poll.data: Synthetic Polling Data

In senseweight: Sensitivity Analysis for Weighted Estimators

Synthetic Polling Data

Description

• pid: party indicator, one of c('Democrat', 'Republican', 'Independent', 'Other')

• educ: education level, one of c('High School or Less', 'Some college', 'College', 'Post-grad')

• age: age of the individual

• age_buckets: grouped age buckets, one of c('18to35', '36to50', '51to64', 'Over65')

• bornagain: indicator for whether the individual identifies as born-again Christian, one of c('Yes', 'No')

• gender: categorical gender, one of c('Men', 'Women')

• race: categorical race, one of c('White', 'Black', 'Hispanic', 'Asian', 'Other')

• vvweight_post: post-election verified voter weight for the individual

Usage

data('poll.data')

Details

A synthetically generated dataset that simulates polling data for a hypothetical support. The covariates are constructed to match commonly used demographic covariates in practice.

References

"Sensitivity analysis for survey weights," Political Analysis, 32(1) (2024), 1-16. #'

Examples

data('poll.data')

senseweight documentation built on Aug. 23, 2025, 1:11 a.m.