The module expects two columns of directional (0-360) or orientational (0-180) data in degrees. For mathematical details, see the Past manual.
Watson-Williams test
The Watson-Williams test for equal mean angle in two samples is a parametric test, assuming von Mises distribution, but is fairly robust. The concentration parameter kappa should be larger than 1.0 for accurate testing. In addition, the test assumes similar angular variances (Rayleigh's spread R values). The significance for the test statistics U is computed by first correcting U according to Mardia (1972a). The p value is then given by the F distribution. The combined concentration parameter kappa is maximum-likelihood.
Mardia-Watson-Wheeler test
This non-parametric test for equal distribution is computed according to Mardia (1972b). For n>14, W is approximately chi-squared with 2 degrees of freedom.
References
Mardia, K.V. 1972a. Statistics of directional data. Academic Press, London.
Mardia, K.V. 1972b. A multi-sample uniform scores test on a circle and its parametric competitor. Journal of the Royal Statistical Society Series B 34:102-113.