U.S. flag

An official website of the United States government, Department of Justice.

NCJRS Virtual Library

The Virtual Library houses over 235,000 criminal justice resources, including all known OJP works.
Click here to search the NCJRS Virtual Library

Case Prediction by Staircase Tables and Percentaging

NCJ Number
97084
Journal
Jurimetrics Volume: 25 Issue: 2 Dated: (Winter 1985) Pages: 168-195
Author(s)
S S Nagel
Date Published
1985
Length
28 pages
Annotation
Staircase prediction and percentaging are two new methods for systematically predicting the outcome of court cases by providing insights which supplement tradition legal and social science analysis.
Abstract
Staircase prediction applies when two crucial predictive variables are used; percentaging, when there are more than two main predictive variables. Both methods use the same first four steps: finding one or more key cases, deciding the variables or case characteristics associated with winning or losing those cases, deciding subdivisions for each variable, and arranging subdivision order from least favorable to most favorable. The next steps in the staircase method are to prepare a two-dimensional matrix showing one variable along the top and the second variable along the side, to insert the predicted outcomes for each combination of variables, and to draw a staircase showing in what situations the liberal side will win and in what situations the liberal side will lose. The percentage method also uses a table, but assigns percentages to each combination of variables based on the number of variables. The methods have advantages over traditional methods of prediction because a large, random sample of cases is not needed, they represent deductive analysis and thus avoid the need to generalize and the danger of interference by the presence of other variables, and they stimulate careful and logical thinking about variables and categories of variables and their influence on decisionmaking. Case illustrations of the prediction methods using one variable, multiple variables, and unequal variables are provided. A description of trichotomy prediction using one variable is included. Illustrative examples relate to due process, equal protection, and the first amendment. Tables and footnotes are supplied.

Downloads

No download available

Availability