Published online by Cambridge University Press: 01 July 2024
The controlled experiment is undoubtedly the ideal instrument for testing the effectiveness of a new rule of law or legal procedure: have one random half of the cases operate under the new rule, the other half under the old rule, and see what difference there is in the outcome.
1. See, for example, R. S. James, The Jury and the Defense of Insanity (1967).
2. See The Case for the Official Experiment in Zeisel, Kalven, Buchholz, Delay in the Court ch. 21 (1959).
3. The idea was first developed to overcome not legal but factual obstacles to experimentation. To test the effectiveness of television commercials, it is clearly impossible to have a random group of homes view the program while the control group refrains from viewing. An indirect experiment was designed, making non-exposure the experimental variable. See I. M. Towers, L. A. Goodman, H. Zeisel, A Method of Experiments 87-110 (No. 4, Studies in Public Communication, U. of Chicago, 1962).
4. See The Cross-tabulation Explains in H. Zeisel, Say It With Figures ch. 9 (5th ed. 1968).
5. See H. Zeisel, The Law, in The Uses of Sociology 81, esp. 91 n. 12 (P. F. Lazarsfeld ed. 1967).
6. See M. Rosenberg, The Pre-trial Conference and Effective Justice (1964). Also H. Zeisel, Optional vs. Obligatory Pre-trial in 4 Trial 11 (Jan. 1968).
7. See Delay in the Court, supra note 2, at 99.
8. The result was corroborated by another important finding that explained it: that in one third of all cases the verdict against liability obviated litigation of the damages; and where liability was found, the parties were as a rule quick to settle the damages without further trial.