Difficulty can arise when we wish to define what is 'desirable' and 'fun' in a usability lab setting.
Desirable can be defined as: Worth having or seeking out, as by being useful, advantageous or pleasing or you can replace with words such as: fun, engaging, natural, enjoyable, valuable, essential, suitable, worthwhile, beneficial, likeable, in demand, amusing and appealling.
Formal usability testing is great at working out whether users can complete tasks, however the technique is less effective when measuring intangible aspects such as desirability.
One approach is to set a questionnaire with Likert scales, but as these questions are often set by the UX practitioner, they mean less to the user.
Another approach is the interview, and whle this approach can reveal some interesting data, some candidates struggle with giving honest answersor particularly negative feedback. In addition this approach can be time consuming and the resulting data can be difficult to evaluate and translate into actionable points.
A third approach is the brainstorm session which can be augmented with word cards that detail reaction. This helps users quantify their feedback without the need to generate words or rely on an ambiguous questionnaire. As there is usually a positive bias in feedback (users don't like to be negative) the word cards need to be skewed towards more negative words.
The cards chosen can then be ranked according to times chosen and the resulting qualitative data displayed using a chart.
At best desirabllity testing is a qualitative excercise and analysis of users reactions need to match this form of enquiry. Attaching any quantitative overlay is often spurious and not particularly helpful going forward.