Experimental design (2)

Scenario experiment is better than self-report, but scenario experiment is not as good as the neuroscience study, which probably uses different devices to assess people. But scenario experiment is very good for us particularly for students, if you don’t have some good resources, if you don’t have some partners to work with you for that scientific study. I think scenario will be a good way, because this is the thing that can reduce the bias from memory lapses. Why we always think self-report is not good? For example, please recall the last six months, recall your recent visit or stay at the hotel. I recall yesterday, some people recall three months ago, some people recall six months ago, but we put all data together.

And the hotel I stayed at is different from the hotel he or she stayed at, I stayed at a luxury hotel and he stayed at a motel, so the experience probably will be totally different. Always by memory recall you, recall the memory, even I stayed at hotel yesterday, today I moved to another hotel, that hotel is much better than yesterday. So I will not give comments on the hotel I stayed yesterday, or even other way around. So that memory recall is a very challenging part for trying to get a true value of responses, which is particularly we call the measurement error. What you measure is not true value.

Scenario, at least when you read the scenario or you had the virtual reality test, maybe you immediately respond. And every group is facing similar situation, not like the above case. I stay at the five-star hotel, the other people stay at two-star hotel, but we respond the same question. So I think that’s really a good way to do research. I have quite many students who did the scenario-based study, and tried to get response from the participants.