Reviewed and revised 26 August 2015
Confidence intervals are the range around the sample mean within which you predict the means of the population to lie
- using a standard normal distribution curve, 95% of the sample means should lie within 1.96 SE above and below the population mean. This the 95% CI for the population mean.
- Standard Normal Distribution Curve: CI = sample mean +/- (z0.05 x SE)
- T-test: CI = sample +/- (t0.05 x SE)
CAUSES OF WIDE CONFIDENCE INTERVALS
- small sample
- large variance within sample
- precision of sample mean as an estimation of the population mean.
- more information on significance of p-value.
- allows clinical relevance to be made.
- allows for graphical representation between several means.
- if OR presented with CI and includes 1 -> there is no difference between the two groups.
- if p value presented with CI that includes 0 -> there is no difference between the two groups.