It is worth noting, scientists never talk about their hypothesis being "right" or "wrong." Instead, they say that their data "supports" or "does not support" their hypothesis. This goes back to the point that nature is complex—so complex that it takes more than a single experiment to figure it all out because a single experiment could give you misleading data. For example, let us say that you hypothesize that earthworms do not exist in places that have very cold winters because it is too cold for them to survive. You then predict that you will find earthworms in the dirt in Florida, which has warm winters, but not Alaska, which has cold winters. When you go and dig a 3-foot by 3-foot-wide and 1-foot-deep hole in the dirt in those two states, you discover Floridian earthworms, but not Alaskan ones. So, was your hypothesis right? Well, your data "supported" your hypothesis, but your experiment did not cover that much ground. Can you really be sure there are no earthworms in Alaska? No. Which is why scientists only support (or not) their hypothesis with data, rather than proving them. And for the curious, yes there are earthworms in Alaska .

In statistical hypothesis testing, two hypotheses are compared. These are called the null hypothesis and the alternative hypothesis . The null hypothesis is the hypothesis that states that there is no relation between the phenomena whose relation is under investigation, or at least not of the form given by the alternative hypothesis. The alternative hypothesis, as the name suggests, is the alternative to the null hypothesis: it states that there * is* some kind of relation. The alternative hypothesis may take several forms, depending on the nature of the hypothesized relation; in particular, it can be two-sided (for example: there is * some* effect, in a yet unknown direction) or one-sided (the direction of the hypothesized relation, positive or negative, is fixed in advance). [23]