General Concepts and Planned Comparisions and Post Hoc Tests

Question 1

What is Post Hoc?

Answer 1

no specific hypothesis about where the differences might lie so we look at the comparisons between possible means

Question 2

What are planned comparisons?

Answer 2

specific contrasts of interests decided on A PRIORI to test specific predictions

Question 3

How do multiple comparisions effect the Type 1 Error Rate?

Answer 3

increase probability of making error

Question 4

What is the trade-off between Type 1 and Type 2 errors?

Answer 4

when we decrease the likelihood of Type I errors, we decrease power, and increase the likelihood of Type II errors

Question 5

What is the most common technique for controlling for Type 1 errors in Planned Comparisions?

Answer 5

Bonferroni

Question 6

What is the most common technique for controlling for Type 1 errors in Post Hoc?

Answer 6

Tukey

Question 7

What is the rule for whether or not a Bonferroni adjustment is needed for Planned Comparison?

Answer 7

If the number of planned comparisions doesn’t exceed df-effect then there is no need to adjust the alpha level

Question 8

What bonferroni adjustment do you use for planned comparisons?

Answer 8

.05 / number of comparisons e.g. if there are 4 comparisons = .05 / 4 = .0125

Question 9

What is the interaction effect?

Answer 9

it is the extent to which the effect of one factor differs across the levels of the other factor

Question 10

What does it mean if the lines on a graph are parallel?

Answer 10

no significant main effect or interaction effect

Question 11

What does it mean if the lines on a graph are not parallel?

Answer 11

Significant interaction
- needs following up