t Test for Independent and Dependent Groups
Matched t test
A dependent t test design where the participants are paired or matched on some critical variables that are known to be related to the dependent variable in the experiment.
Repeated measures t-test
A dependent t-test design where the participants are the same in both groups and usually are measured pretreatment and posttreatment.
Cohen's d
A measure of effect size that estimates the magnitude of the differences between the means in terms of a standard deviation.
Dependent t test
A test designed to determine the statistical difference between two means where the participants in each group are either the same or matched pairs.
Independent t test
A test to determine whether there are significant differences between two independent groups' means on the same dependent variable.
Heterogeneity of variance
A violation of the assumption that two groups have equal or approximately equal variances in the independent groups t test; in this case, the variances are not equal.
Homogeneity of variance
An assumption in the independent groups t test that the two groups have equal or approximately equal variances.
Effect size
How strongly the independent variable affects the dependent variable. Effect size ranges from small to large.
Controlled experiment
The most powerful experimental design because it allows for the inference of causation. The participants are randomly chosen from a population and randomly assigned to one of two (or more) groups, usually an experimental group that receives some treatment and a control group that receives a placebo.
Order of presentation
The order in which levels of the independent variable are presented to the participants. If the conditions are not counterbalanced, the participants may become fatigued or bored and do more poorly in the later conditions.
Power
the probability that the null hypothesis will be rejected correctly. In significance testing, the power of a test is estimated by 1 - Beta, where Beta is the probability of committing a Type II error (retaining the null hypothesis when false). The standard for minimum power in psychological research is .80.
