From Description to Causation Department of Government London - - PowerPoint PPT Presentation

▶
from description to causation
SMART_READER_LITE
LIVE PREVIEW

From Description to Causation Department of Government London - - PowerPoint PPT Presentation

Sep 01, 2022 260 likes •788 views

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality From Description to Causation Department of Government London School of Economics and Political Science Correlation vs. Causation Over-Time Changes Getting

slide-1
SLIDE 1

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

From Description to Causation

Department of Government London School of Economics and Political Science

slide-2
SLIDE 2

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

1 Correlation vs. Causation 2 Over-Time Changes 3 Getting Systematic about Causality

slide-3
SLIDE 3

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

What makes something a cause?

Write for 1 minute.

slide-4
SLIDE 4

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

1 Correlation vs. Causation 2 Over-Time Changes 3 Getting Systematic about Causality

slide-5
SLIDE 5

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Correlation

Correlation is the non-independence of two variables for a set of observations

slide-6
SLIDE 6

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

slide-7
SLIDE 7

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Correlation

Synonyms: correlation, covariation, relationship, association Any correlation is potentially causal

X might cause Y Y might cause X X and Y might be caused by Z X and Y might cause Z There may be no causal relationship

slide-8
SLIDE 8

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Flashback!

Two Categories of Inference:

1 Descriptive Inference

What are the facts?

2 Causal Inference

Why does something occur?

slide-9
SLIDE 9

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Correlation is Causation?

The mind tends to interpret correlations and patterns as evidence

  • f causal relationships!

But this is rarely correct!

slide-10
SLIDE 10

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Source: Wikimedia Commons

slide-11
SLIDE 11

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Source: Wikimedia Commons

slide-12
SLIDE 12

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Source: The Economist, 8 July 2016

slide-13
SLIDE 13

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Source: The Economist, 8 July 2016

slide-14
SLIDE 14

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Source: Randal Olson

slide-15
SLIDE 15

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Source: Ministry of Justice, “Statistics on Race and the Criminal Justice System 2010”

slide-16
SLIDE 16

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Source: StackExchange

slide-17
SLIDE 17

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Source: TylerVigen.com

slide-18
SLIDE 18

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Naive Causal Inference

Correlations are not necessarily causal Our mind thinks they are because humans are not very good at the kind of causal inference problems that social scientists care about Instead, we’re good at understanding physical causality

slide-19
SLIDE 19

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Physical causality

Action and reaction

slide-20
SLIDE 20

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Physical causality

Action and reaction Example:

Picture a ball resting on top of a hill What happens if I push the ball?

slide-21
SLIDE 21

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Physical causality

Action and reaction Example:

Picture a ball resting on top of a hill What happens if I push the ball?

Features:

Observable Single-case Deterministic Monocausal

slide-22
SLIDE 22

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

1 Correlation vs. Causation 2 Over-Time Changes 3 Getting Systematic about Causality

slide-23
SLIDE 23

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Pre-Post Change Heuristic

Our intuition about causation relies too heavily on simple comparisons of pre-post change in outcomes before and after something happens

No change: no causation Increase in outcome: positive effect Decrease in outcome: negative effect

slide-24
SLIDE 24

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Pre-Post Change Heuristic

Our intuition about causation relies too heavily on simple comparisons of pre-post change in outcomes before and after something happens

No change: no causation Increase in outcome: positive effect Decrease in outcome: negative effect

Why is this flawed?

slide-25
SLIDE 25

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Threats to Validity

Campbell and Ross talk about six “threats to validity” (i.e., threats to causal inference) related to time-series analysis

slide-26
SLIDE 26

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Flaws in causal inference from pre-post comparisons

1 Maturation or trends 2 Regression to the mean 3 Selection 4 Simultaneous historical changes 5 Instrumentation changes 6 Monitoring changes behaviour

slide-27
SLIDE 27

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Maturation or trends

Is a shift in an outcome before and after a policy change the impact of the policy

  • r a small part of a longer time trend?

Case Study: Connecticut crackdown on speeding (1955)

slide-28
SLIDE 28

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

slide-29
SLIDE 29

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Regression to the mean

Is a shift in an outcome before and after a policy change the impact of the policy

  • r a function of statistical variation?

Case Study: Connecticut crackdown on speeding (1955)

slide-30
SLIDE 30

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

slide-31
SLIDE 31

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Selection

Is a shift in an outcome before and after a policy the impact of the policy or the result of the policy being implemented when outcomes are extreme? Case Study: Connecticut crackdown on speeding (1955)

slide-32
SLIDE 32

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

slide-33
SLIDE 33

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Simultaneous changes

Is the shift in an outcome before and after a policy the impact of the policy or the result of a simultaneous historical shift? Case Study: US Great Depression Policy

slide-34
SLIDE 34

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

slide-35
SLIDE 35

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Instrumentation changes

Is the shift in an outcome before and after a policy the impact of the policy or a change in how the outcome is measured? Case Study: Age-adjusted mortality rates

slide-36
SLIDE 36

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

slide-37
SLIDE 37

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Monitoring changes behaviour

Is the shift in an outcome before and after a policy the impact of the policy or a change in response to measuring the

  • utcome per se?

Case Study: Educational testing

slide-38
SLIDE 38

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

The more any quantitative social indicator is used for social decision-making, the more subject it will be to corruption pressures and the more apt it will be to distort and corrupt the social processes it is intended to monitor. – Donald T. Campbell (1979)

slide-39
SLIDE 39

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Flaws in causal inference from pre-post comparisons

1 Maturation or trends 2 Regression to the mean 3 Selection 4 Simultaneous historical changes 5 Instrumentation changes 6 Monitoring changes behaviour

slide-40
SLIDE 40

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

slide-41
SLIDE 41

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

slide-42
SLIDE 42

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

1 Correlation vs. Causation 2 Over-Time Changes 3 Getting Systematic about Causality

slide-43
SLIDE 43

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Directed Acyclic Graphs

Causal graphs (DAGs) provide a visual representation of (possible) causal relationships

slide-44
SLIDE 44

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Directed Acyclic Graphs

Causal graphs (DAGs) provide a visual representation of (possible) causal relationships Causality flows between variables, which are represented as “nodes” Variables are causally linked by arrows Causality only flows forward in time

slide-45
SLIDE 45

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Directed Acyclic Graphs

Causal graphs (DAGs) provide a visual representation of (possible) causal relationships Causality flows between variables, which are represented as “nodes” Variables are causally linked by arrows Causality only flows forward in time Nodes opening a “backdoor path” from X → Y are confounds “Selection bias” or “Confounding”

slide-46
SLIDE 46

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Smoking Cancer

slide-47
SLIDE 47

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Smoking Cancer Age Environment Genetic Predisposition Parental Smoking

slide-48
SLIDE 48

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Smoking Cancer Age Environment Genetic Predisposition Parental Smoking

slide-49
SLIDE 49

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Smoking Cancer Age Environment Genetic Predisposition Parental Smoking

slide-50
SLIDE 50

Correlation vs. Causation Over-Time Changes Getting Systematic about Causality

Causal Inference

Causal inference (typically) involves gathering data in a systematic fashion in

  • rder to assess the size and form of

correlation between nodes X and Y in such a way that there are no backdoor paths between X and Y by controlling for (i.e., conditioning on, holding constant) any confounding variables, Z.

slide-51
SLIDE 51