NCJ Number
85293
Date Published
1982
Length
48 pages
Annotation
Three step-by-step methods for jail population forecasting can be completed without a computer. The first two can be used on any calculator, while the third requires a statistical or programmable calculator.
Abstract
Method I bases future jail size on three trends: ratio of State to national population, the ratio of county to State population, and the ratio of jail population to county population. It include a peaking factor to estimate jail population during the greatest periods of use and requires 5 years of data. Method II was developed to deal with variations in jail population from day to day and month to month, allowing planners to identify the influence of discretionary decisions within the criminal justice system. Using 5 years of data from jail records, these projections are based on estimated admissions and releases by month. The first two methods have been used successfully to project bed space needs. Method III -- population projections controlled for seasonal variations -- can be used to estimate several different statistics, such as total bookings, pretrial versus sentenced residents, and resident days. It needs as much data on bookings as possible, preferably 5 years worth grouped by month. Common problems associated with population forecasting methodologies -statistical error, incarceration cycles, relationships between past assumptions and present practices, and organizational and external factors affecting forecasts -- can be partially circumvented using strategies described. Tables, graphs, sample worksheets, and two references are appended.