## How to test if the difference between probabilities is statistically significant

How do I test if the difference between probabilities is statistically significant. I know the population size and population variation. A simple t-test or ANOVA would work for variables, but how do I test for probabilities?

A completed a Monte Carlo simulation which returns a probability, $p$, given a set a variables, $x$. I need to know if the probabilities for different set of variables, are statistically different.

-
In about 4 hours, a new Stack Exchange site focusing on statistics will become public: stats.stackexchange.com I think your question will be better-suited there. – Adrian Petrescu Jul 26 2010 at 14:46
What do you mean by "test if the difference between probabilities is statistically significant" can you formulate that mathematically? Does testing the value of the parmeter of a Bernouilli would fit to your question? I think this question is not precise enough to be a real question. – robin girard Jul 26 2010 at 15:19
do you assume a parametric model and want to test if the two samples come from the same distribution inside this parametric family ? – Alekk Jul 26 2010 at 18:21