Class Notes (836,128)
Canada (509,645)
POLC78H3 (5)
Lecture

Lec 4 jan 30th.docx

5 Pages
196 Views
Unlock Document

Department
Political Science
Course
POLC78H3
Professor
Renan Levine
Semester
Winter

Description
th Lec 4 jan 30 Inference and causal relationship 2 types of inference: • Descriptive inference-eg indepth studies looking at neighbourhood change and how some ethnic goups disapperared as neighbor hoods changed • Collect information abt just boston but really talking abt everywhere else; so its abt filling in gaps; looking at few things and appling it wholly • Causally inference-mechanism as to why things are happening • Key problem:Shively reading-storks and babies-storks-universal symbol of fertility-storks are enormous birds and are not in cities but rural areas; rural areas have higher birth rates; storks in Poland and not Canada ; catholics in maj in Poland; so does storks cause higher birth rate? NO. but we have have phenomenas coinciding • 3 reasons when phenomena coincide • 1) causal relationship • 2)logical necessity or tautoligcal determination- one comes after the other- winter comes after summer • 3) spurious cause- when it appears that 2 variables/ phenomensas are correlated but in rd reality there is 3 phenomena/variables that actually cause both of them so there isn’t actually a causal relationships; eg blonds are more likely to be republicans in US. • Is there a causal relationship b/w hair color and poltical preference? No • Blonde test • How can we eliminate alt causal interpretations? Two ideas that coincide that blondes are more likely to vote for republicans • Relation b/w hair color and partisanship is spurious • To test for supriousness; we need to identify another variable and test it against the 2 ; eg maybe is it someones background-control variable; so take 2 ppl frm same background, one blonde and one not then see wht their preference is;so if they have same partisanship, we can say that hair color doesn’t matter; introduce a variable and hold it constant-holding it at same level • Controlling blondes • Frm reading-Another possible variable- WASPs • if there was a causal relationship, why would it be diff for WASPs?- • DOES INCOME INFLUENCE THE US VOTE? • Race is antecedent to both income and vote • Not strongly income doesn’t influence us vote; what if we control with race • Spuriousness and Causality • Ability to rule out alt cause may not always be possible • Categorizing Phenomena • We can never be 100% sure abt relationshops • Indeterminate • Some relationships may prove to be indeterminate. • WHY? • Data or measurement issues- sampling didn’t work; asked a question in a poorly worded manner • Variation in multiple ind variables renders them too similar to separate- we have diff ind variables and using to test them and we cant tell which one it is • Causal order is unclear or unstable- dnt knw what comes first changes over time • Too many explanations, not enuff observations • Causal order is uncler or unstable • Due to reciprocal causation • Error term-part of an analysis that we cant explain eg random noise, person didn’t understand the question • Policy preference- eg civil war fought over natural resources or are natural resources presence driving the civil war • Overdetermination=more e
More Less

Related notes for POLC78H3

Log In


OR

Join OneClass

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

Sign up

Join to view


OR

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.


Submit