Why is my 95% UCL so high?
The size of your confidence interval of the mean is related to the spread of your data and sample size.
A higher GSD or a smaller sample size will result in a larger 95% UCL.
Therefore, if you want to have smaller 95% UCLs in your statistical analysis, either find ways to reduce the variability in exposures or collect more samples.
You can reduce exposure variability by:
Controlling exposures so they are more consistent.
Splitting your SEGs into smaller, more homogeneous groups.
If exposures are already well controlled, and SEGs well defined, sometimes the only necessary step to demonstrate compliance using confidence intervals is collecting a few more samples. Because of the relationship between sample size and UCL, going from 5 to 10 samples can have a huge difference, whereas going from 35 to 40 samples will change very little.
