one-tailed test

Terms from Statistics for HCI: Making Sense of Quantitative Data

A test where data is only compared in one direction. For example, when comparing two systems you may only care about seeing if the new system is better. If they are the same or if the new system is worse, these are equally failures of the new design. So, you care that the new system has a faster response time than the old one (one direction), or if it has a lower error rate (one direction). Contrast with two-tailed test.

Defined on page 47

Context: Crucially, do you care about direction? Imagine you are about to introduce a new system that has some additional functionality. However, you are worried that its additional complexity will make it more error prone. Before you deploy it you want to check this. Here your question is: “Is the error rate higher?” If it is actually lower that is nice, but you don’t really care so long as it hasn’t made things worse. This is a one-tailed test; you only care about one direction.

Used on pages 47, 90