Abstract
Usability studies are important for developing usable, enjoyable products, identifying design flaws (usability problems) likely to compromise the user experience. Usability testing is recommended for improving interactive design, but discovery of usability problems depends on the number of users tested. When usability is critical, an extended statistical model would help estimate the number of undiscovered problems; researchers incrementally add participants to the study until it discovers all problems. For estimating required sample size, usability researchers often resort to either magic numbers or the geometric series formula; inaccurate for making predictions, both underestimate required sample size. Most usability practitioners will likely continue to use strategies of iterative low-budget evaluation where quantitative statements are unreliable but also unnecessary.
Original language | English |
---|---|
Pages (from-to) | 64-70 |
Number of pages | 7 |
Journal | Communications of the ACM |
Volume | 55 |
Issue number | 4 |
DOIs | |
Publication status | Published - 2012 |
Keywords
- METIS-288872
- IR-83286