Research Design Flashcards

1
Q

ATE

A

E(y1) - E(y0)

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

NATE

A

E(Y1 I D=1) - E(Y0 I D=0)

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

ATE

A

π x ATT - (1-π) x ATC

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

ATT

A

E(Y1 I D=1) - E(Y0 I D=1)

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

ATC

A

E(Y1 I D=0) - E(Y0 I D=0)

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

Total bias

A

NATE - ATE

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

BB

A

E(Y0 I D=1) - E(Y0 I D=0)

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

DTEB

A

(1-π) x [[E(Y1 I D=1) - E(Y0 I D=1)] - [E(Y1 I D=0) - E(Y0 I D=0)]]

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

Internal validity

A

able to identify the causal effect in the study sample

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

external validity

A

ability to generalize the results to other contexts

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

confounder

A

when Z points outwards (do D and Y or other variables)

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

mediator

A

when one arrow points into Z and another out of Z (on the path between D and Y or other variables)

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

collider

A

when arrows point to Z, both variables effect Z (path closed)

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

direct path

A

when there is a causal open path between D and Y

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

non-causal open path

A

when there is an open path between D and Y but not causal (backdoor paths)

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

condition

A

introducing info about a variable, can open or close path dependent on if variable was collider or not

17
Q

IPW

A

D

18
Q

IPW

A

Dx1 / P(D=1, X=X) + (1-D)x1 / P(D=0, X=x)

19
Q

estimate counterfactual

A

mean from the results, ex (D=1, X=1)