Abstract
The purpose of this study was to compare the accuracy of the method used by the California State Allocation Board for the projection of student enrollment with four other projection methods in an effort to determine the most useful of the methods. A random sample of twenty-five school district's State Allocation Board Form 411s from the school year 1984-85 were provided by the California Office of Local Assistance. These forms provide student enrollment projections, as determined by the state projection process, for the school year 1986-87. The same enrollment data from the Form 411s for the school year 1984-85 was then applied to four other student projection models to determine the accuracy of each model. Correlation coefficients were then compared to determine which model correlated best. The information was used to answer the research questions. (1) What is the correlation between the enrollment numbers produced by the processes of the California State Allocation Board for 1984-85 school districts projected to the 1986-87 school year and the actual enrollment in a sample of 1984-85 school districts? (2) What is the correlation between the actual 1986-87 enrollment numbers in the sample and those produced by applying the state growth percentage from 1984-85 to 1985-86 to the 1984-85 numbers? (3) What is the correlation between the actual 1986-87 enrollment numbers and those produced by applying a straight line projection of the sample numbers for the years 1981-82 through 1984-85? (4) What is the correlation between the actual 1986-87 enrollment numbers and those produced for the sample by applying a method using the ratio of the largest grade growth to the average total growth of the school districts in the sample? (5) What is the correlation between the actual 1986-87 enrollment numbers in the sample and those produced by applying the four year growth percentage of the actual 1981-82 through 1984-85 sample numbers and applying the results to the 1984-85 actual numbers? (6) What is the correlation between the sample numbers produced by the above five applications? (7) What is the accuracy of each district's projection given in percent of error?