Using p Values to Estimate the Probability of a Statistically Significant Replication
- 2 May 2002
- journal article
- Published by Taylor & Francis in Understanding Statistics
- Vol. 1 (2), 101-112
- https://doi.org/10.1207/s15328031us0102_03
Abstract
Many commentators have discussed misinterpretations of null hypothesis significance tests. Although probability levels are often misinterpreted, there is a valid interpretation that is not well-known. With the wide access to statistical programs it has become possible to use the exact probability levels of tests of statistical significance to estimate the probability that replications of completed studies would be statistically significant. It is shown that the difference between an observed value of t and the critical value of t can provide such estimates. The method is applied to tests of the statistical significance of correlation coefficients also. Both a graphical and a tabular method of converting a probability level of an initial study into the probability of a statistically significant replication are provided.Keywords
This publication has 7 references indexed in Scilit:
- Null hypothesis significance testing: A review of an old and continuing controversy.Psychological Methods, 2000
- Statistical methods in psychology journals: Guidelines and explanations.American Psychologist, 1999
- Why scientists valuep valuesPsychonomic Bulletin & Review, 1998
- Statistical Significance and ReplicabilityTheory & Psychology, 1998
- Effect sizes and p values: What should be reported and what should be replicated?Psychophysiology, 1996
- The earth is round (p < .05).American Psychologist, 1994
- Theoretical risks and tabular asterisks: Sir Karl, Sir Ronald, and the slow progress of soft psychology.Journal of Consulting and Clinical Psychology, 1978