G*Power for nursing research is a software workflow: choose the procedure that matches your agreed plan, enter its inputs and retain a record of the output. This guide concentrates on the settings and checks inside that workflow, using the application’s documented approach (Faul et al., 2007; Faul et al., 2009).
Establish your recruitment rationale with the nursing sample-size planning guide. Use the separate statistical power article for the meaning and limitations of power. Have those decisions ready before entering values here.
Prepare a software input record
Create a blank record with these labels: software version, test family, statistical test, analysis type, input values, output values and saved-file location. Copy labels as they appear in the procedure you actually use. This avoids a methods paragraph referring to one test while the retained screen shows another.
Check the test family and exact procedure
Begin with the name of the planned statistical test. Then locate the corresponding family and exact procedure. Check distinctions such as independent versus paired observations and the particular regression comparison. Do not choose a menu item simply because it contains a familiar word from your topic.
If the procedure does not represent the design, stop and resolve that mismatch. Selecting the nearest-looking option and entering more precise numbers will not make the model suitable.
Select the analysis type before entering values
The analysis type determines which quantity the software is asked to return. Check that this matches your task before completing the displayed fields. For the methodological distinction, see a priori, post hoc and sensitivity analysis in nursing.
Audit the fields currently displayed
Work from the agreed statistical plan and record each displayed input. Check the effect-size label rather than assuming every procedure uses the same measure. Where the procedure asks for allocation, predictors or repeated-measure inputs, those fields need their own justification.
After changing a test or analysis type, inspect the input panel again. A useful practical rule is to read each field name aloud against your planning record before selecting Calculate. Keep a note of any unresolved input rather than silently accepting a convenient value.
Read the returned output selectively
Identify the quantity requested by the selected analysis type. Check any reported total and group-specific numbers, and retain the other output relevant to that procedure. Do not copy a number into your dissertation without its label and the settings that produced it.
Build a reproducibility check
| Saved evidence | Question to check |
|---|---|
| Input panel | Are all entered values visible and linked to the agreed plan? |
| Procedure labels | Do the family, exact test and analysis type agree? |
| Output panel | Is the reported quantity labelled correctly? |
| Methods draft | Does the written account describe this actual calculation? |
| Version record | Can another reader identify the software used? |
Troubleshoot a mismatch before reporting
If a repeated calculation returns a different result, compare the saved settings field by field. Check the procedure and analysis type first, then the entered values. Record the reason for any change so the final output can be distinguished from an earlier trial.
Keep exploratory scenarios separate from the final calculation record. A comparison across plausible inputs is useful only when the assumptions are labelled; it should not conceal how the final settings were chosen.
Frequently asked questions
Which version should I name?
Name the version actually used and keep the associated output.
Does a screenshot replace a written explanation?
No. Explain the procedure and assumptions in the methods section; use the saved screen as supporting evidence where appropriate.
What if my design is not represented?
Seek advice on a suitable calculation approach instead of forcing the study into an unrelated procedure.
Get feedback on an existing calculation
For a project-specific review, send the planned test, input rationale and saved output through sample-size and power analysis support. Explain which setting or result you need checked.
References
- Faul, F., Erdfelder, E., Lang, A.-G. and Buchner, A. (2007) ‘G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences’, Behavior Research Methods, 39, pp. 175–191.
- Faul, F., Erdfelder, E., Buchner, A. and Lang, A.-G. (2009) ‘Statistical power analyses using G*Power 3.1’, Behavior Research Methods, 41, pp. 1149–1160.
- Kang, H. (2021) ‘Sample size determination and power analysis using the G*Power software’, Journal of Educational Evaluation for Health Professions, 18, 17.
- Lakens, D. (2022) ‘Sample size justification’, Collabra: Psychology, 8(1), 33267.
