Quiz 5
The Friedman's ANOVA by rank test is
The Friedman's ANOVA by rank test is a nonparametric test that is similar to repeated-measures ANOVA and can be used when the data do not meet the assumptions. A post hoc test is used to verify results. There are no ad hoc tests.
When using the Kruskal-Wallis H-statistic test, one of two tables can be used to determine whether the two groups are significantly different. If each group contains at least five subjects, then the ______ table is used.\
The chi-square table is used if each group contains at least five subjects. If each group contains fewer than five subjects, then the Kruskal-Wallis H-table is used to obtain an exact p-value. The ANOVA table and t test table both show the final computations of the data.
The independent variable(s) in the repeated-measures design is/are
The independent variable of the repeated-measures design is implicit, (eg, not actually measured as a number that is used in computation). The dependent variable is explicitly measured. It is also defined by one of four general situations—the same measurement is taken on the same subject at different points in time; the same measurement is taken under different experimental conditions; the independent variable is "type of medicine" and has three levels; and a study has a case and multiple matched control groups in which each group is a level of the independent variable. The dependent variable is explicitly measured.
If the correlation between two variables is r = 0.50, the variance of the residual is
Variance of the residuals is the amount of variance in the dependent variable not explained by the dependent variable and covariates included in the analysis. It is calculated as variance of the residual = 1 - r2 = 1 - 0.52 = 0.75 or 75%.
It is fairly common for a health research question to involve a comparison of several groups on a particular measure. To examine the differences among the groups through an analysis that considers the variation across all groups at once, we can use the
When we have more than two groups and are interested in the differences among these groups, we are dealing with different combinations of means or distributions; thus, the one-way ANOVA would be the appropriate test. If using the t test, we would need to do a number of independent sample t tests. The Mann-Whitney U-test and Kruskal-Wallis H-test both are nonparametric analogs.
The Summary of ANOVA table displays all the elements necessary to compute the f-ratio between the within-group and between-group variance. Typical columns include all the following, except degrees of freedom. t test. sums of squares. mean Square.
t test.
After a significant f-test has been obtained for repeated measures ANOVA, the null hypothesis is ________.
After a significant f-test has been obtained for repeated-measures ANOVA, the null hypothesis is rejected, and it can be said that at least one of the group means is different from the others. Not rejecting the null hypothesis could result in a type I error. There is no need to update or reevaluate the null hypothesis.
