Table 2.

Statistical Measures in Bayesian Probability

Notation/Abbreviation	Full Name	Interpretation
Prior	Prior distribution	Distribution of the effect size, as assumed prior to data collection/analysis
Posterior	Posterior distribution	Actual distribution of the effect size after the data at hand have been analysed
P(M)	Prior model probability	Probability of this particular statistical model being supported by the data at hand, as assumed prior to data collection/analysis
P(M|data)	Posterior model probability	Posterior probability of this particular model being supported by the data at hand, after they have been analysed
BF	Bayes factor	The strength of evidence in favour of a given statistical model, relative to another statistical model (see below)
BF01	Bayes factor 0/1	The strength of evidence in favour of Model 0, relative to Model 1
BF10	Bayes factor 1/0	The strength of evidence in favour of Model 1, relative to Model 0
BF10 > 100		‘Extreme evidence’ favouring Model 1, relative to Model 0
BF10 > 30		‘Very strong evidence’ favouring Model 1, relative to Model 0
BF10 > 10		‘Strong evidence’ favouring Model 1, relative to Model 0
BF10 > 3		‘Moderate evidence’ favouring Model 1, relative to Model 0
BF10 = 1		Model 1 and Model 0 are equally supported by the evidence
BF10 < 0.33		‘Moderate evidence’ against Model 1, relative to Model 0 (equivalent to BF01 > 3)
BF10 < 0.10		‘Strong evidence’ against Model 1, relative to Model 0 (equivalent to BF01 > 10)
BF10 < 0.03		‘Very strong evidence’ against Model 1, relative to Model 0 (equivalent to BF01 > 30)
BF10 < 0.01		‘Extreme evidence’ against Model 1, relative to Model 0 (equivalent to BF01 > 100)
Error%	Stability of the BF	The range of the BF over the chosen Markov chain Monte Carlo iterations, e.g. BF10 = 10 with error% = 20 means that the BF10 ranged from 8 to 12
95% CI	Credible interval	With 95% certainty, the true effect size lies within these bounds