Section 3:
Means and Medians
Trial Award Patterns
Standard Deviation
Normal Distributions
Awards Test

Means and Medians

Preliminary definitions

Variables can be categorical or quantitative

  • The 2 x 2 tables and statistical techniques already discussed are suitable for analyzing data that fit conveniently into mutually exclusive categories–judge trial vs. jury trial; win vs. lose; black vs. white; life vs. death. Such data often are referred to as categorical. (Such data can be nominal, in the sense of fitting into named categories for which there is no natural ordering–an example being flavors of ice cream. Or such data can be ordinal, in the sense of being ordered–as, for example, when people might agree with a statement, neither agree nor disagree with a statement, or disagree with a statement.)
  • Data of course often exist in other forms. Everyday measures such as temperature measured or distance traveled are not categorical data. Such quantitative measures can take on many values and have a natural ordering, and so require different statistical techniques.

Quantitative variables can be continuous or discrete

  • Continuous variables, such as temperature, can take any value within a range.
  • Discrete quantitative variables could be the number of motions filed in a case or the number of children in a family.

Copyright © 2002 by Theodore Eisenberg & Kevin M. Clermont
Cornell University
Cornell Law School
Cornell University
Comments to ted@teddy.law.cornell.edu
Last updated: September 2002