Sampling Bias Quotes

It always cracks me up when program directors or music directors or companies will say, 'Well, we did research, and we interviewed 25 people in our focus group, and this is what they said.' And I'm like, 'I've talked to 25 people in two hours! I talk to 50, 60, 70 people a night! Five or six days a week!'

Sampling, statisticians have told us, is a much more effective way of getting a good census.

If you aren't taking a representative sample, you won't get a representative snapshot.

The opinions of a convenience sample... may not represent all users.

One should be weary of drawing too many inferences from a single poll. You can find wildly disparate results with two different polls. There is no question that American Muslims remain one of the most marginalized and demonized groups in United States. There has been a sustained propaganda campaign against Muslims for over a decade and it doesn't disappear over night. Those attitudes are hardened. But one of the things that polling often doesn't measure is the intensity of opinions.

Exit polls have gone notoriously wrong in the past.

Let us suppose that an ichthyologist is exploring the life of the ocean. He casts a net into the water and brings up a fishy assortment. Surveying his catch, he proceeds in the usual manner of a scientist to systematise what it reveals. He arrives at two generalisations: No sea-creature is less than two inches long. (2) All sea-creatures have gills. These are both true of his catch, and he assumes tentatively that they will remain true however often he repeats it.

You can't just take and sample size a few hundred people and decide that what you think 200 - almost 200 million people are going to vote and go do. It's a complex country.

It is impossible, by the way, when picking one example of anything, to avoid picking one which is atypical in some sense.

The problem is that when polls are wrong, they tend to be wrong in the same direction. If they miss in New Hampshire, for instance, they all miss on the same mistake.

One of many problems with survey research in general is that you can only survey the survivors. In other words, if you were to do a survey of people who were known to have played Russian Roulette and you sent out the questions before the time they were going to play and then you come back six months after they played Russian Roulette, you would probably discover that among the people who did come back there was no harm done.

I wanted us to be careful about, going to the corner diner, interviewing three people and saying, "here's the mood of the public."