Let’s say I’m a researcher (yay, I am!) who is interested in finding out the effect of cigarette smoking on people’s health (not me, but let’s play pretend).
Let’s also say that because I have read lots and lots of other people’s research (which is the bare minimum for carrying out a study of this sort) I suspect - nay, hypothesise - that the effect of cigarette smoking is going to be detrimental to health.
I also, from my reading, hypothesise that these detrimental effects will probably only be seen after many years of smoking.
Do I,
A) Give a small group of teenagers who are desperate to smoke, two cigarettes per day for two years, and, at the end of the two years compare heath factors to those in two other groups - another group who are desperate to smoke, but who have only been given cigarettes for a year, and a group who have no interest in smoking and indeed have never smoked;
or
B) Collect health data on as many factors as I can (factors that will be defined by, what’s that? Oh yes, reading lots of other research) from as large a group as possible of adults who have smoked for a significant amount of time, and compare this to outcomes on the same factors for a large group of adults who have never smoked?
Additional question for bonus points: Given that my hypothesis is that smoking is going to have negative effects on health, do you think that an ethics committee should allow me to give two groups of children cigarettes at all?