Sorry if this is trivial - I'm arguing with someone on Facebook who is claiming that r-squared values, or correlation coefficients represents "chance" or "probability". I've never heard of this. I just need a simple yes or no answer along with a short explanation why it is or isn't probability. He says: "It is the probability that there is a connection between two different things. A Correlation Probability of 1 means that these two things ALWAYS occur together, even if there is absolutely no causal links between them. A Correlation Probability of ½ (or .5, or 50%) means that the two things are connected 50% of the time. Or, roughly what we would see in "Chance" for most things, depending upon what two things we are looking at." I say: "I think you have something wrong here. I don't know what a correlation probability is and I'm struggling to find it online. The ".55" value and other correlations in the literature are correlation coefficients, they are r-squared values calculated using a least squares regression analysis. R-squared estimates the fraction of the variance in IQ scores that is explained by % genetic relation - it has nothing to do with chance. Chance of randomness is given by the p value, calculated by a t test on the same data. The chance of randomness depends on the data sets you're attempting to correlate and the variances of the data points. I still don't actually understand where you've gotten 50% from." He says: "An R-Squared IS a PROBABILITY (Describing the Chance/Probability that two variables are correlated)." Is r-squared the same thing as probability? I've not learned it that way, I've learned it *only* as how much variance in one thing explains variance in another.