Ordinal tests for two or more samples

  • George Argyrous


The previous two chapters discussed situations where two or more samples were compared in terms of a variable measured at the interval/ratio level. However, in social science we often do not work with interval/ratio data but ordinal-level data instead. Sometimes this ordinal-level data looks like interval/ratio, especially with attitude scales that have a large number of points. For example, we could construct an ‘index of satisfaction’, whereby we ask individuals to rate themselves on a scale of I-to-lO , with 1 indicating ‘Not at all satisfied’ and 10 indicating ‘Extremely satisfied’:


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© George Argyrous 1997

Authors and Affiliations

  • George Argyrous
    • 1
  1. 1.School of Social Science and PolicyThe University of New South WalesAustralia

Personalised recommendations