UMUX, the Usability Metric for User Experience, is a short questionnaire for perceived usability. The original instrument has four items on a seven-point agreement scale. UMUX-Lite is a related two-item version focused on perceived capability and ease of use. Their brevity can help when a longer questionnaire would impose too much burden, but the versions and scoring methods need to remain distinguishable.

A short instrument still measures a reported judgement. It does not establish whether people completed their tasks, understood the consequences or encountered accessibility barriers. Those questions need relevant behavioural and contextual evidence alongside the score.

Choose the version before planning comparisons

The original UMUX includes positive and negative items. UMUX-Lite retains two positive items, so its scoring does not use the same reversal pattern. Later variants, including UX-Lite, make further changes and should be named accurately rather than treated as interchangeable labels.

MeasuringU’s account of the evolution from UMUX-Lite to UX-Lite explains why version and scoring choices matter. A team comparing an existing series should first identify the wording, scale and transformation used in earlier waves.

Choose according to the construct, participant burden and comparison needs. The System Usability Scale has ten items and an extensive body of reference data, but its additional items do not automatically diagnose the particular interface problem. Either instrument benefits from observation and follow-up questions.

Keep the calculation attached to its instrument

For the original four-item UMUX with responses from 1 to 7, positive items contribute the response minus 1 and negative items contribute 7 minus the response. The total is divided by 24 and multiplied by 100, producing a score from 0 to 100.

For the two positive UMUX-Lite items on the same seven-point scale, subtract 2 from their sum, divide by 12 and multiply by 100 for a linearly rescaled score. As a scoring check, two middle responses of 4 give 50. The score is not a percentage of users who found the product usable.

A published regression adjustment has also been used to estimate SUS from UMUX-Lite. MeasuringU’s earlier explanation gives that approach, while later research discusses alternatives and limitations. A transformed estimate is not an observed SUS score. Name the transformation and avoid silently mixing adjusted and unadjusted results.

Interpret change with the study conditions

Administer the instrument after relevant experience, using the documented wording and response scale for the chosen version. Specify missing-data handling and keep recruitment, timing and task context comparable where the purpose is tracking.

Report the sample size and an appropriate confidence interval. A small difference between two scores may reflect sampling variation, changed respondents or a changed administration process rather than a product improvement.

Use the result to identify questions for further investigation. A low capability judgement may concern a missing function, an undiscovered function or a mismatch with the participant’s needs. Usability testing and interviews can help separate those possibilities, giving the compact score a more useful interpretation than a grade alone.