We found that neither scale format (numeric vs. slider) nor device (desktop vs. mobile) strongly affected Likelihood-to-Recommend rating behaviors. measuringu.com/ltr-numeric-s…
We applied three estimation methods to two sets of data to see which of them best matched concurrently collected UX-Lite™ and SUS scores. measuringu.com/accuracy-of-s…
The inverse square relationship for the drop in intensity as you move away from an energy source is analogous to the drop in precision that occurs when sample sizes are smaller. measuringu.com/inverse-squar…
Probably most of us who have taken an introductory statistics class have heard the rule of thumb that your sample size should be at least 30. But is that right? measuringu.com/do-statistics…
The UX-Lite is a two-item standardized questionnaire adapted from the UMUX-Lite. In this article, we trace the evolution of the UX-Lite starting from two standardized questionnaires created in the mid-1980s. measuringu.com/evolution-of-…
Is it realistic to expect consistency in #usability problem discovery? Is consistency in problem discovery necessary for usability testing to be effective? measuringu.com/testing-effec…
Short words and sentences are generally easier to understand than long ones, so we checked if changing the perceived usefulness item of the UMUX-Lite from 13 to 8 syllables affected its measurement properties. The simple version worked just as well. measuringu.com/simpler-umux-…
It is important for companies to track negative word of mouth, so we reviewed data in the relevant literature to understand whether the Net Promoter Score adequately measures recommending against a company. measuringu.com/does-the-nps-…
Does adding a third dimension to bar charts affect how quickly people can compare bar heights? It does a little, but it has a powerful effect on preference—and maybe not in the way you might expect. measuringu.com/are-3d-bar-gr…
Using “will recommend” instead of “would recommend” resulted in slightly higher scores, but the result was statistically significant for only one of four brands in our study. measuringu.com/will-would