powerSignificance {ReplicationSuccess} | R Documentation |
Computes the probability that a replication study yields a significant effect estimate in the same direction as in the original study.
powerSignificance(po=NULL, to=p2t(po, alternative=alternative), c=1, level=0.05, designPrior="conditional", alternative="two.sided")
po |
P-value from original study, default is |
to |
test statistic from original study. Is calculated from |
c |
The ratio of the variances of the original and replication effect estimates. This is usually the ratio of the sample size of the replication study to the sample size of the original study. |
level |
The significance level of the replication study. Default is 0.05. |
designPrior |
either |
alternative |
either |
This extends the 'replication probability', the probability of repeating a statistically significant result by Goodman (1992), to the case of possibly unequal sample sizes, see also Senn (2002).
The probability that a replication study yields a significant effect estimate in the same direction as in the original study.
Leonhard Held
Goodman, S. N. (1992). A comment on replication, p-values and evidence, Statistics in Medicine, 11, 875–879.
Senn, S. (2002). Letter to the Editor, Statistics in Medicine, 21, 2437–2444.
Held, L. (2019). A new standard for the analysis and design of replication studies (with discussion). To appear in Journal of the Royal Statistical Society, Series A. https://arxiv.org/abs/1811.10287
powerSignificance(po=0.005, c=2) powerSignificance(po=0.005, c=2, designPrior="predictive") powerSignificance(po=0.005, c=2, alternative="one.sided") powerSignificance(po=0.005, c=2, designPrior="predictive", alternative="one.sided") powerSignificance(po=0.005, c=1/2) powerSignificance(po=0.005, c=1/2, designPrior="predictive") powerSignificance(po=0.005, c=1/2, alternative="one.sided") powerSignificance(po=0.005, c=1/2, designPrior="predictive", alternative="one.sided")