9. Understand when to use Linear versus Logistic Regression (if the dependent variable is binary, use logistic regression). 10. Understand and recognize the R command for building a linear regression model. Ex. : > fit = glm(d$donationbinary ~ d$age + d$householdincome + d$pctemployment + d$householddependents, family="binomial") 11. Understand what statistical significance means and how to test for it. What does it mean if a variable is not statistically significant? 12. To test for significance at standard levels of significance (e.g., a=0.05) use stars or p-values (if p-value is less than the level of significance a, then the variable/coefficient is statistically significant). P-values allow you to test statistical significance at non-standard levels of significance (e.g., a=0.0375). 13. Understand how to build a logistic model equation after you have tested for statistical significance, using the R output coefficients (the left side of the model equation is always the log odds ratio associated with the dependent variable. Note: you need to define what p is). 14. Make sure you know how to interpret the log graphs (slide 10-14 in Lecture 11) and understand the relationships between odds ratio, log odds ratio, and probability of success/failure. For example, what does it mean when log odds ratio is greater (or less) than zero? What does it mean when odds ratio is greater (or less) than 1. What happens to log odds ratio if you increase an independent variable with positive (or negative) coefficient (Lecture 12)? What happens to the probability of success/failure? 15. Understand ALL assumptions and requirements and how to check for each one (you will not be asked to test linearity using logs - Assumption 1). Understand what charts and tests to use and how to interpret the different charts and tests. 16. Make sure you know how to test for No Complete Separation and how to plot a binary variable on the y-axis vs. an independent variable on the x-axis (lecture 13) 17. Know how to interpret R output for Linear and Logistic regression (e.g., statistical significance, interpretation of intercept and coefficients) and how to interpret visual tests for each assumption and requirement. 18. Models with not adequate sample size (eg. less than 150 observations) will have poor performance, therefore, you need to collect more data