Lecture 4 Flashcards Preview

Metho 3 > Lecture 4 > Flashcards

Flashcards in Lecture 4 Deck (13)
Loading flashcards...
1
Q

why not use multiple unpaired t tests when a factor has 3 (or more) levels?

A

> testing all differences pairwise is cumbersome

> chance capitalization for repeated tests on the same data

2
Q

ANOVA: whats in numerator? whats in denominator?

A

ANOVA:

numerator: variance between groups
denominator: variance within groups

3
Q

how to calculate the SStotal?

A

SStotal

> sum of deviation of all observations from the overall mean

4
Q

how to calculate SSeffect?

A

SS effect

> sum of the deviations of the group means from the overall mean

5
Q

how to calculate SSerror?

A

SS error

> sum of the deviations within a group from the group mean

6
Q

3 types degrees of freedom in one way ANOVA

(between subjects design)

> how to calculate

A

total: df = N-1
effect: df = number of levels - 1
error: df = (n -1) x number of levels

total = effect + error

7
Q

ANOVA:

> what F value to expect when there is null effect

> why?

A

ANOVA:

> typical null effect has an F value around 1

> if F = 0, the conditions are identical

> but MSeffect also contains error, so F = 0 is unlikely

8
Q

how does F test relate to t test?

A

F- test is a generalization of the t test

> F with df = (1,x) equals t² with df = x

9
Q

4 types degrees of freedom in ANOVA

(within subject design)

> how to calculate

A
  1. total = ( a x n ) -1
  2. effect = a - 1
  3. subject = n-1
  4. error = (n-1)(a-1)
10
Q

why does a within anova (assuming the same data) have more power than between?

A

withing subjects anova partitions the error variance into within subject variance and pure error variance

> MS effect = SS effect / SS error

> SSeffect is the same, but SS error is smaller

11
Q

what are post hoc tests?

why are they important?

A

post hoc: anova tells you that there is a difference between two or more of the groups you tested

> does not tell you which groups, and if more than one

post hoc tests:

> determine which groups differ, and take chance capitalization into account

12
Q

what are 2 common post hoc tests?

A

post hoc tests

  1. bonferroni

> divide alpha by number of tests (conservative!)

  1. tuckey HSD

> less conservative

13
Q

what are two non parametric alternatives for one way ANOVA?

> when use which?

A

distribution free alternatives

  1. kruskal wallis test

> use for between subject design

  1. friedman test

> use for within subject design

>>> both provide chi square values