Ch. 10: The t Test for Two Independent Samples

Independent-measures research design

A research design that uses a separate group of participants for each treatment condition (or for each population) is called an independent-measures research design or a between-subjects design

Between-subjects design

A research designs that uses a separate group of participants for each treatment condition (or for each population) is called an independent-measures research deisgn or a between-subjects design

Repeated-measures research design

A research method where the same participants are tested in all conditions of the experiment. For example, you might measure people’s memory before and after caffeine. This designs controls for individual differences because each person serves as their own control

Within-subject design

Basically the same idea as repeated-measures: each particpant experiences every condition in the study. The focus is on how each person’s performance changes across conditions rather than comparing different groups

Independent-measures t statistics

A statistical teset used when comparing the means of two separate groups (different participants in each group). It tells you if the difference between group means is big enough to be unlikely due to chance. Example: comparing test scores of a group that studied vs. a group that didn’t

Estimated standard error of M1-M2

This measures how much difference you’d expect between two sample means just by chance. It’s like a “margin of error” for the difference between groups—the smaller it is, the more confidently you can say your groups are truly different

Pooled variance

When comparing two groups, this is the average of their variances, adjusted for sample size. It’s used when you assume both groups have similara variability. It helps get a more stable estimate of how spread out the scores are overall

Homogeneity of variance

An assumption that both groups you’re comparing have about the same amount of variability (their scores spread out roughly equally) if one group’s scores vary a lot more than the other’s, this assumption is violated, which can affect the accuracy of your t-test