two-tailed test

Terms from Statistics for HCI: Making Sense of Quantitative Data

A test where data is compared in two directions. For example, assume you are choosing which of two systems, A and B, to purchase for you company and want to check if either is measurably better than the other. The results matter in both directions: whether A is better than B or whether B is better than A. If comparing response times your test has to look both at when System A is significantly faster and at when it is significantly slower. Contrast with one-tailed test.

Defined on page 47

Used on pages 47, 90