When formally stating a hypothesis, scientists call it an alternative hypothesis, or HA. This is a bit confusing at first because the alternative sounds like it should be the alternative to the hypothesis. Actually it is the alternative to something called the null hypothesis, or H0. The null hypothesis is, as its name suggests, the hypothesis that nothing is going on. If your alternative hypothesis is that Stellar’s Jays are larger than Grey Jays, the null hypothesis would be that the two species are the same size. If your alternative hypothesis is that larger tomato plants produce larger flowers, the null hypothesis would be that there is no relationship between the size of the plant and the size of the flower. Whatever you think is going on, the null hypothesis says it’s not.
The value of the null hypothesis is that it gives us a hypothesis that we can show to be false (at least with high probability). Remember that we can never prove a hypothesis is true. Let’s take the comparison between Stellar’s Jays and Grey Jays as an example. Suppose we weigh 50 birds of each species, and find that on average the Stellar’s Jays weigh about 420 g, while the Grey Jays weigh about 290 g. Have we proven our hypothesis? No. Every bird is somewhat different in size and it could be that just by chance we happened to get larger birds in our sample of Stellar’s Jays. We also certainly made some measuring errors. Perhaps they are responsible for the difference. It’s hard to ask: How likely is it the two species are really different in size? But it’s easy to get at the question from the other direction: How likely is it that the average weight of the birds we looked at would be as far apart as they are, if the two species were really the same weight ? (Remember this is our null hypothesis.) Statistical tests can accurately tell us this probability (if we know something about the variation in weight from bird to bird) and if the null hypothesis has only a small probability of being true, then that is support for the alternative. (For more on basic statistics go here.)