I'm also new to AP Stats and I'd like to add on to this question...
If someone says:
According to the model, an R-squared of .86 means that 86% of the variation in y can be accounted for by x.
Is that OK? The language in my textbook and the AP review books seems inconsistent on the best way to phrase this and I wouldn't want my students to get penalized on the exam.
Also, does anyone have any examples with data that illustrate an extremely high R-squared (like 99%), but nonlinear data? Or a very low R-squared (like 10%), but a useful model? We discussed this issue in class and one of my students was skeptical that these situations could occur with real data.