1 / 7

Power and Sample Size

Power and Sample Size. David M. Thompson, Ph.D., P.T. Associate Professor Dept. of Biostatistics and Epidemiology College of Public Health, OUHSC. Power and sample size calculations. POWER. SAMPLE SIZE. EFFECT SIZE. Specify or estimate any two to calculate the third. Power.

chayton
Télécharger la présentation

Power and Sample Size

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Power and Sample Size David M. Thompson, Ph.D., P.T. Associate ProfessorDept. of Biostatistics and EpidemiologyCollege of Public Health, OUHSC

  2. Power and sample size calculations POWER SAMPLE SIZE EFFECT SIZE Specify or estimate any two to calculate the third.

  3. Power • The probability of rejecting the null hypothesis when a specified alternative is true. • Both null and alternative hypotheses must be specified; the alternate hypothesis specifies a clinically meaningful effect size.

  4. Sample Size To calculate sample size, first must specify: •  • Power = (1-) • Effect size • Difference in means • Difference in proportions

  5. Effect size Meaningful and detectable difference in group means, proportions, etc. … …including estimate of within group variability

  6. In study planning: • Power is customarily prespecified at .8 or .9 • A minimal clinically important difference focuses the decision on the effect size, from which we calculate sample size. • If practical conditions constrain the sample size, we specify that, then calculate the effect size we have prespecified power to detect with that sample.

  7. Do it yourself resources Web primer on “hypothesis testing and statistical power” moon.ouhsc.edu/dthompso/CDM/power/hypoth.htm Java applets that explore influences of sample and effect sizes on power http://www.stat.uiowa.edu/~rlenth/Power/ Power and sample size programs (UCSF) http://www.biostat.ucsf.edu/sampsize.html

More Related