BST 322 WEEK 3 VERIFIED STUDY GUIDE
chi-square test of independence - Answers - - basic non-parametric test
- determine if two categorical variables are independent; basically if they have a
relationship
- compares the observed and expected cell frequencies of a contingency table
expected frequencies - Answers - - we would expect to see if the subjects were
distributed through the table randomly
observed frequencies - Answers - - those counts we literally got in the course of some
research
Yate's correction - Answers - - a variation on the chi-square test when the expected cell
frequency in any cell in a 2x2 table
Fisher's exact test - Answers - - a variation on the chi-square test
- only used on a 2x2 table
Phi coefficient - Answers - the correlation coefficient used when both measured
variables are dichotomous and nominal
Cramer's V - Answers - - way of calculating correlation in tables which have more than
2x2 rows and columns
- post-test to determine strengths of association after chi-square has determined
significance
chi-square goodness of fit test - Answers - - analogous to the one-sample t-test and is
performed when you wish to see if your distribution of categorical variables matches
that of a known population
Mann-Whitney U-test - Answers - - used when the dependent variable is ordinal and is
the analog of the independent groups t-test
Kruskal-Wallis test - Answers - - the non-parametric counterpart to the simple ANOVA,
involves ranking, and should be used only when there are at least five cases per group
McNemar test - Answers - - the chi-square analog to the paired t-test
- subjects are related in some way
Cochran's Q test - Answers - - the non-parametric analog of RANOVA
- used to test for differences in proportion when the dependent variables is dichotomous
and there are three or more repeated observations or correlated groups
chi-square test of independence - Answers - - basic non-parametric test
- determine if two categorical variables are independent; basically if they have a
relationship
- compares the observed and expected cell frequencies of a contingency table
expected frequencies - Answers - - we would expect to see if the subjects were
distributed through the table randomly
observed frequencies - Answers - - those counts we literally got in the course of some
research
Yate's correction - Answers - - a variation on the chi-square test when the expected cell
frequency in any cell in a 2x2 table
Fisher's exact test - Answers - - a variation on the chi-square test
- only used on a 2x2 table
Phi coefficient - Answers - the correlation coefficient used when both measured
variables are dichotomous and nominal
Cramer's V - Answers - - way of calculating correlation in tables which have more than
2x2 rows and columns
- post-test to determine strengths of association after chi-square has determined
significance
chi-square goodness of fit test - Answers - - analogous to the one-sample t-test and is
performed when you wish to see if your distribution of categorical variables matches
that of a known population
Mann-Whitney U-test - Answers - - used when the dependent variable is ordinal and is
the analog of the independent groups t-test
Kruskal-Wallis test - Answers - - the non-parametric counterpart to the simple ANOVA,
involves ranking, and should be used only when there are at least five cases per group
McNemar test - Answers - - the chi-square analog to the paired t-test
- subjects are related in some way
Cochran's Q test - Answers - - the non-parametric analog of RANOVA
- used to test for differences in proportion when the dependent variables is dichotomous
and there are three or more repeated observations or correlated groups
