Question 2: What methods could be used to calculate the confidence interval for the difference in mean alcohol consumption? Which should we use here and why?
We could use the large sample Normal method or the two sample t method.
The data clearly do not have a Normal distribution and the variance is not particularly uniform, which does not indicate the t method. The samples are both fairly large, however, both greater than 50. We can use the large sample Normal method.
Note that SPSS, like many statistical programs, does not do this explicitly. Instead, we use the two sample t method with unequal variances, which gives the same answer when the sample is large.
Back to Exercise: Alcohol consumption and sex.
To Applied Biostatistics index.
To Martin Bland's M.Sc. index.
To Martin Bland's home page.
This page maintained by Martin Bland.
Last updated: 31 October, 2006.
Back to top.