Confidence limits for non normal distribution
I'm new to this forum. I have a question which I hope someone can answer.
I'm analyzing data from a survey (opinion survey at a local Catholic parish). The possible answers for each question are: strongly agree (1), agree (2), undecided (3), disagree (4) and strongly disagree (5). Right now, I have about 550 data points. When all surveys have been entered, I will have about 750 data points.
One of the first thing I noticed is that there is a lot more variability in some answers than others. The standard deviation of the responses varies from .6 to 1.1.
I would like to calculate confidence intervals for each question, but I'm not sure how to do it. I don't believe the responses are normally distributed. The results are very skewed to positive responses. About 69% of the responses are "agree" or "strongly agree". If I calculate a 95% confidence interval for each question as: the mean answer +/- 1.96 standard deviations, I get a range for some questions that says the answer could be anything from strongly agree to strongly disagree.
I have some statistics background and I'm comfortable doing regression analysis, but this is way beyond my statistical knowledge.
Thanks in advance for your help.