Chapter 14 Chi-Square Tests
Hypothesis testing procedures for nominal variables (whose values are categories) Focus on the number of people in different categories
Chi-Square Statistic Observed frequency distribution Expected frequency distribution Chi-square statistic (χ 2 )
Chi-Square Statistic Chi-square distribution
Chi-Square Statistic Chi-square table
The Chi-Square Test for Goodness of Fit Levels of a single nominal variable
The Chi-Square Test for Independence Two nominal variables, each with several categories Contingency table
The Chi-Square Test for Independence Independence –No relation between the variables in a contingency table Sample and population
The Chi-Square Test for Independence Determining expected frequencies
The Chi-Square Test for Independence Figuring chi-square Degrees of freedom
Assumptions for Chi-Square Tests No individual can be counted in more than one category or cell
Effect Size for Chi-Square Test for Independence 2 X 2 contingency table –Phi coefficient (φ) –small φ =.10 –medium φ =.30 –large φ =.50
Effect Size for Chi-Square Test for Independence Contingency tables larger than 2 x 2 –Cramer’s phi –Effect size for Cramer’s phi
Power for Chi-Square Test for Independence (.05 significance level)
Approximate Sample Size Needed for 80% Power (.05 significance level
Controversies and Limitations Minimum acceptable frequency for a category or cell Small expected frequencies –At least 5 times as many individuals as categories (or cells) –Reduce power
Chi-Square Tests in Research Articles χ 2 (2, n = 101) = 11.89, p <.005