Critical incidents and critical threads in empirical usability evaluation

Empirical usability evaluations (particularly “formative” evaluations, Scriven, 1967) hinge on observing and interpreting critical incidents (Flanagan, 1954) of use: the causes of such critical incidents can often be found in the immediate contexts of their occurrence and can guide specific design changes. However, it can also happen that the causes of a critical incident are temporally remote from its context of occurrence or distributed throughout the user’s prior experiences. We propose augmenting critical incident methods by analysis of what we call “critical threads”: sets of causally related user episodes that, taken together, define major usability themes.