Class Notes (837,023)
Canada (510,044)
Sociology (4,081)
SOC350H5 (9)
Lecture 5

SOC350H5 Lecture 5: Lecture 5

3 Pages
Unlock Document

David Pettinicchio

Lecture 5  We want to interpret slope and if model is a good model – explained variance  Because it’s a bivariate mode – only one variable explaining  Conceptually there is no outcome that can be explained by just one variable  Multitude of factors – other variables that are supposed to influence the dep variable  Control for other expected factors that would explain that outcome  To do so you have to do multivariate reg  Pay attention to slope, sig, r sq  Want to know if you make model better or worse when you add other variation  Stan beta – compare slopes  Stand coefficient – allows you to talk about magnitude of effect Multivariate reg  Additive models – full model table  Which includes all the indep variable tha you pla on including  I des with multiple reg is to see what happens to dep variable as you add  You aren’t just putting in variables for just putting variables  This allows you to design models that speak to competing explanation for why things look the way they do  Take into account, other effects  How much that helps explain the outcome  What does lit say – have proof for including variables  You want to add variables that are theoretically meaningful  Whatever you do for putting in models has to be motivated by general belief in lit review  Bivariate regression – a reg where we predict (y) – with inequality Multivariate reg  Ineq but also involve GDP  2 x variable  inequality is no longer statistically sig  When you start with hyp that says ineq influences homicide  You add other variables  Are you explaining more variation than other variable? Increase in R sq  This is a better model than the first one  Model can suffer because of things you didn’t control for variables that other people say matter  When you get the SPSS output these values have to be presented in certain way  Show the title – inequality GDP  Slope (has * because it means that it sig at .05) second model is full – includes new r2 new slope, new variable, stan error, intercept  When you look at reg – want to look at: size of sig of original, r2 and what happens to new variable – sig?  Additional tools that help contextualize findings and say whe
More Less

Related notes for SOC350H5

Log In


Join OneClass

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

Sign up

Join to view


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.