A single Likert item is ordinal, and your examiner knows it

Here is a trap almost every dissertation walks past: a questionnaire item scored 1 to 5 looks numeric, so it gets a mean, an SD, and a t-test. But a single Likert item is ordinal, the gap between 'agree' and 'strongly agree' is not guaranteed to equal the gap between 'neutral' and 'agree'.

So are Likert scales off-limits for parametric tests?

No, and this is the nuance worth learning properly. A summed multi-item scale (say, a 10-item anxiety questionnaire scored 10–50) is conventionally treated as interval in psychology. That convention is pragmatic and defensible (Field, 2018), and departments rely on it. The slip is treating a single item the same way.

The sentence to say in your viva

“Individual items are ordinal, but the summed scale score is treated as interval, consistent with standard practice in psychological measurement.”

One sentence, said calmly, converts a trap question into evidence that you understand your own measurement. That is the entire game: not memorising rules, but being able to narrate why your choices were reasonable.

If you want to drill this until it is automatic, it is Lesson 2 of Unit 1 in the course, the free unit.