genesis in psychology and are generally attributed to Spearman (1904). He first hypothesized that the correlations among a set of intelligence test scores could be generated by linear functions of a single latent factor of general intellectual ability and a second set of specific factors representing the unique characteristics of individual tests. Thurston (1945) extended Spearman’s model to include many latent factors and proposed a a method, known as the centroid method, for estimating the coefficients of different factors (usually called factor loadings) in the linear model from a given correlation matrix. Lawley (1940), assuming normal distribution for the random
vector X, estimated these factor loadings by using the method of maximum likelihood.
