pop_cces: CCES Sample that can be used as a quasi-sample
In kuriwaki/svysim: Realistic Simulations of Biased Survey Sampling

The cumulative CCES stacks CCES common content for all years and harmonizes the variables, which makes it ideal for using it for MRP. This is a sample for illustration; see ccesMRPprep::get_cces_dv to get the full data.

pop_cces

A dataset of 402220 observations and 17 columns. See the CCES cumulative codebook for more explanation of the variables.

ID: Identifier by DeclareDesign
year: Year of the common content
case_id: Respondent identifier (unique within each year)
state: State (in the form of state.name)
cd: Congressional district at the time of the survey.
gender: Gender (equivalent to sex in ACS for the purposes of this package)
age: Age, in integers
race: Race and ethnicity
educ: Education level
newsint: Frequency of following the news
pid3_leaner: Partisan Identification
ideo5: Ideology Self-Placement
Y: A simulated continuous outcome that is highly positively correlated with liberal and Democratic identification. See the source file data-raw for exact simulation
Z: A simulated binary outcome, sampled from passing Y with a inverse logit link and using its probailities.

This is encoded as a RDS file with some variables stored in the Stata-based integer + labelled class instead of as factors. See the CCES cumulative guide for the difference between the two and how to go from one to another.

Kuriwaki, Shiro, 2018, "Cumulative CCES Common Content (2006-2018)", https://doi.org/10.7910/DVN/II2DB6, Harvard Dataverse

kuriwaki/svysim documentation built on Jan. 27, 2021, 5:37 a.m.