Confidence interval (CI) is the range around a study's result within which we would expect the true value to lie.
CIs account for the sampling error between the study population and the wider population the study is supposed to represent.

Centre for evidence-based medicine. Glossary. Accessed on June 19, 2016.