powerSignificance {ReplicationSuccess}R Documentation

Replication probabilities for P-values

Description

Computes the probability that a replication study yields a significant effect estimate in the same direction as in the original study.

Usage

powerSignificance(po=NULL, to=p2t(po, alternative=alternative), c=1,
level=0.05, designPrior="conditional", alternative="two.sided")

Arguments

po

P-value from original study, default is NULL

to

test statistic from original study. Is calculated from po, if necessary.

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 "conditional" or "predictive". Defaults to "conditional".

alternative

either "one.sided" or "two.sided". Defaults to "two.sided". Specifies if the P-value is one-sided or two-sided. If the P-value is one-sided, then power calculations are based on a one-sided assessment of significance.

Details

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).

Value

The probability that a replication study yields a significant effect estimate in the same direction as in the original study.

Author(s)

Leonhard Held

References

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

Examples

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")


[Package ReplicationSuccess version 0.1-1 Index]