> I've read 25% of married Americans report having an affair (meaning .75 * .75 = 56% chance of a marriage not having an affair) Only if "married but having an

> I've read 25% of married Americans report having an affair (meaning .75 * .75 = 56% chance of a marriage not having an affair) Only if "married but having an affair" is an independent variable, which I would doubt -- I would be shocked if one partner having an affair weren't predictive of the other partner having an affair. Interesting to see the spread of numbers by different studies though; I wonder how close to reality any of them are.