# Pearson's Correlation, for comparing a PRNG?

Other than uniformity tests on random numbers of which can be done with other methods, I had wondered if the result of the Pearson Product Moment Correlation function would be an effective means to convey randomness, based on its purpose (of measuring correlation)

My two questions are,

Would this be a less relevant test of my random generations to compare to say an HRNG, than other statistical tests of randomness to show a potential client?

Would I be correct using the function in this manner, by generating two new arrays (of the same RNG method) for the function ρ(x, y) to show the correlation or is it used differently?

-

Computing Pearson's correlation is one of several ways to measure correlation (see wiki's article on correlation and dependence). There are other measures, like Spearman's rank correlation coefficient $\rho$ and Kendall's rank correlation coefficient $\tau$ among others.