PS102Lecture23

PS102Lecture23 - Multivariate Regression Multivariate PS...

Info iconThis preview shows pages 1–8. Sign up to view the full content.

View Full Document Right Arrow Icon
Multivariate Regression Multivariate Regression PS 102 Introduction to Political Inquiry Lecture 23
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Multivariate Analysis Multivariate Analysis What if Y has more than one cause? We can simultaneously investigate relationships with more than two variables Including more than two variables allows us to “control” for a spurious relationship Statistical control: account for common variation among X’s in association with Y Not the same as EXPERIMENTAL control: Random treatment assignment
Background image of page 2
Multivariate Cross-tabulation Multivariate Cross-tabulation In a multivariate cross-tabulation, you can control for a third variable by holding it constant. Example: Consider the relationship between party identification (party id) and candidate preference. Candidate Preference = b 0 + b 1 Party Identification + e
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Multivariate Cross-tabulation Multivariate Cross-tabulation Using a bivariate cross-tabulation, examine the two variables of interest, party id and candidate preference. But, we what if we think that gender also matters? Candid ate pref. Kerry Bush N Republi can 33% 67% 800 Democ rat 68% 32% 750 N 774 776 1,550 Candidate Pref. = b 0 + b 1 Party ID + b 2 Gender + e
Background image of page 4
Multivariate Cross-tabulation Multivariate Cross-tabulation To control for a third variable (gender), we first group the observations by their values on gender Then observe the original relationship within each group Effective, but cumbersome for many variables Candid ate Pref. Kerry Bush Total Male Republi can 17% 83% 400 Democ rat 44% 56% 375 Total 233 542 775 Female Republi can 60% 40% 400 Democ rat 80% 20% 375 Total 540 235 775
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Multivariate Regression Multivariate Regression We have developed a bivariate regression estimator Provides good summary measures of the strength and fit of the relationship between X and Y Developed methods to use the estimator to test hypotheses derived from theories about X and Y But we have only 1 X (and only 1 b) The world is more complicated than that!
Background image of page 6
Multiple Regression Analysis Multiple Regression Analysis We can develop a simple extension of
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 8
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 10/21/2011 for the course POL SCI 102 taught by Professor Gelpi during the Spring '11 term at Duke.

Page1 / 23

PS102Lecture23 - Multivariate Regression Multivariate PS...

This preview shows document pages 1 - 8. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online