Assignment 11

.pdf

School

California State University, Chico *

*We aren’t endorsed by this school

Course

105

Subject

Economics

Date

Jan 9, 2024

Type

pdf

Pages

Uploaded by ProfessorUniverseTarsier7

Example Assignment Eleven: Using multiple regression to analyze the gender pay gap Part I: Use APA style and formatting for all assignments, references, and citations. Yes, have a cover page, too, as well as a running head. Try Purdue Owl for an example APA style paper: https://owl.english.purdue.edu/owl/resource/560/18/ For this final analysis you get to bring together many of the variables we have been using this term to better understand difference in income. In particular we want to explain the gender pay gap between women and men. For Ordinary Least Squares (OLS) regression analyses, which we are using for this assignment, you want to have at least one interval/ratio independent variable and an interval/ratio dependent variable. Your dependent variable is pincp. Your independent variables will be sex, agep, and schl. But, there are other variables that might explain variation in income. For this analysis we will add race to our independent variable list. However, as your book tells you, for the nominal variables we need to do a little recoding into “dummy” variables so we can use OLS regression more effectively. We will recode sex into a dummy variable called “Male.” And, we will recode rac1p into a dummy variable called “White.” Also, know that there are more tests that need to be done to come to firmer conclusions from an OLS analysis. For example, two independent variables might also have a strong association where one predicts the other to a large degree. Might this be the case for sex and schl? When this happens it is known as multicollinearity or just collinearity and it can impact OLS regression results. There are ways to test for it and correct the problem, but we are not going to do that in this course. Just know that there is more to OLS regression than what you practice here. You are practicing running and interpreting the analysis. 1. What is the measure (nominal, ordinal, or interval/ratio) of each of your independent variables and your dependent variable? Dependent variable, pincp: I/R Independent variable, rac1p: Nominal Independent variable, sex: Nominal Independent variable, agep: I/R Independent variable, schl: I/R I answer for you because I want you to treat this as an I/R variable for years of schooling even though it is not exactly year for year the years of schooling. You can check the data dictionary for schl to see how the answers are coded. They are coded from 1 to 16 where each number means progressively more education. 2. Using your 2014-2018 ACS data file, recode your nominal independent variables as instructed in the text under 17.2 Recoding to Create Dummy Variables and from past assignments to transform each nominal variable, sex and rac1p, into a new variable. 3. For sex code Male=1 and Female = 0 in a new variable Male. Male is already coded 1, but you need to make 1 = 1 in the new variable anyway. Female is coded as 2, so you have to change the 2 to a 0. The new variable, male, should be numeric when you are done. Here is a screen shot to help you:

4. Next assign labels to the values for your new variable, male. So, 1=Male and 0=Female. We have done assigned labels before. See screen shot below to help guide you. 5. Save your file with the new variable, male. 6. For rac1p code white=1 and nonwhite = 0 in a new variable white. Recoding rac1p is a little more complicated to recode thnt sex was because it has many values-white, black, native, etc. If

you want to see the coding for rac1p, it is in the data dictionary starting on page 101. White is already coded 1, but you need to make 1=1 in the new variable. All other race categories are coded 2-9, so you have to change them all together to =0. Race categories are much more complicated than white/nonwhite. We are coding them this way for ease of practice. The new variable, white, should be numeric when you are done. Here is a screen shot to help you: 7. Next assign labels to the values for your new variable, white. So, 1=White and 0=Nonwhite. We have done assigned labels before. See screen shot below to help guide you.

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version