Fig 1.
Number of people in each region (represented by the heights of the columns on the vertical axis) with each regional percentage possessing the genetic marker being represented on the horizontal axis and the equal number of random selection record cards represented by the rectangular distribution below the baseline.
Fig 2.
Comparing effects of prior probabilities of 0.01 and 0.001 for true proportions on the likelihood and posterior distributions of observed proportions of [40\99] and [60\99].
Fig 3.
Comparison of distributions of posterior probabilities of true proportions conditional on observed proportions and binomial and Gaussian likelihood distributions.
Fig 4.
Comparing the probability of non-replication conditional on an observed proportion and one-sided P-values conditional on Null Hypotheses.
Fig 5.
Comparison of probabilities of non-replication for observed proportions of r/100 with one-sided P-values for different null hypotheses of 0.5 and 0.4.
Fig 6.
Calculating the posterior probability distribution of true outcomes from a prior probability distribution and a likelihood distribution Θi.
Fig 7.
Taking into account the probability of impeccable reproducibility for an observation [50\100].
Fig 8.
The effect of a probability of impeccable reproducibility on the probability of replication based on Z scores representing Θi.
Table 1.
Some comparisons between probability of replication, P-values and confidence intervals.