assumption that any specific forecast will be true. To improve future forecasts of crime rates and the number of offenders, the panel recommends the following:

  • forecasts should be accompanied by warnings of their inherent inaccuracy and cautions about their appropriate use;

  • forecasts should guard against continuity biases or at least explicitly recognize their presence in projections of which the objective is to draw out implications of recent trends;

  • forecasts should take into account uncertainty in the predictions by developing upper and lower bounds within which paths of crime rates are expected to lie;

  • the forecast time period should be shortened as much as the purpose for which the forecasts are produced will allow; and

  • forecasts should be updated frequently.

The incorporation of these characteristics into crime forecasts should result in more realistic uses and assessments of the forecasts. Nevertheless, current capacity to forecast crime rates is very limited. Errors in forecasts over even relatively short time periods of 2 to 3 years, let alone for a decade or more, are very large.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement