Introduction to Panel Data: Does the Death Penalty Reduce Homicides?: Causal Inference Bootcamp

Often we have data on units at multiple points in time——that’s called panel data. We introduce the main approach to using panel data to learn about causality in this module. We illustrate this idea by looking at the effect of the death penalty on homicide rates.

[barovierkevinallybose]

I am just learning this topic but at 4:15 where you talked about the assumptions, shouldn't we be considering other assumptions, such as the number of states from which we garner these numbers and other external factors?

[sj-oswx]

Sir how to contact you??? You are not present on duke university social science institute website?? I need your guidance. You teach so well.

[orjihvy]

But if economic slowdown had a greater negative effect on one state than another leading to a different environment and consequent homicide rates, then I guess the common trend assumption would not hold in that scenario?

[orjihvy]

It seems like no matter what, statistical interpretation of data is very much an art or judgement. And can’t be as conclusive as physical sciences.

[faanlouw]

Why should the death penalty reduce murders? There is no logical reason to assume that the death penalty will be in the thinking of the person while commuting his crime. What is the moral and ethical reasoning behind the death penalty? Justice, revenge, hate, believe systems could be better arguments than the argument you offered. You are trying to use numbers to motivate an argument that is not founded in numbers but something else, a perfect example of not having identified your theoretical causal effects correctly.