The Bayesian Correlation analysis allows for the estimation of the population correlation, as well as the comparison of hypotheses. The three comparisons are (i) between the null hypothesis (H0) that the correlation between pairs of variables equal 0, and the alternative hypothesis (H1) that the population correlation takes its value between -1 and 1; (ii) between H0 and the alternative hypothesis (H+) that the population correlation is positive; and (iii) between H0 and the alternative hypothesis (H-) that the population correlation is negative. All possible pairs of the specified variables are analyzed.
Kendall's tau-b: Kendall's tau-b rank-order correlation coefficient to quantify the monotonic association between two variables.
Scatterplot: Displays scatterplots for each specified pair of variables.
Prior and posterior: Displays the prior and posterior distribution of the correlation under the alternative hypothesis for each specified pair of variables.
Testing info: Adds the Bayes factor computed with the user-defined prior; adds a probability wheel depicting the odds of the data under the null vs. alternative hypothesis (under the assumption that the null and alternative hypotheses were equal probable a priori, i.e., P(H0)=P(H1)=1/2); adds indicators representing the prior and posterior at the test point (e.g., rho=0), the ratio between these two values equals the Bayes factor.
Bayes factor robustness check: Displays the Bayes factor as a function of the width of the stretched beta prior on the correlation for each specified pair of variables. The width of the kappa prior is varied between 0 and 2.
Additional info: Adds the Bayes factor computed with the user-defined prior and the maximum obtainable Bayes factor.
Sequential analysis: Displays the development of the Bayes factor as the data come in using the user-defined prior for each specified pair of variables.
Credible interval: Central credible intervals for the correlation coefficient.
Flag supported correlations: Correlations that are supported to the Bayes factor are marked with (see Jeffreys (1961) for evidence categories).
Displays the scatterplot of the correlation pairs.
Prior and Posterior:
Displays the prior (dashed line) and posterior (solid line) distribution of the correlation under the alternative hypothesis; The horizontal solid line represents the width of the 95% credible interval of the posterior distribution.
Bayes factor robustness check:
Displays the Bayes factor as a function of the width of the beta prior on the correlation.
Sequential analysis:
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.