Logo Utrecht University

Informative Hypotheses


BOCOR is an abbreviation of Bayesian testing of Order constraints on CORrelations. BOCOR can be used to compute Bayes factors for testing order constrained hypotheses on bivariate correlation coefficients, e.g., H1:rho(1,2)>rho(3,1)>rho(3,2)>0, where rho(p1,p2) denotes the bivariate correlation between the p1-th and p2-th dependent variable. An extension is under development for testing partial, polychoric, biserial and other types of correlations.

When using BOCOR for your research you can give credit to the developer by referring to

Mulder, J. (2016). Bayes factors for testing order-constrained hypotheses on correlations. Journal of Mathematical Psychology, 72, 104-115.

BOCOR is licensed under the GNU General Public License Version >=3. It is a beta version, so if you find a bug let us know.

BOCOR can be downloaded from www.jorismulder.com.