Class Notes (836,562)
POL3371 (20)
Lecture

3 Pages
95 Views

School
Department
Political Science
Course
POL3371
Professor
Ivan Katchanovski
Semester
Winter

Description
March 25, 2014 Logistic Regression Binary Logistic Regression  Appropriate when the dependent variable is a dummy variable o Dummy variable: a variable the includes two categories which assume values 1 and 0 o Example: “Voted”: Yes = 1, No = 0  One or many independent variables  Assumes non-linear relationship Regression Coefficients  Interpretation is similar to interpretation of unstandardized regression coefficients in linear regression o Effect of ca change of one unit of an independent variable on the logged odds of the dependent variable. Odds Ratio … Statistical Significance  Statistical significance of a regression coefficient: o Statistically significant if p(obtained) < .05 or .01 or .001 o Statistically non-significant if p(obtained) > .05 o For smaller samples .1 level of statistical significance can also be used. Pseudo R Square  R square analogs in logistic regression o Power of independent variables in predicting the dependent variable  Cox & Snell R square o Ranges between 0 (no association) and less than 1 (perfect association)  Nagelkerke R square o Adjusts Cox & Snell R square so that its maximum value can equal 1 o Ranges between 0 (no association) and 1 (perfect association) Binary Logistic Regres
More Less

Related notes for POL3371
Me

OR

Join OneClass

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

Join to view

OR

By registering, I agree to the Terms and Privacy Policies
Just a few more details

So we can recommend you notes for your school.