In one of my classes this week I was explaining the scientific method to my students.
"The first step is observation, which is basically just that you notice something and start thinking about the reason why that's what you see. Then you develop a hypothesis that might explain your observations, and make predictions based on that hypothesis. Next you design an experiment to test the hypothesis, you analyze the results of the experiment, and draw conclusions based on those results. You can determine if the results support the hypothesis, and if they don't, you come up with an alternative hypothesis and test that."
Then I gave them an example of something that I'd been talking about in my other classes this week: sexual selection in guppies.
"Male guppies have colored spots all over their bodies, which turns out to be a way that females choose which males to mate with. People have done experiments to show that, but it all started with someone making the observation that males are brightly colored and females aren't. Then they came up with a hypothesis about it being related to female mate choice, and designed a set of experiments to test that. Think about it. You could basically set up a tank with a female and two different males, one brightly colored one and one dull one, and sit and watch to see which one the female mated with. It'd be like watching fish porn. But that's probably not the kind of stuff you want to go look up on YouTube."
I probably should have left the last two sentences unspoken.