Hostname: page-component-8448b6f56d-42gr6 Total loading time: 0 Render date: 2024-04-24T12:00:28.526Z Has data issue: false hasContentIssue false
  • English
  • Français

The Merits of Confidence Intervals Relative to Hypothesis Testing

Published online by Cambridge University Press:  21 June 2016

Leon F. Burmeister ,
David Bimbaum  and
Samuel B. Sheps
David Bimbaum
Affiliation:
Applied Epidemiology, Sidney, British Columbia
Samuel B. Sheps*
Affiliation:
University of British Columbia, Vancouver, British Columbia
*
Department of Health Care and Epidemiology, 5804 Fairview Ave., Vancouver, BC, V6T 1Z3
Get access Rights & Permissions [Opens in a new window]

Extract

A variety of statistical tests of a null hypothesis commonly are used in biomedical studies. While these tests are the mainstay for justifying inferences drawn from data, they have important limitations. This report discusses the relative merits of two different approaches to data analysis and display, and recommends the use of confidence intervals rather than classic hypothesis testing.

Formulae for a confidence interval surrounding the point estimate of an average value take the form: d= ±zσ/√n, where “d” represents the average difference between central and extreme values, “z” is derived from the density function of a known distribution, and “a/-∨n” represents the magnitude of sampling variability. Transposition of terms yields the familiar formula for hypothesis testing of normally distributed data (without applying the finite population correction factor): z = d/(σ/√n).

Type
Statistics for Hospital Epidemiology
Information
Infection Control & Hospital Epidemiology , Volume 13 , Issue 9 , September 1992 , pp. 553 - 555
Copyright
Copyright © The Society for Healthcare Epidemiology of America 1992

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1. Birnbaum, D, Sheps, SB. Validation of new tests. Infect Control Hosp Epidemiol. 1991;12:622–624.CrossRefGoogle ScholarPubMed
2. Burmeister, LF Proving the null hypothesis. Infect Control Hosp Epidemiol. 1992;13:55–57.CrossRefGoogle ScholarPubMed
3. Sheps, SB, Bimbaum, D. Aspects of truth, statistics, bias, and confounding. Infect Control Hosp Epidemiol. 1992;13:418–420.CrossRefGoogle ScholarPubMed
4. Rothman, KJ. Modern Epidemiology. Boston, Mass: Little, Brown and Co.; 1986.Google Scholar
5. Lilliefors, HW. On the Kolmogorov-Smimov test for normality with means and variance unknown. J Am Stat Assoc. 1967,64:399–402.CrossRefGoogle Scholar
6. Freiman, JA, Chalmers, TC, Smith, H, et al. The importance of beta, the Type II error and sample size in the design and interpretation of the randomized control trial. Survey of 71 “negative” trials. N Engl J Med. 1978;299:690–694.CrossRefGoogle ScholarPubMed
7. McGill, R, Tukey, JW, Larsen, WA Variations of box plots. The American Statistician. 1978;32:12–16.Google Scholar
8. Johnson, RA, Wichem, DW. Applied Multivariate Statistical Analysis. 2nd ed. Englewood Cliffs, NJ: Prentice-Hall; 1988.Google Scholar
9. Holland, BS, Copenhaver, MD. An improved sequentially rejective Bonferroni test procedure. Biometrics. 1987;43:417–423.CrossRefGoogle Scholar