8. Interactions: Categorical*Categorical Flashcards

1
Q

What is a categorical*categorical interaction?

A

A difference in differences between groups across levels of a second factor

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

When visualising a categorical*categorical interaction, what does it mean if lines are parallel?

A

There is no interaction

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

When visualising a categorical*categorical interaction, what does it mean if lines are not parallel but not crossing?

A

Ordinal interaction

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

When visualising a categorical*categorical interaction, what does it mean if lines are not parallel but are crossing?

A

Disordinal interaction

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

What is the value of beta 0 in a categorical*categorical interaction?

A

Value of y when x + z are 0

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

What is the value of beta 1 in a 2x2 categorical interaction?

A

Difference between x and z =0

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

What is the value of beta 2 in a 2x2 categorical interaction?

A

Difference between levels of z when x = 0

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

What is the value of beta 3 in a 2x2 categorical interaction?

A

Difference between x across levels of z

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

How do you visualise categorical*categorical interactions?

A

Via cat plot

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

What do we do if there are more 2x2 variables?

A

Need additional interaction terms

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

In the hospital example, a 3 x 2 interaction, what was the model and how were the coefficients interpreted?

A

yijk = b0 + b1D1 + b2D2 + b3D3 + b4D13 + B5D23

b0 = mean of treatment a in hospital 1
b1 = Difference between TA and TB in H1
b2 = Difference between TA and TC in H1
b3 = Difference of TA in hosp 1 and 2
b4 = Difference between TA and TB in H1 and H2
b5 = Difference between TA and TC in H1 and H2

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

What is the rule on number of interaction terms?

A

(r-1) * (c - 1)

Row = Number of levels in first categorical variable
Column = Number of levels in second categorical variable

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

Why are three way interactions not as common?

A

Requires a lot of data to interpret as effects are often small
Extends issue of power
Only test if strong and theoretical reason to

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