Key points at a glance
- Justify the interaction conceptually.
- Read main terms at reference values.
- Plot conditional relationships.
State the question before adding a product
An interaction means the association between X and the outcome varies with Z. Do not add a product term just to find an interesting p-value. Identify substantively comparable groups or values with enough observed data.
Interpret coefficients conditionally
Example: A grade model includes study time, course mode and study time × mode. The study-time coefficient describes its association in the reference mode; the interaction shows how that association changes in the other mode. It is not a standalone “effect of two variables”.
Show interpretable predictions
Plot modelled values at sensible study times for both groups and display uncertainty. Avoid extrapolation outside observed ranges. With a continuous moderator, centring and named reference values can aid readability.
Check model limits
Report coding, reference group, main terms, interaction and model assumptions. A main-term coefficient applies at its reference value in an interaction model. Observed moderation alone does not prove causation.
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
- Hypothesis stated.
- Reference values defined.
- Conditional predictions checked.
- Observed range respected.
Common mistakes
- Adding a product without rationale.
- Treating a main term as constant everywhere.
- Interpreting beyond observed values.
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
Is interaction just correlation?
No. It describes how one modelled relationship changes over another variable.
Should I remove main terms?
They usually remain with a product term; justify exceptions.
How should I present it?
Use conditional estimates or a clearly labelled interaction plot.