This is probably one of the bigger examples of correlation not equating to causation I have read today. There are a metric ton of factors here that haven't been factored in, like education. There's just not enough data to support such a conclusion. Or deny it. There's no real way to tell based on the data you linked. Hell, let's do a quick google search of sorts: South Africa has a population of ~49 million, and the United States has a population of roughly ~300 million as of 2009, for simplicity's sake. According to Avert (On US statistics), there's an estimated 1 million or so with the virus. This obviously isn't exact, and it could be a lot higher or lower. That's roughly an infection rate of 3% Now, as for information on South Africa, a short and simple summary is that in 2008, a national survey was conducted where roughly 90% of households involved took part, 89% of that agreed to answer the survey, and 64% agreed to take an HIV blood test. If we multiply those percentages, we get a rough picture of 50% of the population, and their infection rate was much higher, going as high as ~27% in the 31-35 age group. It decreased below that for other ages, but it never quite dipped below the 3% estimate in the United States. There is some obvious variance for the SA survey, as we aren't sure about the other 50% relatively. The rate could potentially be half or double. The conclusion of sorts: Circumcision wasn't factored in, but I'm not sure if that's as large of a factor as a lack of education on the matter. As for the higher rate in the United States, that honor actually goes to South Africa assuming infection rates stay roughly close to the same as they do now. And besides, South Africa does have a higher infection rate in males AND females. Huh, I might have accidentally verified your point instead of disproving it. But education is a much larger factor than penis skin, and there's no way to tell if circumcision affects HIV rates. Not ethically anyway.