So let's think about doing an experiment on meningicocci. First of all, let's think about where we should do the experiment:
In the lab...
In real people...
And how many different conditions do we want to test?
Here is where a lot of beginners (and also quite a few people who should know better) dig themselves into a really deep hole... Let's listen in:
Student A: Well, I think the main difference is that Frank's blood has more iron. I say we do one dish with iron, and one without. And maybe a third dish with a super-high iron level, just to be safe.
Student B: No way, I think the difference is that Frank's blood has sugar in it, and his nose doesn't.
Student A: Ewwww.
Student B: So let's make 3 dishes, with zero, medium, and high levels of sugar.
Student C: You're both wrong. I bet its the nasal spray Frank's been using. Let's try one dish with spray and one without.
Student A: Iron!
Student B: Sugar!
Student C: Spray!
Student D: Arrrgh. You've been sitting here arguing all day and lab is over in 2 minutes. How about we just do them all.
Student A: Sure, why not?
The problem is, not only have students A-C proposed 3+3+2 = 8 different treatments, but they would also need to think about how the treatments affected each other. So, they will need
a high sugar-high iron dish, and
a high sugar-low iron dish, and
a high sugar-zero iron dish, and
a mediumn sugar-high iron dish, and ...
Altogether they will have
3*3*2 = 18
different dishes. Just imagine what would happen in Student D had yet another idea involving 3 or 4 treatments!
Remember the deep hole I was talking about? That comes when you are counting 18 dishes and your classmates are only counting 2. So, keep your life simple, think carefully about what you want to test, then test only one question.