lecy
commented
4 years ago Not a dumb question at all.
The goals for the lab are (1) be able to interpret a regression that includes dummies, and (2) think about which dummy variables you need to omit to test specific hypotheses. So it a fully-interpretative lab. Everything you need is in the regression table already. We worried that including the data might make people think they need to run additional models to answer the questions, which you do NOT.
If you want to play with the data, though, it is here:
URL <- "https://raw.githubusercontent.com/DS4PS/cpp-523-spr-2020/master/labs/data/engineer-salaries.csv"
dat <- read.csv( URL, stringsAsFactors=F )It is synthetic data created for the assignment. If you want to play with some real compensation data similar to that used in the chapter:
# compensation of nonprofit executives
URL <- "https://raw.githubusercontent.com/DS4PS/cpp-523-spr-2020/master/labs/data/np-comp-data.csv"
dat <- read.csv( URL, stringsAsFactors=F )Data Dictionary:
- FILEREIN - org tax ID
- FILERNAME1 - org name
- NTMAJ12 - nonprofit subsector (arts, education, environment, etc.)
- NPAGE - age of the nonprofit
- TAXYR - year in which salaries reported
- STATE
- RULEDATE - year in which org created (used to create NPAGE)
- REVENUE - total revenues of nonprofit
- ASSETS - net assets
- PERSONNM - name of the director
- TITLETXT - title reported on tax forms
- AVGHRS - reported hours worked each week
- SALARY - annual pay of director
- GENDER
- PROPORTION_FEMALE - proportion of board members that are female
- M2012CEO - was the outgoing CEO a male? (1/0)
- TREAT - were they replaced by a male CEO? (1/0)
- POST - is the data from before or after the 2012 transition? (1/0)
Sean-In-The-Library
I know that we don't need the source data for Lab05 to answer the questions but will we get access to it so that we can try to crunch some of this stuff in R if we want?
Sorry if this is a dumb question (or if the point of the lab is specifically to focus on the formulas and not the data).