 |
Chapter 6: Analysing the Data |
 |
Chi-SquareUp to this point, all of our discussion of statistical procedures has been focused on the analysis of "quantitative" data. But some of the variables in behavioural research are not quantitative. Instead, they are "qualitative" in nature. A qualitative variable is one where the possible "measurements" are not measurements as such but quantities or frequencies of things that occur in categories. Religious affiliation, gender, or which brand of dishwashing detergent you prefer are examples. A very useful test for this type of data is the Chi-square test. It uses counts or frequencies as data rather than means and standard deviations. So we can compare how many people or responses or things fall into one category instead of another compared to some hypothesised or expected number. Unfortunately SPSS does not do one sample chi-square test. However, because it is such a useful and simple test it is included here. The tests involving the use of |
|
© Copyright 2000 University of New England, Armidale, NSW, 2351. All rights reserved
|
2 are usually considered as part of the branch of nonparametric statistics since we examine the statistics of category membership (nominal or ordinal measurement) rather than the statistics of means and standard deviations (interval or ratio measurement). The assumption of normality is not required in 