You are one of...current visitors on the English part - also ...current visitors on the Swedish part

(the number of current visitors is automatically updated every 4 minutes)

Cite this page as:

-

G*Power (ENG)

-

First published:

on:

INFOVOICE.SE

Last updated:

If you want to share information about this web page...

This web-page provides an introduction to the software G*Power used to estimate sample size. Reading this will give you enough knowledge to be able to try doing simple things in G*Power.

Recommended prereading are the pages Introduction to statistics, Choosing statistical analysis och Sample size estimation.

When performing inferential statistics, you have observations with a measure of central tendency and a known measure of dispersion around it. When running the observations through your chosen statistical method, you obtain an effect size and often also a p-value. When estimating sample size, you do all of this in reverse. You use the same statistical method but start from assumed values for the effect size, measure of central tendency, and measure of dispersion. You then obtain the answer to how many observations are needed.

G*Power is a computer program that can perform sample size estimation and estimate power. The program can handle most common statistical tests except Cox regression. G*Power is free and can be downloaded from the Dussedorf University website.

Example – Comparing two unmatched groups

Example 1 of sample size estimation to compare two unmatched groups – T-test and Mann-Whitneys test (Click to see)

Example 2 of sample size estimation to compare two unmatched groups – T-test and Mann-Whitneys test (Click to see)

Example – Logistisc regression

Example 3 – of sample size estimation for logistisc regression when the independent variable is binary (like sex) (Click to see)