POLS 2400 Lecture Notes - Lecture 1: Ceteris Paribus, Causal Inference, Counterfactual Conditional
19 Oct 2018
School
Department
Course
Professor
POLS 2400
9/9/18
Quantitative Techniques Introduction: Causality
1.) Cause and effect
a. What does it mean to say that something caused something else?
b. Example from Quantitative Social Sciences: Researchers applied to jobs with 2 “high
quality” and 2 “low quality” resumes. The resumes were randomly assigned one of each
type of name (a white name or black name). The first hypothesis was that people are less
likely to hire African Americans while the second hypothesis was that people are more
likely to hire African Americans.
c. Counterfactuals: “If things had been the other way, we would have had the other
outcome.” Fundamental problem of causal inference. Counterfactuals don’t exist, we can
never observe them, yet we use them to construct our entire theory of cause and effect.
All other things being equal (ceteris paribus). Example from Quantitative Social
Sciences: Yi (1) - Yi (0): Yi (African-American) - Yi (White).
2.) 4 Basic Concepts for Causality
a. Association of Correlation: The variables must go together consistently. Necessary and
sufficient conditions. Proportional and probabilistic.
b. Time order or temporality: Cause must come before effect
c. No alternative explanations, quantitative techniques helps us sort through these.
d. Cause and effect are connected.
Document Summary
Quantitative techniques introduction: causality: what does it mean to say that something caused something else, example from quantitative social sciences: researchers applied to jobs with 2 high quality and 2 low quality resumes. The resumes were randomly assigned one of each type of name (a white name or black name). Counterfactuals don"t exist, we can never observe them, yet we use them to construct our entire theory of cause and effect. Sciences: yi (1) - yi (0): yi (african-american) - yi (white). 4 basic concepts for causality: association of correlation: the variables must go together consistently.
