One of the doctors on your floor tells you that he's been looking through some patient data and found a very strong positive correlation between a patient's migraine frequency and the age at which they had chickenpox. You're a bit dubious but intrigued. The doctor knows that you have more statistics know-how than they do, so he sends you the data set and asks if you can come up with the same strong correlation. You're doubtful, but you construct a scatter plot and, as expected, there appears to be absolutely no relation whatsoever. You return to the doctor to tell him your results and ask him to explain how he came up with the strong correlation.

He tells you that whenever he creates a scatter plot for a pair of variables, he first sorts the data in each column from lowest to highest. That usually results in the strongest correlations. You, aghast at this response, think about how to respond.

please answer the following:

1. What is wrong with the doctor's procedure?
2. Why did the doctor achieve such a strong correlation in this way?
3. How would you explain to the doctor his mistake?