An exit poll asks what you have just done. Simple, right? Let’s do one in your local supermarket parking lot. Put aside such matters as that the participants have must have cars, shop at grocery stores etc. Just assume that, but put it in your commentary. Instead, we proceed with our questions.

Ask your respondent to pick their weight as being above or below healthy. Then, ask your respondent to characterize their purchases as being “fresh and healthy, consisting of fruits, vegetables and fish” or, “easy and convenient”, consisting of “prepared microwavable and snacks”

If you then checked their register tape, do you think what they did, and what they said match? The difference between an exit and an entrance poll may be great, but which should be the more accurate? I hope you said exit. An exit poll requires that the sample be sufficient, honest and be able to remember. Oddly, even in consumer research, what people say, vs the register receipts differs. If people feel a difference between their action and their own self image, they will lie to a stranger. If they know what they are supposed to say, they will say that.