Wk4: Interactions Flashcards

(11 cards)

1
Q

Main effects have a _________ effect on the DV

A

Additive

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

Define order of interaction

A

Number of IVs (not levels)
e.g. IV1 = 4 levels, IV2 = 5 levels, IV3 = 2 levels;
3 orders total

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

Combination formula

A

nCr = n!/ (r! (n-r)!)

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

Name the interactions of a 2x2x2 design

A
  • 3 main effects (A, B, C)
  • 3 two-way interactions (AB, BC, AC)
  • 1 three-way interaction (ABC)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Name the interactions of a 4-way design

A
  • 4 main effects (A, B, C, D)
  • 6 two-way interactions (AB, AC, AD, BC, BD, CD)
  • 4 three-way interactions (ABC, ABD, ACD, BCD)
  • 1 four-way interaction (ABCD)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Disordinal interactions must have:

A
  1. Slopes have different direction
  2. Intersection of lines
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What’s stronger: ordinal or disordinal?

A

Disordinal interactions are stronger.

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

Problems with ordinal interactions

A
  • Scaling artifacts
  • MSMT artifacts: floor ceiling effects
  • MSMT artifacts: regression to the mean
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What are scaling artifacts?

A
  • The scale used may not accurately reflect real-world differences.
  • Present in ordinal-scales and below (interval and ratio do not have scaling artifacts)
  • E.g. in a 0-10 attractiveness scale, the difference between 4-5 is less than 8-9.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What are floor/ ceiling effects?

A
  • The measurement used may have results that “pool” near the top/ bottom of the scale, meaning results cannot increase/ decrease a significant amount.
  • E.g. In a reading test, comparing 1st and 4th quartile students after a program, the 1st quartile improved much more than the 4th, but only because the 4th could not test higher than 100% on tests.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is regression to the mean?

A
  • The tendency for scores to become more average over time
  • E.g. Those who score high on a true/false quiz without studying will likely score closer to 50% on the next try because they were always likely to score 50%, it was just chance the first time.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly