Key points at a glance
- An effect measure describes the magnitude of a difference or association.
- A confidence interval expresses uncertainty under the procedure’s assumptions.
- A p-value does not measure effect size or practical importance.
Ask three different questions about a result
A statistical report should answer more than whether a result is significant. Ask how large the estimated difference or association is, how precisely it has been estimated and what that magnitude would mean in the research context.
The American Statistical Association emphasises that a p-value does not measure effect size or practical importance. Nor is it the probability that the null hypothesis is true. A threshold alone therefore cannot provide a complete interpretation.
Choose an appropriate effect measure
Match the measure to your question, scale and analysis. A mean difference retains the measurement unit, while a standardised measure relates a difference to a measure of variability. Ratios and correlation coefficients answer different questions again.
Explain direction and reference. Which group was subtracted from which? Does a higher score represent a favourable or unfavourable outcome? A correctly calculated estimate can still be unclear without this information. Avoid universal size labels that lack disciplinary context.
Read a confidence interval accurately
A frequentist 95% confidence interval is produced by a procedure that, under its assumptions and repeated sampling, covers the fixed target parameter in 95% of cases. It is not a range containing 95% of individual observations.
For a particular result, consider the interval’s location and width alongside the model assumptions. A wide interval indicates limited precision. It does not establish that every included value is equally plausible or equally important in practice.
An original example of a clear report
Invented example: Two groups differ by an estimated 4 scale points, with a 95% confidence interval from 1 to 7 points. You can describe direction and magnitude, but you must explain what a point means on that scale. Without this context, practical importance is difficult to assess.
If an interval includes both negligible and substantively important values, discuss that uncertainty. If an interval for a difference includes zero, failure to establish an effect is not proof that no effect exists.
Report the measure, estimate, confidence level and relevant sample size. Identify the calculation method where necessary for understanding. Check that the table and prose describe the same comparison in the same direction.
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
- Match the effect measure to the question.
- Explain units, direction and reference.
- Identify the confidence level and procedure.
- Discuss practical importance separately from significance.
- Align text, tables and sample sizes.
Common mistakes
- Interpreting a small p-value as a large effect.
- Reading a confidence interval as the distribution of individual values.
- Treating a non-significant result as proof of equality.
Print and bind your finished thesis
Once you have checked the content and final PDF, configure the printed copies to match your submission requirements.
Configure printing and bindingRelated content
Explore related guidance on thesis structure, sources, formatting and final submission.
Frequently asked questions
Should I always use standardised effect sizes?
No. An unstandardised measure in familiar units may be easier to interpret. The choice depends on your question, model and disciplinary conventions; explain it clearly.
Does a wide interval make the study worthless?
No. It indicates limited precision under the procedure used. Report that openly and avoid conclusions that require a more precise estimate.
Can I omit the p-value?
That depends on your analysis plan and reporting requirements. Aim for complete and consistent reporting. Do not select statistics afterwards according to which best supports your preferred conclusion.