Dear all, sorry if this is a dumb question but I haven't found the answer in textbooks yet: I have a stratified sample for which I have counts for each stratum, to be combined into a total value. As this is a sample I would like to calculate the 95% confidence intervals for my total value. I know it's straightforward to calculate confidence intervals for each stratum, but intuitively it seems like the upper and lower confidence limits of the total will be too conservative if I use e.g. the lower 95% limit for each stratum to calculate the lower limit for the total. Any advice most welcome. Richard.