Week 3: ANOVAs Flashcards

1
Q

what type of numbers do ANOVAs use?

A

ANOVAs use the F statistic

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

if variance between samples is small, F will be ______

A

small

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

if variance within samples is small, F will be ______

A

large

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

ANOVAs will compare

A

3 or more groups (= levels of 1 IV)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

One-way ANOVA definition

A

one IV with 3+ levels: dry needling vs massage vs sham dry needling; separate groups for each intervention

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

One-way repeated measures ANOVA

A

one IV with 3+ levels: dry needling vs massage vs sham dry needling; everyone gets all interventions

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Two-way ANOVA

A

two IVs: Dry needling vs sham dry needling AND stretching vs no stretching; separate groups for each intervention

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Two-way repeated measures ANOVA

A

two IVs: Dry needling vs sham dry needling AND stretching vs no stretching; everyone gets ALL combination of IVs

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Mixed Model ANOVA

A

Two IVs: dry needling vs sham dry needling AND time (pretest, 4-weeks post, 8-weeks post); two intervention groups, but all participants are measured at all points in time

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Total Variance in a One-Way ANOVA is what two groups

A

between groups and within groups

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

between groups variance is

A

differences between means

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

within groups variance is

A

between subjects error variance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

comparison of group means

A

ANOVA looks at distance of each group mean from the grand mean (total group)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Interpreting F statistic

A

“omnibus test”; overall; nonspecific; tell you a difference exists, but will not specify WHERE

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Multiple comparison tests will tell you ________ the difference exists

A

WHERE the difference exists; if null is not rejected, no multiple comparison tests are needed

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

effect size =

A

how much the IV affected the DV

17
Q

effect size indices for the ANOVA

A

eta squared (n^2) and Cohen’s f

18
Q

small effect size: n^2

19
Q

medium effect size: n^2

20
Q

large effect size: n^2

21
Q

small effect size: f

22
Q

medium effect size: f

23
Q

large effect size: f

24
Q

designs for repeated measures has what characteristic

A

the same people in each level of the IV

25
simplest example of a repeated measures design is
one-way repeated measures design
26
true/false: subjects act as their own controls in designs for repeated measures
true
27
multiple comparison tests are used to
determine "where" difference is after using ANOVA; also called "pairwise comparisons"
28
what are the 2 different strategies for multiple comparison tests
post-hoc and planned comparisons
29
post-hoc info
- performed after ANOVA (only if significant) - most common - test every difference, therefore are "exploratory"
30
planned comparisons info
- performed instead of ANOVA (a priori) - focused only on specific comparisons - you won't see this used very often
31
types of multiple comparison tests for independent groups
- Fisher's least significant difference - Duncan multiple range test - Newman-Keuls Method - Tukey's honestly significant difference - Bonferroni t-test - Scheffé's comparison
32
Fisher's Least Significant Difference information
essentially unadjusted t-tests (LSD)
33
Tukey's Honestly Significant Difference info
"middle of the road" in terms of risk and most commonly used
34
Bonferroni t-test info
simply divides alpha by # of comparisons (also called Bonferroni adjustment or correction
35
multiple comparison test for Repeated Measures
LSD, Sidak, Bonferroni correction
36
LSD definition
unadjusted paired t-tests
37
Sidak definition
adjusted, but good balance of type 1 and type 2 error protection; MOST COMMON
38
Bonferroni correction definition
divides alpha by # of comparison
39
We only need Multiple Comparisons IF there is a _______________ in the omnibus test
significant difference