Key points at a glance
- Define related tests.
- Choose error control in advance.
- Report raw and adjusted results clearly.
Understand the multiple-testing issue
Testing many comparisons for the same question increases the chance of at least one accidental positive result. Define which tests form a family. Not every statistic in your results necessarily belongs to one giant family.
Apply Bonferroni transparently
Example: With four prespecified related comparisons and family alpha 0.05, the Bonferroni threshold is 0.0125 per test. Alternatively, adjust p-values. State which approach and how many tests you used; defining the family after seeing results can mislead.
Understand the trade-off
Bonferroni controls the chance of any false positive within the family but can reduce power when many tests are run. Other prespecified methods may fit a particular design better. Discuss complex choices with a statistics supervisor.
Report more than a verdict
Give the number of comparisons, raw and adjusted p-values where relevant, effect estimates and uncertainty. A nonsignificant adjusted result does not establish no effect. Label exploratory analyses honestly.
Final submission context
Formal details can feel like a separate writing task, but they matter just as much for a printed submission. Anything missing, misplaced or inconsistently formatted in the document will also appear in the bound copy.
Use this guide together with your cover page, table of contents, page numbers, source notes and appendices. Prepare the final PDF for printing and binding only after the complete file has been checked.
Practical check before PDF export
- Do headings, chapter structure and page numbers match?
- Are sources, figures and tables included completely?
- Are required elements such as declarations, appendices or the cover page included where required?
- Did you open and check the final PDF after exporting it?
Checklist
- Defined test family.
- Justified method.
- Calculated threshold.
- Reported effect estimates.
Common mistakes
- Defining family after results.
- Mixing raw and adjusted p-values.
- Interpreting nonsignificance as equality.
Prepare the finished thesis for submission
Once the content and university requirements are checked, review your PDF and the available printing and binding options.
Review printing optionsRelated content
Explore related guidance on thesis structure, sources, formatting and final submission.
Frequently asked questions
How is the threshold calculated?
Divide family alpha by the number of tests in that family.
Is correction always required?
It depends on the question, plan and interpretation.
Is Bonferroni conservative?
Yes, especially with many or dependent tests.