Create Polling Crosstabs and Export to Excel

This function takes survey data and creates formatted crosstabs showing weighted percentages of row variables across different subgroups. The results are exported to an Excel file with proper formatting.

Usage

create_polling_crosstabs(
  data,
  row_vars,
  subgroup_vars,
  wt_var,
  min_n = 75,
  file_name,
  sheet_name,
  overwrite_existing_sheet = TRUE,
  summary_string = ""
)

Arguments

data

A data frame or survey design object containing survey responses

row_vars

Character vector of variable names to display as rows

subgroup_vars

Character vector of variable names to use as column subgroups

wt_var

Character string specifying the name of the weight variable. Ignored when data is a survey.design or svyrep.design object, in which case weights are taken from the design.

min_n

Integer specifying minimum unweighted sample size for a subgroup to be included (default: 75)

file_name

Character string specifying the Excel file path/name

sheet_name

Character string specifying the sheet name

overwrite_existing_sheet

Logical indicating whether to overwrite existing sheet with same name (default: TRUE)

summary_string

Character string to display in the top-left merged cell (default: "")

Value

Data frame with columns: subgroup_var, subgroup_var_label, subgroup_val, n, row_var, row_var_label, row_val, pct (as decimal), moe (as decimal). All categorical columns are factors with levels in the order passed to the function. Sorted by subgroup variables then row variables.

Details

Weighted percentages are calculated via get_freqs(), which supports plain data frames with a weight column as well as survey.design and svyrep.design objects.

Examples

if (FALSE) { # \dontrun{
results <- create_polling_crosstabs(
  data = survey_data,
  row_vars = c("job_approval", "direction"),
  subgroup_vars = c("gender", "race"),
  wt_var = "weight",
  min_n = 75,
  file_name = "crosstabs.xlsx",
  sheet_name = "Results",
  summary_string = "January 2026 National Poll"
)
} # }