Pass or fail? Binomial-related hypothesis testing and confidence intervals using paired samples
Abstract
In this chapter we introduce what to do when we are looking at “pass/fail” data and have “paired data,” or a “within-subject design.” “Paired data” or “paired samples” are terms that are often used when there is a within-subject comparison with only two entities. Put more simply, within-subjects design means that we have the same people trying to complete the same multiple tasks, or choose the best design, or a myriad of other scenarios in the life of a UX researcher. Of course, as we noted in Chapter 2 where we introduced the t-test for comparing means with independent samples, it is not always feasible to have paired data, due to ...
Get Improving the User Experience through Practical Data Analytics now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
