Skip to main content

Non-parametric tests for dependent samples

  • Chapter
Statistics for Social Research
  • 319 Accesses

Abstract

The choice of tests for dependent samples, as with tests for independent samples, is affected by the level at which the test variable is measured. For interval/ratio data, in Chapter 17, we used a t-test for the mean difference. This chapter will consider non-parametric tests that can be applied to dependent samples compared in terms of nominal or ordinal data:

  • The McNemar test/Sign test These are used when the test variable is arranged into a binomial distribution (often after collapsing values).

  • The Wilcoxon signed-ranks test This is used when the test variable is measured at least at the ordinal level.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Institutional subscriptions

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Author information

Authors and Affiliations

Authors

Copyright information

© 1997 George Argyrous

About this chapter

Cite this chapter

Argyrous, G. (1997). Non-parametric tests for dependent samples. In: Statistics for Social Research. Palgrave, London. https://doi.org/10.1007/978-1-349-14777-9_18

Download citation

Publish with us

Policies and ethics