Using G*Power for statistical power analysis
The main purpose underlying statistical power analysis is to help the researchers to determine the smallest sample size that is suitable to detect the effect of a given test at the desired level of significance.
Developed at the Institute for Experimental Psychology in Dusseldorf Germany, the free software G*Power was designed as a general stand-alone power analysis program for statistical tests commonly used in disciplines such as agriculture, animal studies, health science, social and behavioral research, etc.
Power analysis is often intentionally or unintentionally ignored because it involves several statistical concepts which may not be necessarily straight forward to understand and the power analysis formula is very model specific. G*Power provides an effective and very user-friendly solution for promoting the application of power analysis as part of the routine statistical data analysis procedure. It offers a wide variety of calculations related to power analysis along with graphics and protocol statement outputs. The actual teaching and learning time is five hours divided into three sessions.
Fee-based workshop
Quantitative Consulting Unit (QCU) runs this one-day hands-on practice workshop for learning power analysis using GPower 3.1 from time to time based on the request from the researchers (HDR students and staff members) at CSU. It is a fee-based workshop for the purpose that anyone who would register and pay for the workshop is most likely to commit him/her time and effort for the expected learning outcomes.
This is a fact-to-face workshop and we will run the workshop as sketched above with minimum two confirmed attendees. The upper limit of the number of attendees for each workshop is 10 for a better learning and practice result. We are flexible in running this one-day workshop multiple times and on different campus as long as the demand requires.
Workshop program
The actual teaching and learning time is five hours divided into three sessions.
- 9:30am – 10am: registration and morning tea
- 10am – 12pm: learning and practice session 1 – attendees install the GPower software onto their laptop computers; basic concepts of statistical power analysis will be introduced; a first touch with GPower; apply GPower to perform power analysis for five different statistical models.
- 12pm – 1pm: the lunch hour
- 1pm – 2:30pm: learning and practice session 2 – attendees continue to learn and practice power analysis using GPower for six more statistical models including the ANOVA model with more than two categorical factors.
- 2:30pm – 3pm: afternoon tea break
- 3pm - 4:30pm: learning and practice session 3 – attendees continue to learn and practice power analysis using GPower for five more statistical models including a simple repeated measure regression model case.
- 4:30pm: close of the workshop
The GPower user manual (both the hard copy and the electronic copy) will be prepared for the attendees for the workshop.
Cost
- CSU students and staff, and DPI staff: $100 (including GST) per person.
The cost includes morning tea, a light lunch, and afternoon tea.
Expressions of interest
Please send us your Expression of Interest:
- by email: qcu@csu.edu.au
- or talking to us about your request by phone: 02-69332229 (John).
Resources
You can download the G*Power software and access the manual and tutorial to help you get started.