ANOVA (between-participants) Flashcards
What are between-participant designs
The participants each contribute 1 data point only to each dependent variable
What stats tests could be conducted with below example:
Q - what is the effect of group status on perceptions of intergroup conflict
H - group members less likely to perceive conflict with another group when their ingroup has high status
IV - ingroup status (2 conditions: high & low)
DV - perceptions of intergroup conflict (measured by questionnaire)
Independent-samples t-test
OR
One-way ANOVA
What stats tests could be conducted (and any jamovi notes) with below example:
Q - what is the effect of group status on perceptions of intergroup conflict
H - group members less likely to perceive conflict with another group when their ingroup has high or equal status (relative to an outgroup)
IV - ingroup status (3 conditions: high vs equal vs low)
DV - perceptions of intergroup conflict (measured by questionnaire)
There’s 3 conditions in the IV so there is only one analysis option: One-way ANOVA
- because it allows an overall test of the effect of the independent variable
- and it tests the specific comparisons between individual conditions (eg high vs low, equal vs low)
When setting this up in jamovi, you’ll want to change ‘simple’ in the dropdown box to ‘helmert’ which will ask for contrasts of low vs equal and high combined, and then equal vs high
What are the independent variable(s) in the below example:
Q - what is the effect of group status on perceptions of intergroup conflict
H - group members less likely to perceive conflict with another group when their ingroup has high status, BUT only when addressing an outgroup audience rather than ingroup
And what is an interaction effect in relation to this example?
IV1 - ingroup status (high vs low)
IV2 - audience (ingroup vs outgroup)
if the effect of IV1 depends on IV2 then there is an interaction effect
What stats tests could be conducted (and any jamovi notes) with below example:
Q - what is the effect of group status on perceptions of intergroup conflict
H - group members less likely to perceive conflict with another group when their ingroup has high status, BUT only when addressing an outgroup audience rather than ingroup
IV - ingroup status (2 conditions: high vs low), and audience (ingroup vs outgroup)
DV - perceptions of intergroup conflict (measured by questionnaire)
This is a more complex design with 2+ IVs -
Two-way ANOVA
- tests the main effects of the two IVs (effect of status collapsed across audience, and vice versa)
- tests the interaction between the two IVs (extent to which effect of status varies depending on audience)
This output is the result of a two-way ANOVA - what are the take-aways from this?
Main effects of ingroupstatus and audience are both non-significant
But the interaction between the two effects is significant
F(1, 67) = 5.44
p = .023
hp2 = .08 (partial eta squared, what this should look like is attached)
We cannot tell from this analysis if ingroup status would only affect perceived conflict in the outgroup audience condition
What analysis do we need to conduct to understand if high ingroup status would only affect perceived conflict in the outgroup audience condition
We need the analysis of the effect of status of each level of the audience variable
Simple main effects analysis
- tests the main effect of IV1 at each different level at IV2
This is the analysis of a simple main effects analysis - what does it tell us?
The simple main effect (perceptions of intergroup conflict according to ingroup status?) aren’t significant in ingroup audience condition
But they are significant in the outgroup audience condition
This is a simple main effects analysis - pairwise comparisons. What does it tell us?
High and Equal conditions significantly differ from the Low condition - but only in the outgroup audience condition
What are the independent variable(s) in the below example:
Q - what is the effect of group status on perceptions of intergroup conflict
H - group members less likely to perceive conflict with another group when their ingroup has high status, BUT only when addressing an outgroup audience rather than ingroup, AND when perceived threat is low rather than high
And what is an interaction effect in relation to this example?
IV1 - ingroup status (high vs low)
IV2 - audience (ingroup vs outgroup)
IV3 - threat (high vs low)
The effect of IV1 will depend on IV2, but only at specific levels of IV3 - this is a 3-way interaction effect
What stats tests could be conducted (and any jamovi notes) with below example:
Q - what is the effect of group status on perceptions of intergroup conflict
H - group members less likely to perceive conflict with another group when their ingroup has high status, BUT only when addressing an outgroup audience rather than ingroup, AND when perceived threat is low rather than high
IV - ingroup status (2 conditions: high vs low), audience (ingroup vs outgroup), and threat (high vs low)
DV - perceptions of intergroup conflict (measured by questionnaire)
Three-way ANOVA
- tests the main effects of the 3 IVs
- tests the 3 two-way interactions between each pair of IVs (status x audience, threat x audience, status x threat)
- tests the three-way interaction between all 3 IVs
