Chapter 10: Comparing Two Means

Question 1

Independent Samples t-test

Answer 1

two different group means being compared

Question 2

Paired Samples t-test

Answer 2

same people in groups receive all conditions (they’re their own control), compare two means from the same people from one time to another

Question 3

what is the rationale for the t test?

Answer 3

as two sample means get further and further apart, it is less likely that they could be coming from a sample with the same mean (come from different populations)

Question 4

independent t test equation

Answer 4

t = X1-X2 / estimate of the standard error
t will be small if the difference between means is small, and large if the difference is large
bigger t = smaller p-value

Question 5

t test assumptions

Answer 5

1) normality: sampling distribution of the differences between scores, not the scores themselves
2) homogeneity: all comparison groups have the same variance
3) independence: errors from each individual case are unrelated to eachother

Question 6

independent samples t test alternatives

Answer 6

1) Mann whitney
2) robust tests of two independent means
3) bayesian test of two means

Question 7

difference between between groups (independent) and repeated measures (paired) t tests?

Answer 7

repeated measures has more power, more sensitive (likely to find an effect if one exists, lessens individual variation
however, watch out for order effects (i.e., practice/fatigue effect)
- counterbalance conditions to remove confound effect, but can change effect size
-

Question 8

standard error of differences

Answer 8

SD of the sampling distribution

Question 9

variance sum law

Answer 9

the variance of a difference between two independent variables is equal to the sum of their variances (variance of the sampling distributions between two samples means will be equal to the sim of the variances of the two populations from which the samples were taken)

Question 10

grand mean

Answer 10

the mean of an entire set of observations