POL 3371 Lecture Notes - Linear Regression, Statistical Significance, Regression Analysis

• School: University of Ottawa
• Department: Political Science
• Course: POL 3371
• Professor: Ivan Katchanovski

Document Summary

Multiple regression: similar to bivariate regression, but assesses effects of many independent variables on the dependent variable. March 18, 2014: widely used in political science research, simultaneous test of many hypotheses, interval/ratio variables, ordinal variables can also be used for practical reasons, nominal variables cannot be used (need to transform them into dummy variables) Unstandardized regression coefficients: the amount of change produced in the dependent variable by a unit change in an independent variable keeping other independent variables constant. Direction of association of regression coefficients: positive, negative. Statistical significance of unstandardized regression coefficient: statistically significant if p(obtained) < 0. 05 or 0. 01 or 0. 001, statistically non-significant if p(obtained) > 0. 05, for smaller samples criteria of statistical significance also include 0. 1. Multiple regression: spss commands: analyze regression linear, dependent box: select dependent variable, independent box: select independent variable, method: enter , estimates of regression coefficients , model fit . V64 dependent percentage of women in lower house of parliament (2009)