By following the logic of the experimental method, we can discover cause-and-effect relationships. This allows us to predict what will happen under the same conditions in the future.
For example, suppose we want to test the hypothesis that a smoking awareness programme for 11 year olds causes them not to take up smoking by age 14. We would need two identical groups of 11 year olds. For example, they would all need to be non-smokers, and perhaps the same gender, class, etc. We then divide them randomly into an experimental group, who are given the programme, and a control group, who are not.
If by age 14 significantly fewer of the experimental group had become smokers compared with the control group, we might conclude that we had discovered a causal relationship, namely that the effect of the programme was to discourage smoking. Or perhaps this is showing just a correlation – would we need a lot more research evidence to claim causality?
Now answer the following questions:
- Apart from gender and social class, are there any other possible variables that might affect the results of the experiment? Would it be possible by selection of participants to control for any of these variables?
- There is a gap of three years before the effect of the programme can be assessed. Does this matter and, if so, why?
- If the results show differences between the two groups in taking up smoking, can we assume a causal relationship between the programme and smoking levels?
- Are there any other variables that might affect who does not become a smoker, other than exposure to the programme?