cces_30_df

A dataset used for modeling support for abortion coverage in insurance plans in the United States, 
combining individual- and district-level covariates from a 2018 survey.
Out of all U.S. states, only data from 30 states are included in this dataset.
A data frame with 49,095 rows and 6 variables:<dl>
 <dt>abortion</dt>
<dd>Allow employers to decline coverage of abortions in insurance plans (0: Oppose, 1: Support).</dd> <dt>state</dt>
<dd>A factor variable representing the 30 selected U.S. states. The raw data includes states with state_fips values ranging from 1 to 30.</dd> <dt>eth</dt>
<dd>An ethnicity variable stored as a factor (White, Black, Hispanic, Other).</dd> <dt>gender</dt>
<dd>A factor variable representing gender (Female, Male).</dd> <dt>age</dt>
<dd>A factor variable dividing respondents into age groups (18–29, 30–39, 40–49, 50–59, 60–69, 70+).</dd> <dt>educ</dt>
<dd>A factor variable representing educational attainment (No HS, HS, Some college, 4-Year College, Post-grad).</dd>
</dl>

datasets

Complex machine learning models are often difficult to interpret. Shapley values serve as a powerful tool to understand and explain why a model makes a particular prediction. This package computes variable contributions using permutation-based Shapley values for Bayesian Additive Regression Trees (BART) and its extension with Post-Stratification (BARP). The permutation-based SHAP method proposed by Strumbel and Kononenko (2014) <doi:10.1007/s10115-013-0679-x> is grounded in data obtained via MCMC sampling. Similar to the BART model introduced by Chipman, George, and McCulloch (2010) <doi:10.1214/09-AOAS285>, this package leverages Bayesian posterior samples generated during model estimation, allowing variable contributions to be computed without requiring additional sampling. The BART model is designed to work with the following R packages: 'BART' <doi:10.18637/jss.v097.i01>, 'bartMachine' <doi:10.18637/jss.v070.i04>, and 'dbarts' <https://CRAN.R-project.org/package=dbarts>. For XGBoost and baseline adjustments, the approach by Lundberg et al. (2020) <doi:10.1038/s42256-019-0138-9> is also considered. The BARP model proposed by Bisbee (2019) <doi:10.1017/S0003055419000480> was implemented with reference to <https://github.com/jbisbee1/BARP> and is designed to work with modified functions based on that implementation. BARP extends post-stratification by computing variable contributions within each stratum defined by stratifying variables. The resulting Shapley values are visualized through both global and local explanation methods.

Dong-eun Lee

bartXViz

Visualization of BART and BARP using SHAP

Eun-Kyung Lee

cces_30_df function

A dataset used for modeling support for abortion coverage in insurance plans in the United States, 
combining individual- and district-level covariates from a 2018 survey.
Out of all U.S. states, only data from 30 states are included in this dataset.
A data frame with 49,095 rows and 6 variables:<dl>
 <dt>abortion</dt>
<dd>Allow employers to decline coverage of abortions in insurance plans (0: Oppose, 1: Support).</dd>

 <dt>state</dt>
<dd>A factor variable representing the 30 selected U.S. states. The raw data includes states with state_fips values ranging from 1 to 30.</dd>

 <dt>eth</dt>
<dd>An ethnicity variable stored as a factor (White, Black, Hispanic, Other).</dd>

 <dt>gender</dt>
<dd>A factor variable representing gender (Female, Male).</dd>

 <dt>age</dt>
<dd>A factor variable dividing respondents into age groups (18–29, 30–39, 40–49, 50–59, 60–69, 70+).</dd>

 <dt>educ</dt>
<dd>A factor variable representing educational attainment (No HS, HS, Some college, 4-Year College, Post-grad).</dd>


</dl>

cces_30_df: Survey Data on Public Opinion about Abortion Coverage in Insurance Plans(2018)

Description

Arguments

References