Key points at a glance
- A single Likert-type item is not automatically a multi-item scale.
- Specify coding and missing-response rules before calculating scores.
- Choose summaries that fit the question, measure and disciplinary conventions.
Distinguish an item from a composite scale
An individual item may offer five ordered responses from “strongly disagree” to “strongly agree”. A composite scale combines several items intended to represent a shared construct. The terms are often used loosely, but the difference matters when you analyse the answers.
Sullivan and Artino discuss how Likert-type responses can be interpreted. For a published instrument, examine its original guidance. Which items belong together, how are they coded and what does a high total score mean? Changes you make can affect comparability.
Check responses before calculating scores
Code the response categories in their intended order. Check whether some statements are phrased in the opposite direction and require reverse coding under the instrument rules. A single unreversed item can undermine a composite score.
Original example: Three invented usability statements receive responses from 1 to 5. “I find the interface difficult” points in the opposite direction from “The interface is easy to use.” If all scores should point the same way, explain and implement the recoding consistently.
Select a useful presentation
For an individual item, counts or proportions show how answers are distributed across categories. A median and range or quartiles can describe an ordered centre. Some disciplines also use means for individual items; explain the interpretation and assumptions if you do so.
A multi-item scale may specify a sum or mean score. This requires the selected items to belong together conceptually and the calculation rule to be clear. Follow the source instrument when using an established measure.
Also decide how many completed items are needed for a scale score. Missing does not mean zero. Report usable observations by item and scale so that differences between tables and charts remain understandable.
Interpret only what was measured
High agreement with a statement initially means high agreement with its particular wording in the sampled group. It does not establish actual behaviour or a property of all students. Tie each inference to the measurement and sample.
Report response anchors, score direction and aggregation. A chart without category labels leaves readers guessing whether a high number is favourable. Check that “neutral” and “no answer” were handled separately.
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
- Distinguish items from composite scales.
- Record anchors and score direction.
- Check reverse-coded items.
- Specify missing-response rules.
- Match summary and interpretation to the measure.
Common mistakes
- Calling every five-point question a validated scale.
- Averaging oppositely worded items without recoding.
- Coding “no answer” as a neutral response.
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
Can I calculate a mean for one Likert-type item?
Practices differ by discipline and purpose. At least show the response distribution and explain what a mean of coded categories represents and which spacing assumptions you are making.
How do I calculate a composite score?
Follow the documented rule for an established instrument. For your own scale, justify the items, direction, calculation and required number of valid answers.
Does high internal consistency establish validity?
No. Consistent answers alone do not show that the scale measures the intended construct. You need further conceptual and empirical support.