Abstract
When multiple significance tests are computed, a certain number of “significant” findings will emerge simply because of chance fluctuations. In the present paper, some factors affecting the number of nominally significant results are elaborated and a general method is suggested which permits unbiased inference as to the significance of a set of findings,as a set. The method advocated employs a high speed computer to generate empirically a sampling distribution tailormade to a particular data matrix. The method is illustrated in the case of dichotomous response to inventory items, where it is found that the statistical model still often used as a basis for estimation is overly conservative. Some problems in the application of the method are discussed.
Similar content being viewed by others
References
Arthur, A. O. Random digit generation.Computing News, 1956,4, 85: 5–85: 14.
Brozek, J. and Tiede, K. Reliable and questionable significance in a series of statistical tests.Psychol. Bull., 1952,49, 339–341.
Johnson, D. L. Generating and testing pseudo-random numbers on the IBM Type 701.Mathematical tables and other aids to computation, 1956,10, 8–13.
Jones, L. V. and Fiske, D. W. Models for testing the significance of combined results.Psychol. Bull., 1953,50, 375–382.
Latscha, R. Tests of significance in a 2 × 2 contingency table: extension of Finney's table.Biometrika, 1953,40, 74–86.
Meyer, H. A. (Ed.)Symposium on Monte Carlo methods. New York: Wiley, 1956.
Mosteller, F. and Bush, R. R. Selected quantitative techniques. In G. Lindzey (Ed.),Handbook of social psychology. Cambridge, Mass.: Addison-Wesley, 1954. Pp. 289–334.
Sakoda, J. M., Cohen, B. H. and Beall, G. Test of significance for a series of statistical tests.Psychol. Bull., 1954,51, 172–175.
Tukey, J. W. Comparing two small samples on many items.Memo. Rep. 54, Statist. Res. Group, Princeton Univ., 1954.
Wilkinson, B. A statistical consideration in psychological research.Psychol. Bull., 1951,48, 156–158.
Author information
Authors and Affiliations
Additional information
This investigation, part of a larger research project, was supported primarily by research grant M-1078 from the National Institute of Mental Health of the United States Public Health Service. Gratitude is also extended to the National Science Foundation for making available research time on the IBM Model 701 computer. The present paper has benefited greatly from the comments on an earlier version offered by John Tukey and an editor ofPsychometrika.
Rights and permissions
About this article
Cite this article
Block, J. On the number of significant findings to be expected by chance. Psychometrika 25, 369–380 (1960). https://doi.org/10.1007/BF02289754
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF02289754