Class Notes (810,290)
Canada (494,047)
Sociology (3,990)
SOC350H5 (9)
Lecture 9

SOC350H5 Lecture 9: Lecture 9

4 Pages
Unlock Document

University of Toronto Mississauga
David Pettinicchio

Lecture 9 Omitted variable bias  Left out variables that are negatively affecting outcomes  What are known predictors influencing outcome  Something destabilizing model (lack of it)  It’s a problem with x highly related to each other  Because cannot decide what variable has the effect cuz you can t separate the two Model specification  Parameters that you decide are boundaries of outcome  Part of understanding outcome is first am I putting in irrelevant things  Not affecting outcome but also screwing up outcome  Jeopardize ability to predict outcome  When understanding multivariate model – saying stat sig are interpreted in relation to all x  Standard error  When you have omitted variable bias  Can make things be non sig when they may be sig but fact that model is on shaky grounds because of variables that you select  This happens when you add irrelevant x or miss important ones Omitted Variable bias  Solution – add variable that you think are important  You cannot substitute something else in there if data doesn’t contain it  Find another way to deal with problem  Adjusted r2 very low – problem  People rely on comparing r2 to adj r2 to see is variable bias  You saying x is increasing y in a linear way but non linear relationship could be problem  A lot of concern about what you do when you don’t know how to construct model based on unavailability of data  Look at r2 with caution SPSS OVB  Requires extra step  Create a variable –  Create a model  Ask spss to sq outcome of predicted values  It’ll create a new variable  This is a method that is more reliable than just looking at adjusted r sq  But first have to produce original model first  If you suspect OVB have to sq predicted values as new variable  Re run original dep regression – with predicted value and sq value  Two outcomes  DON’T WANT SQUARED PREDICTED TERM TO BE SIGNIFICANT  When values are sq, relationships are not linear  When you run reg with predicted values and they are sig – there is a problem with structure of model – variables causing model to violate assumption  This is an indication that you must readjust or fix kind of variables you have in model  If its not doing anything, why is it in there  Basically to look at omitted variable bias – squared predicted values shouldn’t be sig  Have sq predict value and unstand predicted value – 0.05 (not a problem)  Does not reach significance at alpha of .05 What if spv is sig  Adding variables  Enter variables that were omitted  What is an important predictor that might be missing  Is it causing multicolinearity  Outliers?  In project – have to be honest about what you did – what you thought was the problem Multicollinearity  Independent variables not detached from each other in social world  Problem is not whether x are correlated but whether its just too high  Pearson correlation matrix  Are your x variables correlated to each other  MC problem with x so closely related that you cant separate what is closely related  If you have something super correlated that SPSS takes it out for you  Diagnostic to look at whether you should worry about ti  Cant figure out what effects are when cant untangle – PROBLEM  Interaction term is variable term  Lecture 4 looked at pearson correlation – see how
More Less

Related notes for SOC350H5

Log In


Don't have an account?

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.