The classroom experiences of children in two model preschool programs were compared using an observational technique which was developed for a study of Head Start classrooms. The main objective of the study was to determine if the observational scale could discriminate between two programs of widely differing philosophy. The second goal of the study was to determine the number of independent dimensions which differentiated the two programs. Univariate analyses of variance showed that the programs differed on a number of variables. In order to determine if these variables were all measuring the same dimension, multivariate analyses were used. The results of these analyses indicated that more than one dimension was required to properly describe the classrooms but that the precise nature of the dimensions depended on whether one was interested in differences between the programs or variations within a program. (Author)