descriptive_table

Get a Descriptive Table for every Data Frame, to easy document your Data

Easily analyze and visualize differences between samples (e.g., benchmark comparisons, nonresponse comparisons in surveys) on three levels. The comparisons can be univariate, bivariate or multivariate. On univariate level the variables of interest of a survey and a comparison survey (i.e. benchmark) are compared, by calculating one of several difference measures (e.g., relative difference in mean), and an average difference between the surveys. On bivariate level a function can calculate significant differences in correlations for the surveys. And on multivariate levels a function can calculate significant differences in model coefficients between the surveys of comparison. All of those differences can be easily plotted and outputted as a table. For more detailed information on the methods and example use see Rohr, B., Silber, H., & Felderer, B. (2024). Comparing the Accuracy of Univariate, Bivariate, and Multivariate Estimates across Probability and Nonprobability Surveys with Population Benchmarks. Sociological Methodology <doi:10.1177/00811750241280963>.

Bjoern Rohr

sampcompR

Comparing and Visualizing Differences Between Surveys

Barbara Felderer

descriptive_table function

<dl><dt>dfs</dt>
<dd>A character vector, containing the names of the data frames.</dd>
<dt>variables</dt>
<dd>A character vector containing the variables in the data frame
that should be described.</dd>
<dt>varlabels</dt>
<dd>A character vector containing the Labels for every variable in variables.</dd>
<dt>weight</dt>
<dd>A character vector, containing either the name of a weight in the
respective data frame, or NA, if no weighting should be performed for this data frame.</dd>
<dt>strata</dt>
<dd>A character vector, containing either the name of a strata in the
respective data frame, or NA, if no strata should be used when weighting this
data frame.</dd>
<dt>id</dt>
<dd>A character vector, containing either the name of a id in the
respective data frame, or NA, if every row is unique for this data frame.</dd>
<dt>value</dt>
<dd>A character vector indicating what descriptive value should be displayed
for the data frame. It can either be "mean", "percent", "total", or "total_percent".</dd>
<dt>digits</dt>
<dd>A numeric value indicating the number of digits that the Descriptive table
should be rounded to.</dd></dl>

Arguments

Get a Descriptive Table for Every Data Frame — descriptive_table

<dl>

<dt>dfs</dt>
<dd>A character vector, containing the names of the data frames.</dd>


<dt>variables</dt>
<dd>A character vector containing the variables in the data frame
that should be described.</dd>


<dt>varlabels</dt>
<dd>A character vector containing the Labels for every variable in variables.</dd>


<dt>weight</dt>
<dd>A character vector, containing either the name of a weight in the
respective data frame, or NA, if no weighting should be performed for this data frame.</dd>


<dt>strata</dt>
<dd>A character vector, containing either the name of a strata in the
respective data frame, or NA, if no strata should be used when weighting this
data frame.</dd>


<dt>id</dt>
<dd>A character vector, containing either the name of a id in the
respective data frame, or NA, if every row is unique for this data frame.</dd>


<dt>value</dt>
<dd>A character vector indicating what descriptive value should be displayed
for the data frame. It can either be "mean", "percent", "total", or "total_percent".</dd>


<dt>digits</dt>
<dd>A numeric value indicating the number of digits that the Descriptive table
should be rounded to.</dd>

</dl>

descriptive_table: Get a Descriptive Table for Every Data Frame

Description

Usage

Value

Arguments