Question 4: Why would a correlation coefficient be difficult to interpret for this graph?
The distribution of alcohol consumption is clearly positively skew and will have a long tail on the right.
We cannot estimate a confidence interval for r unless both variables follow Normal distributions, so it is difficult to interpret the value of a correlation coefficient.
The test of significance would be OK, provided BMI followed a distribution which was approximately Normal.
Back to Exercise: BMI and alcohol consumption in the 2005 questionnaire.
To Applied Biostatistics index.
To Martin Bland's M.Sc. index.
This page maintained by Martin Bland.
Last updated: 5 December, 2006.