I think one thing that's important to note is that they don't show the actual data in this link, and in the summary they seem to be including using a very broad definition of cheating that includes "emotional" infidelity as well as things like having a sexually explicit conversation with a stranger online. I'm not saying those things shouldn't be included, however, it's definitely something to keep in mind.
I've seen the actual study before and one thing it definitely DOESNT say is that everyone is having affairs. The levels vary, it's hard to tell how much actual betrayal is involved. And precisely how the respondents perceive both their actions and their responses within the study.