My high school earth/space science teacher had just assigned a report. He said if we were having trouble choosing a topic to think about a scientific question we wanted to answer.
He told us, "In one of the other classes, someone asked, 'Why are clouds white?' The whole class stopped, and people started asking themselves, 'Yeah, why are clouds white?' I happen to know why clouds are white."
A student raised his hand and asked, "So, why are clouds white?"
The teacher responded, "I'm not going to tell you. When she finishes writing it, you can read her report."
Admittedly, this tactic was more effective then, during the mid '80s, in pre-Internet search engine times. Thinking about a question you want to answer is still a good way to choose a topic for a report, however.