Block 1: X-centered RDs: Causal strategies: X and Y (Lesson 3) Flashcards Preview

Social Science Methodology > Block 1: X-centered RDs: Causal strategies: X and Y (Lesson 3) > Flashcards

Flashcards in Block 1: X-centered RDs: Causal strategies: X and Y (Lesson 3) Deck (42)
Loading flashcards...
1
Q

What is the golden standard for x-centered research?

A

The experimental template

Outcome (Y) is less important, the focus is on the effect on the population

Any deviations from this template are viewed as sources for bias

2
Q

Definition of ontology

A

The nature of being, existence or reality

3
Q

Definition epistemology

A

Theory of knowledge: It’s nature and the limitations thereof

4
Q

Definition of ethics/ aesthetics

A

What value does an observation have?

5
Q

Different types of causal relationships

A

CEMS LICS PCPCP

Conjunctures
Equifinality
Monotonicity
Sequence

Linearity
Irreversibility
Constancy
Set-theoretic causes

Proximal
Causal chain
Path-dependency
Causal laws
Probibalistic causes
6
Q

Conjunctures (causal relationship)

A

A combination of causes that produce an effect

7
Q

Equifinality

A

Several causes acting independently of each other, but lead to the same effect/ outcome

8
Q

Monotonicity

A

Where an increase (decrease) in X always causes an increase (decrease) in Y

9
Q

Linearity

A

Rise in X causes a predictable rise in Y, explainable by a linear relationship

10
Q

Irreversibility

A

One way relationship

X affects Y as X increases but not as it decreases (or vice versa)

11
Q

Constancy

A

A constant cause operates continually upon an outcome

12
Q

Proximal

A

A proximal cause operates immediately

13
Q

Sequence

A

The effect of X(1-3) on Y depends upon the sequence they appear in

14
Q

Causal chain

A

Multiple factors (M) form a chain between X and Y

15
Q

Path-dependency

A

A single causal intervention has enduring, and perhaps increasing, effects over time

16
Q

Causal laws

A

Exception-less relationships between X and Y

17
Q

Probabilistic causes

A

Relationship with errors, i.e., exceptions, which can be given a certain probability of occurring

18
Q

Set-theoretic causes

A

Where X is necessary and/or sufficient for Y

19
Q

Criteria for good causal analysis on the treatment variable (X)

p
e
e
v
s

s
u
d
s

A
Is X 
P =  proximate to Y,
E = exogenous to Y, 
E = evenly distributed,
V = varying, 
S = simple, 
S = strong, 
U = uniform, 
D = discrete, 
S = scaleable?
20
Q

Criteria for good causal analysis on the outcome variable (Y)

A

Is Y free to vary?

21
Q

Criteria for good causal analysis on the sample

A

Are the chosen observations (a) independent (of one another) and
(b) causally comparable?

22
Q

Definition of independence (sample criteria)

A

Each observation is seperate and gives new evidence of the causal linkage

Changes in Y need to be due to the treatment, not because the units themselves are influncing each other

23
Q

Definition of causal comparability (sample criteria)

A

The average value of Y for a given value of X should remain the same across units and during the period of analysis

24
Q

Incomparabilities which may influence causal comparability

A

Noise (B), which is random and only influences Y

Confounders (C), non-random and influence both X and Y

25
Q

Strategies for causal inference

A

Randomized designs (experimental)

Non-randomized designs (non-experimental, observational)

Beyond X and Y

26
Q

Randomized designs (strategies)

A

– Pre-test/post-test,

– Post-test only,

– Multiple post- tests,

– Roll-out,

– Crossover,

– Solomon four-group,

– Factorial

27
Q

Non-randomized designs (strategies)

A

– Regression discontinuity,

– Panel,

– Cross-sectional,

– Longitudinal

28
Q

Beyond X and Y (strategies)

A

– Conditioning confounders,

– Instrumental variables,

– Mechanisms,

– Alternate outcomes,

– Causal heterogeneity,

– Rival hypotheses,

– Robustness tests,

– Causal reasoning

29
Q

Generally on randomized designs and their internal/ external validity

A

Internal validity and the assignment problem are per definition solved, though post-treatmen threats can be present

External validity can still be problematic

30
Q

Give two examples of pre-treatment/ assignment bias?

A

1) Common-cause confounder: Confounder C affects both the treatment (X) and outcome (Y)
2) Self-selection bias: Treatment assignment is done by the subjects of the study

=> Solved through randomization!

31
Q

Post-Treatment Bias, examples

A

Attrition: The loss of subjects during the course of a study

Noncompliance: When subjects do not comply with instructions

Contamination: Where treatment and control groups are not isolated from one another

Reputation effects: Where the reputation of the treatment in the minds of subjects affects an outcome

Researcher (Hawthorne) effects: Where the condition of being studied affects an outcome

Testing effects: Where responses to a test are influenced by a previous testing experiences, rather than the treatment itself

32
Q

Examples of Pre/Post-Treatment Bias in Longitudinal Studies

A

– History: Where the treatment is correlated with some other factor that affects the outcome, which is to say, where the variation over time is driven by some factor other than the treatment.

– Regression to the mean: Where some change observed over time is a product of stochastic variation rather than the treatment of interest.

– Instrumentation effects: A change in the measurement of an outcome over the course of a study which alters the estimate of X’s effect on Y (learning effects).

=> Remain present even when randomization is used!

33
Q

Stochastic variable

A

Variable which is completely random with no systemic component (statistics)

34
Q

Generally on Non-Randomized Designs

A

– When randomization isn’t possible, the assignment problem has to be dealt with

– Except in case of (perfect) natural experiments, non- randomization is the standard in studies using observational data

– The fundamental problem: We do not know the ‘data generating process’ (DGP), because we did not create the data ourselves

35
Q

Cartwright’s socio-economic machine argument

A

We cannot compare countries because they have different socio-economic backgrounds and we know little about the data-generation process

36
Q

Non-randomized designs:

Regression-Discontinuity Design (RDD)

A

Assignment principle:

(a) known;
(b) measurable (prior to treatment, for all units in the sample);
(c) Has a cutoff point, which defines the assignment of subjects, producing a binary treatment variable;
(d) many units fall on either side of this cutoff;
(e) this principle is maintained throughout the period of study

37
Q

Non-randomized designs:

Panel Design

A

Multiple observations are taken from each unit and there variation in X across time/ units.

Two types:

1) Difference-in-difference (DD)
2) Fixed effect

38
Q

Difference-in-difference (DD) panel design formula

A

Y = B + T + X + T*X

B = covariates 
T = time dummy 
X = treatment
39
Q

Types of autocorrelation

A

1) Spatial autocorrelation: Non-independence because observations are correlated across space (neighborhood effect)
2) Serial autocorrelation: Non-independence because observations are correlated across time (temporal stickiness)

40
Q

Non-randomized designs:

Cross-Sectional Designs

A

Based on post-test observations, where X and Y vary spatially (but not over time)

Advantage: Creates fewer statistical (method) problems

41
Q

Non-randomized designs:

Longitudinal Designs

A

Variation in variables is longitudinal (over time) but not cross-sectional, thus all units are treated

Treatment effect is found by comparing pre-treatment and post-treatment status

42
Q

Sub-types of the longitudinal design

A

1) Interrupted time-series: Single intervention affects the units, which are observed ofer time, before and after the intervention
2) Repeated observations: Units receive the same treatment multiple times