the number of groups involved in the analysis
 the number of pairs of groups involved in the analysis
 the number of dependent variables in the ANOVA
 the number of independent variables in the ANOVA

 True
 False

 With several measures matching is usually unnecessary.
 Multiple before/after t-tests will provide the same data.
 Matching won’t control the error variance in more than two subjects.
 It’s difficult to find enough subjects that have the same characteristics.

 In any ancillary analysis that may be conducted
 Error variance
 In the differences between groups
 As a portion of the treatment effect

 Differences in the level of the treatment applied
 Differences within the groups involved
 The independence of the groups involved in the analysis
 Differences in sample size

 A variable that causes variance but is uncontrolled.
 A variable that is mathematically controlled in an analysis.
 A variable that is controlled but not measured.
 A variable that correlates with other analyses.

 To determine statistical significance
 To determine effect size
 To determine error variance
 To determine which pairs are significantly different

 All variance minus the variance related to measure-to-measure differences
 The measure-to-measure differences minus the residual variability
 The residual variability minus the treatment effect
 All variance minus the treatment effect minus the subject-to-subject differences

 All groups have similar amounts of data variability.
 The same mean value occurs in each group.
 All groups are drawn from the same population.
 All groups have the same sample size.

 The measure-to-measure differences
 The subject-to-subject differences
 The measure times subjects differences
 The residual differences