
Small Sample Just Won't Cut It
Both small and large sample research methods were used to evaluate the design. B|P conducted 20 usability interviews to gain insight into users' behavioral patterns and interactions with the new design. Automated sessions ran simultaneously, where 200 participants posted comments to the prototype on virtual sticky notes. These participants were also asked to find information while their click patterns were recorded and tabulated.
Not only did the unmoderated study support the story that the moderated testing was starting to tell, but it expanded the range of usability issues that we uncovered. With the help of the automated data, we got a greater level of precision into our insights.
For example, one usability finding from the moderated study told us that users had difficulty finding local news. In the automated study, we learned just how often users were unsuccessful in completing this task. We also learned exactly where they most often looked for local news by mistake.
With the help of large sample data, we were able to judge the severity of the usability issue, as well as find out more about what was contributing to the problem.
|