In testing specific hypotheses, single-value estimates are used in describing tendencies and variability of data. Due to the nature of sampling, such estimates may be insufficient in representing the uncertainty associated with them as representations of population parameters. Alternatively, confidence intervals can give a possible range of estimates for the population parameter of interest.


Confidence intervals rely on a critical value from an estimate’s distribution that identifies the cutoff with a proportion of the distribution equivalent to the Type I error rate, or probability of a false positive, with which a researcher is willing to make a false positive. The general form of a confidence interval is given by the estimate of interest minus (for the lower bound) and plus (for the upper bound) the product ...

  • Loading...
locked icon

Sign in to access this content

Get a 30 day FREE TRIAL

  • Watch videos from a variety of sources bringing classroom topics to life
  • Read modern, diverse business cases
  • Explore hundreds of books and reference titles