6.05 Several samples - Kruskal-Wallis test

Course video 44 of 46

In this module we'll discuss the last topic of this course: Non-parametric tests. Until now we've mostly considered tests that require assumptions about the shape of the distribution (z-tests, t-tests and F-tests). Sometimes those assumptions don't hold. Non-parametric tests require fewer of those assumptions. There are several non-parametric tests that correspond to the parametric z-, t- and F-tests. These tests also come in handy when the response variable is an ordered categorical variable as opposed to a quantitative variable. There are also non-parametric equivalents to the correlation coefficient and some tests that have no parametric-counterparts.

4.4(293 个评分) | 31K 名学生已注册

关于 Coursera


Join a community of 40 million learners from around the world
Earn a skill-based course certificate to apply your knowledge
Gain confidence in your skills and further your career