Giovanna De Girolamo

Every school board in Ontario must plan for the future to ensure that its present and future students are properly accommodated. Accommodation planning in recent years has become extremely difficult, as many school boards are faced with declining enrolment in mature, established communities and increasing enrolment in newly built subdivisions. To assist with the accommodation planning process, school board planners develop enrolment projections to determine future facility needs by geographic area. This paper will examine school enrolment trends over a ten year period in mature, fully developed communities within York Region. It will also identify by means of a regression analysis the demographic and housing factors that directly influence school enrolment. This paper will also review the various enrolment projections methodologies currently utilized by various school boards throughout Ontario, and will conduct enrolment projections using both the cohort survival and regression methods. A historical analysis and correlation analysis found that older subdivisions that were built before 1990 tend to yield fewer students compared with newer subdivisions built after 1990. Results further indicate that the regression analysis approach to predicting grade-school enrolment produced results within 8% accuracy, which was superior to the commonly used cohort-survival technique, which produced projections within 12% accuracy. Unlike the cohort-survival technique, which incorporates only existing student data, the regression method utilizes student data along with demographic and housing variables that were found to influence grade-school enrolment. In establishing a regression approach to predicting grade-school enrolment in mature communities, this paper will assist school boards in making decisions regarding pupil accommodation, including school closures, consolidations and boundary changes.