Contact us

INFORMATION REPOSITORY

Introduction to G*Power

Updated on January 10, 2025
In this brief tutorial we will address the most critical functions of the G*Power program.

Learning Goals #

  • Understand the advantages of programming compared to spreadsheet operators.
  • Conduct simple calculations in MATLAB.
  • Work with variables and matrixes. 

1. What to use it for #

Power analysis can be used to evaluate statistical tests. For analytical chemists, the most useful functions are to calculate the required number of repetitions to achieve a specific statistical power. It can also be used to calculate the statistical power from a given hypothesis test.

Read more about power analysis here.

2. The program #

The program G*Power offers a graphical user interface to perform power analysis on statistical tests. The concept of power analysis has been extensively covered in Lesson 6 of our Chemometrics & Statistics course.

G*Power is developed and maintained by the Heinrich Heine Universität Düsseldorf. The program is freely available for both personal and commercial use.

G*Power

The program is created and maintained by the Heinrich Heine Universität Düsseldorf. It is freely available for both personal and commercial use.

Download

3. User interface #

The G*Power user interface is simple and straightforward. Figure 1 features various bits of information about the different interface components.

Figure 1. Overview of the G*Power interface. Note that depending on your operation system and system settings the window may look slightly differently.

There is an additional side panel for the calculation of the effect size that is shown in Figure 2.

Figure 2. Effect size calculation interface.

4. Types of power analysis #

There are four types of power analysis, depending on which you want to solve for.

  1. A priori: Compute the sample size, when designing a study. This provides you the required sample size to detect some level of effect with p-values
  2. Post-hoc: Computer the statistical power when completing a study. This tells you if you had sufficient subjects to detect the actual effect you found.
  3. Criterion: Compute alpha. Obviously provides you the probability of a type-I error. We rarely run this. We’ll discuss soon why.
  4. Sensitivity: Compute effect size when the sample size is predetermined by study constrains. For example, there are only 10 subjects available. With this analysis, we can see what level of effect we can still find with our subjects.
Is this article useful?
Table of contents