Module 7

Question 1

What does the intercept represent?

Answer 1

The mean of the reference category

Question 2

if a researcher waited until the data were collected and then looked at the sample means of the five groups then decided which means to test for significant - why wouldn’t the p-value be valid?

Answer 2

The p-value for the corresponding t-test of a dummy variable coefficients would not be valid because the comparisons were not planned a priori

Question 3

Type 1 error and multiple significance tests

Answer 3

the probability of a type 1 error for a single significance test is set as 0.05.

if one performs multiple significance tests on the same data without proper adjustments then the probability that at least one of the tests produces a type 1 error is greater than .05

Question 4

Formula for type 1 error accumulation

Answer 4

Prob of at least type 1 error = 1-Probability of no type 1 errors = 1-(1-a)^c

where c is the number of significance tests to be performed

Question 5

experiment wise/family wise type 1 error rate

Answer 5

the probability of obtaining at least one Type 1 error in a set of tests run on the same data set is known as this

• probability given is not quite as high as it shows because they are not independent rhyming vs counting and rhyming vs imagery both include rhyming

Question 6

Tukeys HSD

Answer 6

post-hoc test for pairwise comparisons

experiment -wise type 1 error rate is maintained at the alpha level used to test the omnibus null hypothesis regardless of whether the pairwise comparisons were planned a priori

p value is adjusted for the fact we are doing multiple comparisons

Question 7

Bonferroni adjustment

Answer 7

least powerful

experiment-wise alpha level is simply divided by the number of specific hypothesis tests to be performed 0.05/10 = 0.005