Survey results

Our statistics class survey was not only a survey. It was an experiment.

You remember that each survey had a random number X at the top of it, `a random number between 0 and 100'. These numbers were actually chosen at random from the set {10,65}. Half of the surveys had `X=10', and half had `X=65'.

Please think about the following questions: (remember, the numbers X = 10 and X= 65 were assigned randomly)

• Do you think there should be any correlation between the value of X and the answer to question 2?
• Consider the question `what is the typical number of children that a woman has?'
  • Is the average answer to question 3 a good approximate answer to this question?
  • Is the average answer to question 4 a good approximate answer to this question?
• Do you think there should be any correlation between the answers to questions 3 and 4?
• Do you think the distribution of answers to questions 3 and 4 should be different?
• Do you expect the average of the answers to question 3 to be significantly different from the average of the answers to question 4?

Then...

Listen to a 'Yes Minister' play, that explains how opinion pollsters can get whatever answer they want to a question.

See the responses to the survey.