There isn't any evidence to back that claim: correlations may suggest a causation, but they certainly do not prove that there is a causation. Frankly, I think that the thesis of Windows 8 hurting PC sales is absurd. How many people are going to say, "my computer's slow, but I'm not going to buy a faster computer with a faster operating system because the Start Screen is too confusing"? In the more extreme case, how many people are going to say, "my computer's broken, but I'm going to go without a computer because I don't like Windows 8?" I would suggest that very few people would fit into either category. There are far more likely hypotheses like tablets pillaging PC sales. Both of theories are opinions because neither one of us has proof to our claims. One of us may be right, both of us may be right, or both of us may be wrong.