A linear regression model was developed with the incarceration rate per thousand as the dependent variable and year as the independent variable. Once the incarceration rates are plotted, these figures are converted into actual estimates of total assigned inmates for these years. This is done by multiplying the incarceration rate for the 18-34 year-old male population estimate for that year. This procedure yields the yearly estimate of inmates assigned to the Department of Corrections. To monitor the accuracy of the estimates, the actual average inmate population is computed after the completion of the first quarter. This actual population is compared with the projected population for that quarter. The difference between these two observations is calculated by subtracting the estimate from the actual and then dividing this result by the actual number of inmates from that quarter, thereby revealing the error in the estimate. The yearly estimate is adjusted by this value and the second, third, and fourth quarter estimates are recalculated. Reasons why Maine does not have a data system in place are examined; among other reasons, the Department of Corrections has a manual system with unique problems that are not easily solved.
Prison Population Forecasting and Management Information Statistics - 'The State of the Art in Maine'
NCJ Number
94368
Date Published
July 1984
Annotation
A methodology used by the State of Maine to forecast the prison population is described.
Abstract
Date Published: July 10, 1984