StaTips Part VIII: Confidence interval for the sample mean

  • Giuseppe Perinetti University of Trieste


When conducting research on a given type of patients, it is impossible to examine all the existing subjects of that type (population)
to derive the true mean of the parameter of interest. More realistically, by the investigation of a small group of subjects (sample) from
the whole population, researchers can estimate an interval into which the true mean of the population lies. In statistics, such interval is
referred to as confidence interval (CI). The calculation of the CI from a sample mean is simple and gives important information, not only
regarding the true mean of the population, but also on the statistical significance of the difference between groups being compared. For
these reasons, the reporting of the CIs is preferred over the p value alone.


Glantz S. Primer of Biostatistics. 7th ed. Columbus: McGraw-Hill Education; 2011.
Short Communication