Analysis of Categorical Data

An
ANOVA
is used to determine if 2 or more groups are
different. With a dependent variable measured on an interval or ration scale. Know the different
scale used for measurement for the test. A
ttest
is used to determine if 2 groups are different.
With a dependent variable measured on an interval or ratio scale. A
chisquare test
is used to
determine if 2 or more groups are different. With a dependent variable measured on a nominal
scale. A
MannWhitney U Test
is used to determine if 2 groups independent of each other are
different. With a dependent variable measured on an ordinal scale, the groups must be
independent of each other. Cannot perform a MannWhitney U Test in Excel, must use SPSS, Go
to: Analyze, Nonparametric Tests, 2 Independent Samples, Look at the [Asymp. Sig (2tailed)]
result and treat it as a pvalue if it is less than .05 you reject the null hypothesis or you do not
reject if larger.
Wilcoxon Matched Pairs Signed Rank Test
Use if the data are related. Good
for beforeandafter studies. Cannot perform a Wilcoxon test in Excel, must use SPSS, Go to:
Analyze, Nonparametric Tests, 2 Related Samples, Look at the [Asymp. Sig (2tailed)] result and
treat it as a pvalue if it is less than .05 you reject the null hypothesis or you do not reject if
larger.
KruskalWallis Test
Use like an ANOVA when you want to determine if three are
differences among> 2 groups, but data measured on an ordinal scale. Groups should not be
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
This is the end of the preview.
Sign up
to
access the rest of the document.
 Spring '08
 Mendez
 Statistics, Null hypothesis, Statistical tests, Nonparametric statistics

Click to edit the document details