The conceptualization, development, and evaluation of a national categorical enrollment projection model

The projection of student enrollments is a crucial component in budget formulation and operations planning for school systems. In general, student enrollment projections have been confined to aggregate student populations or grade-by-grade enrollments and have failed to provide educational administrators with adequate projections of special educational service requirements such as special education and compensatory education programs.

This dissertation reviews several different methods of enrollment projection: Linear regression, time series, fuzzy time series and survival methods, and evaluates their use in formulating an enrollment projection model. This dissertation then develops a model for projecting enrollments based on the reviewed methods. A hybrid technique is formulated and then evaluated for use in a national enrollment projection model.

The model projects grade-by-grade and special educational service requirement enrollments for six years. The results of the model are compared to the results of projections developed by the United States Department of Education's National Center for Educational Statistics.