Class Notes (838,950)
Canada (511,158)
POL222H1 (44)
Lecture 3

POL222H1 Lecture 3: How to Evaulate Causal Relationships

4 Pages
Unlock Document

Political Science
Kenichi Ariga

POL222 – How to Evaluate Causal Relationship • Reverse causality and simultaneous relationship • Confounding variables o Bivariate relationship between X and Y • X causes Y o We should consider whether there is possibility what could cause X o Is there any credible causal mechanism linking Y to x? o Are both causal mechanisms from X to Y and from Y to X plausible? ▪ E.g., gun ownership  crime ▪ The greater the gun ownership in the neighbourhood, the larger the serious crimes in the neighbourhood ▪ Is there causal mechanism in which crime rate can increase or decrease ▪ Gun ownership  crime • People protect themselves from crimes with guns o A systematic relationship between X and Y, we cannot tell whether X causes Y or vice versa or both o Our causal theory – X causes Y ▪ E.g., economic performance  election outcomes ▪ Economic performance election outcomes • There is no theory that future election outcome will affect economic performance • Time order will define whether the claim is plausible • Confounding variables o Independent variable X  dependent variable Y ▪ E.g., college education  political participation ▪ Causal argument = people will have greater information ▪ College education is likely to increase for political participation ▪ Z  political participation o If there is a relationship between X and Z , Z is called a cofounding variable ▪ It not only causes Y but also causes X o If there is a certain relationship between X and Z which also causes Y, then it is a confounding variable o Focus on X Z, X Y, Z Y o If Z causes X, then it’ll be a confounding variable [whether or not they’d go to college] • Spurious Relationship o When we look at a bivary relationship between X and Y, if we account for a confounding relationship, then relationship between X and Y might be a spurious relationship ▪ E.g., suppose you were an elementary school teacher, trying to improve student’s academic achievement ▪ A sample of elementary school children gr.4-6 ▪ Dependent variable: score of achievement test [math and English] ▪ Independent variable: number of hours spent on homework [math and English] ▪ Various characteristics of children o Spurious relationship is caused by a confounding variable ▪ Height (cm)  math test score ▪ Should not be the relationship between height and math test score ▪ Confounding variable  Grade ▪ Relationship between grade and height, and grade and math test score ▪ The people who do better tend to be taller o No relationship between height and math test score ▪ E.g.2, Data: 1965-1973 Waves of political socialization study ▪ Dependent variable: an index of 8 political participatory acts (voting, campaign worker, demonstration): 1 = lowest, 8 = highest ▪ Independent variable: college attendance, family background, individual characteristics ▪ X = college education ▪ Y=political participation ▪ Z=family background ▪ Simple comparison • Attended college: 2.79 participated • Did not attend college: 1.43 ▪ With confounding variables • Attended college: 2.79 • Did not attend college: 2.77 o Debate whether it’s true or not: seeing the study whether increase or decrease but does not determine our knowledge o The relationship between X and Z may modify an apparent relationship between X and Y • Omitted
More Less

Related notes for POL222H1

Log In


Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.