December 2, 2024
Every way of using correlation as evidence for causality makes assumptions
Solution | How Bias Solved |
Which Bias Removed |
Assumes | Internal Validity |
External Validity |
---|---|---|---|---|---|
Experiment | Randomization Breaks \(W \rightarrow X\) link |
All confounding variables | 1. \(X\) is random 2. Change only \(X\) |
High | Low |
Conditioning | Hold confounders constant |
Only confounders conditioned on |
1. Condition on all confounders 2. Low measurement error 3. Cases similar in \(W\) |
Low | High |
Feinberg, Branton, and Martinez-Ebers compare hate crimes in counties with and without Trump rallies, condition on (hold constant):
After conditioning on population (a confounder): no correlation.
If we know confounding variables, can we find cases with and without rallies that are the same on many confounding variables?
If we don’t know or can’t measure confounding variables, may still be differences between places with and without rallies that produce confounding.
Taking the same data from Feinberg, Branton, and Martinez-Ebers…
DISCUSS
If we compare counties to themselves before and after rallies…
Which confounding variables are held constant?
What are confounding variables that might NOT be addressed in this comparison?
What kinds of confounding variables are held constant in this before/after comparison?
Conditioning removes confounding by:
Design-based solutions remove confounding by:
Which of these possible confounders are held constant in a before-and-after comparison (month after vs month before rally)?
All confounding variables (affect whether a rally occurs; affect hate crimes) that are unchanging over time are held constant.
Before and after comparisons are design based, because…
And like all solutions to confounding: they make an assumption
Just like experiments and confounding, Before and After comparisons plug in for MISSING potential outcomes.
County | Time | \(Y:\) HC(Yes) | \(Y:\) HC(No) | \(X:\) Rally |
---|---|---|---|---|
\(c\) | Before | \(\color{red}{\text{Hate Crimes}_{c,Before}[\text{Rally}]}\) | \(\color{black}{\text{Hate Crimes}_{c,Before}[\text{No Rally}]}\) | No |
\(\Downarrow\) | ||||
\(c\) | After | \(\color{black}{\text{Hate Crimes}_{c,After}[\text{Rally}]}\) | \(\color{red}{\text{Hate Crimes}_{c,After}[\text{No Rally}]}\) | Yes |
County | Time | \(Y:\) HC(Yes) | \(Y:\) HC(No) | \(X:\) Rally |
---|---|---|---|---|
\(c\) | Before | \(\color{red}{\text{Hate Crimes}_{c,Before}[\text{Rally}]}\) | \(\color{black}{\text{Hate Crimes}_{c,Before}[\text{No Rally}]}\) | No |
\(\Downarrow\) | ||||
\(c\) | After | \(\color{black}{\text{Hate Crimes}_{c,After}[\text{Rally}]}\) | \(\boxed{\color{black}{\text{Hate Crimes}_{c,Before}[\text{No Rally}]}}\) | Yes |
We assume \(\color{red}{\text{Hate Crimes}_{c,After}[\text{No Rally}]} = \\ \color{black}{\text{Hate Crimes}_{c,Before}[\text{No Rally}]}\)
That is: if \(X\) had not changed, \(Y\) would not have changed.
Any \(W\) that affects \(Y\) and changes with \(X\) will produce confounding even if it does not cause \(X\).
Mostly constant upward trend; no change when rallies occur
Over-time comparison, we can create confounding from variables that do not cause \(X\) to change, if they also change with \(X\) over time…
Does rally change measurement, but not actual number of hate crimes? (Measurement bias)
Are there are other changes over the same time-frame (change at the same time as \(X\), rallies)?
It may be that the effects are due to changes in measurement: Anti-Defamation League vs. FBI Hate Crimes give different results.
Why have real wages stayed stagnant?
Starting in the 1980s, automation via robotics/software started to grow.
From “before” to “after” growth of automation, we see slowing or even reversal of growth in real wages:
Can we reasonably conclude the machines are to blame?
Compare the same case to itself before and after change in \(X\)
Holds constant all unchanging attributes of the case.
In order to infer \(X\) causes \(Y\) if \(X,Y\) correlated in before/after comparison
This assumption can be violated if…
Solution | How Bias Solved |
Which Bias Removed |
Assumes | Internal Validity |
External Validity |
---|---|---|---|---|---|
Experiment | Randomization Breaks \(W \rightarrow X\) link |
All confounding variables | 1. \(X\) is random 2. Change only \(X\) |
Highest | Lowest |
Conditioning | Hold confounders constant |
Only variables conditioned on |
1. Condition on all confounders 2. Low measurement error 3. Cases same in \(W\) |
Lowest | Highest |
Before and After | Hold confounders constant |
variables unchanging over time |
1. causes of \(Y\) do not change w/ \(X\) |
Lower | Higher |