1. Review
- causality as counterfactual
- potential outcomes
2. Types of Causal Claims
- deterministic causal claims
- probabilistic causal claims
3. Testing Causal Claims
- Fundamental Problem of Causal Inference
October 28, 2024
All causal claims are claims about how the world would be changed in an alternate timeline in which some thing (or things) were different than they actually are.
These alternate timelines/universes are counterfactuals
“The expansion of NATO into Eastern Europe caused Russia to invade Ukraine”
implicitly claims that…
in the counterfactual world where NATO did not expand (the “cause” is not present), Russia would not have invaded Ukraine in February 2022 (the “effect” would be different).
potential outcomes are values for variables that describe the factual world (that has occurred) and counterfactual worlds (that have not).
\(\mathrm{Russian \ Invasion}_{Ukr}(\mathrm{E. \ Europ. \ NATO \ Memb.} = 0) = ?\) \(\mathrm{Russian \ Invasion}_{Ukr}(\mathrm{E. \ Europ. \ NATO \ Memb.} = 14) = ?\)
Which of these potential outcomes is factual? Counterfactual?
\(\mathrm{Russian \ Invasion}_{Ukr}(\mathrm{E. \ Europ. \ NATO \ Memb.} = 0) = ?\) \(\mathrm{Russian \ Invasion}_{Ukr}(\mathrm{E. \ Europ. \ NATO \ Memb.} = 14) = ?\)
What are the values of these potential outcomes if the following claim is true?
“The expansion of NATO into Eastern Europe caused Russia to invade Ukraine”
“The expansion of NATO into Eastern Europe caused Russia to invade Ukraine”
If this causal claim were true: then it implies these potential outcomes:
\(\color{red}{\mathrm{Russian \ Invasion}_{Ukr}(\mathrm{E. \ Europ. \ NATO \ Memb.} = 0) = \mathrm{No}}\) \(\mathrm{Russian \ Invasion}_{Ukr}(\mathrm{E. \ Europ. \ NATO \ Memb.} = 14) = \mathrm{Yes}\)
(red indicates \(\color{red}{\mathrm{counterfactual}}\))
It follows that, all causal claims can be re-stated as counterfactual claims
“News coverage of the discovery of unmarked graves at the Kamloops Residential School increased settler Canadian support for Truth and Reconciliation.”
\[\overbrace{\text{If there had not been news coverage of the graves}}^{\text{If-clause in Subjunctive Mood}}, \\ \underbrace{\text{there would be less support for Truth and Reconciliation}}_{\text{Then-clause in Conditional Mood}}\]
Note: Counterfactual claims get increasingly complicated, the more complicated your causal claim is
With your neighbors: turn these causal claims into counterfactual claims.
Usually… different focus leads to different kinds of causal claims
And different types of causal claims imply different counterfactuals/potential outcomes, different forms of evidence.
deterministic causal claims
claims about what happens with certainty under specific causal conditions
There are several varieties and combinations
A causal claim that there is some cause \(C\) without which the effect \(E\) cannot occur
A claim: “If Canada had not signed the UN Declaration on the Rights of Indigenous Peoples, the Blueberry River First Nation would not have been able to successfully challenge BC’s permitting of industrial activities on their ancestral lands.”
Also can be stated: “The signing of the UNDRIP by Canada was a necessary condition for the Blueberry River First Nation to be able to successfully challenge BC’s permitting of industrial activities on their ancestral lands.”
Head to menti.com and use code \(4959 \ 5430\)
If this claim is true: “The signing of the UNDRIP by Canada was a necessary condition for the Blueberry River First Nation to be able to successfully challenge BC’s permitting of industrial activities on their ancestral lands.”…
The fact that Canada signed the UNDRIP does not mean that the the BRFN’s legal victory over BC was inevitable.
Claims about necessary conditions have specific implications about potential outcomes:
If we say that: “economic crisis is a necessary condition for populist dictatorship to replace democracy”
It implies the potential outcomes in for democratic country \(i\):
\(\mathrm{Dictatorship}_i \ (\mathrm{Economic \ Crisis = No}) = \mathrm{No}\)
\(\mathrm{Dictatorship}_i \ (\mathrm{Economic \ Crisis = Yes}) = \mathrm{Yes} \ or \ \mathrm{No}\)
Something else might need to happen, in addition to economic crisis, for dictatorship to arise.
(In contrast to necessary conditions)
“A military coup that overthrows a democratically elected government is a sufficient condition for large public protests.”
Sufficient conditions also imply specific potential outcomes:
“A military coup that overthrows a democratically elected government is a sufficient condition for large public protests.” implies that for every democratic country \(i\):
\(\mathrm{Protests}_i \ (\mathrm{Military \ Coup = No}) = \mathrm{No \ or \ Yes}\)
\(\mathrm{Protests}_i \ (\mathrm{Military \ Coup = Yes}) = \mathrm{Yes}\)
Aired this ad on YouTube and then surveyed people in “treatment” and “control” conditions to see if they recognize the misinformation tactic
menti.com \(7452 \ 3898\)
Does it make sense to say that “being inoculated” is a necessary condition for spotting misinformation?
Does it make sense to say that “being inoculated” is a sufficient condition for spotting misinformation?
are claims that the presence/absence of a cause \(C\) makes an effect \(E\) more or less likely to occur. Or cause \(C\) increases/decreases effect \(E\) on average
Causality may be deterministic… there are exact conditions for when effect always/never happens.
But in reality, it is almost always complex
Because causality is complex, we do not fully know the deterministic rules…
\(C\) appears to only cause a change in the probability or likelihood of seeing the effect \(E\).
Which are probabilistic causal claims?
Which is a probabilistic causal claim?
Not every probabilistic statement is causal
In this course, we focus on how to provide evidence that that pertain to claims about effects of causes rather than causes of effects.
A claim for today:
NDP’s Bill 44 (abolishing single-family zoning restrictions to permit multi-family units) reduced the average cost of buying a house.
Causal claims imply relationships between potential outcomes
\(\text{Housing Price}_{BC}(\text{Bill 44}) < \\ \color{red}{\text{Housing Price}_{BC}(\text{No Bill 44})}\)
\(\mathrm{Black}\) indicates factual potential outcomes (we observe this state of the world)
\(\color{red}{\mathrm{Red}}\) indicates counterfactual potential outcomes (we do not observe this state of the world)
Average housing price: \(\text{Housing Price}_{BC}(\text{Bill 44}) = \$990,500\)
\(\color{red}{\text{Housing Price}_{BC}(\text{No Bill 44})} = \ \mathbf{????}\):
For BC, we can only observe the potential outcome of \(\text{Housing Price}_{BC}\) for where the value of \(\text{Bill 44} = Yes\): the actual policy that occurred in BC.
We can never observe the other, counterfactual, potential outcomes of \(\color{red}{\text{Housing Price}_{BC}}\) where \(\color{red}{\text{Bill 44} = No}\), because that was not the actual policy.
We can never empirically observe, for BC, whether \(\text{Bill 44}\) caused \(\downarrow \text{Housing Price}\).
By definition, \(X\) causes \(Y\) if the value of \(Y\) were different if we changed \(X\) for the exact same case.
For a specific case, we can only observe the potential outcome of \(Y\) for the value of \(X\) it actually takes.
We never observe the counterfactual potential outcomes of \(Y\) for different possible values of \(X\) that the case did not experience.
We can never empirically observe, for a specific case, whether \(X\) causes \(Y\).
I thought evidence for empirical claims based on observing the world?!
Does this mean that all evidence for causal claims fails weak severity?
Are there “solutions” to this fundamental problem?