Class Notes (811,170)
Sociology (4,001)
SOC350H5 (9)
Lecture 7

# SOC350H5 Lecture 7: Lecture 7 Premium

3 Pages
63 Views

School
University of Toronto Mississauga
Department
Sociology
Course
SOC350H5
Professor
David Pettinicchio
Semester
Winter

Description
Lecture 7  Final project – asks about intermediating relationship  Dummy variable – categorical variable coded as 0/1  How to code those with multiple variables Recall level of measurement  Gender 0 0/1  Categorical data that is not dichotomous  How to interpret them substantively  OLS have dep variable that is interval  When it comes to categorical data as indep – can do a lot  Categorical variable that have category of 0 as reference – dummy  Interpret attributes of variable  0- reference (comparing to 0) Categorical variable  Race/relig – not quantitative but can include them in quant analysis  How do you interpret effects of those with multiple categories – not just comparing male to female  Understand that you are recoding variable into 0/1 category and omitting referent category Dummy variable  Attributes of 0 and 1  Interpret the category that is 1 in reference to what is called 0  There is no rule that this is 0 and this is 1 (do you wanna say males compared to females or females compared to male) – different  Do you want to make white referent category that you compare other races to  Comparison – want to be intuitive – how do you want to interpret matters  Recode variable so you make referent variable 0  By default the software recognizes 0 as referent  Its important to be consistent but when you deal with multiple categories – its important to be consistent  If you return to data, you wont remember otherwise Basic – eg Gender  Male 0/female 1  Comparing female to male  When you interpret the findings you have to interpret it based on referent category  Output from SPSS  Gender/fathers and mothers occupational prestige score  The way you interpret it:  Compared to being male, being female increases respondent’s occupational prestige score score by .606 Multiple categories  A lot of variables are not just 0 and 1  Variable has multiple attributes  Look at race  Being black or other impacts occupational prestige in comparison to white  Pick and recode variable that you want referent – make 0  Recode the rest as different variable  SPSS knows when you make everything 0 – it knows that referent category was white  Everything has to be in the model at same time though  When you recode race – all categories have to be in model at the same time  No longer have “race” in model  Referent is white – in comparison to white, being black decreases occupational prestige score by 4.168 The nature of relationship  The idea of what relationship look like between variables is a theoretical question  SPSS doesn’t tell you  You have to know that  Do I ha
More Less

Related notes for SOC350H5

OR

Don't have an account?

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.