Skip to contents

Summarize survey data with confidence intervals

summarize_data.Rd

Aggregate targets and calculate survey totals and confidence intervals using survey or srvyr package. Use to summarize survey results for reporting and validation.

Usage

summarize_data(
  data,
  id_cols = "hh_id",
  weight_col = "final_weight",
  strata_col = NULL,
  group_col = NULL,
  data_cols = NULL,
  ci_level = 0.9,
  use_survey_package = TRUE
)

Arguments

data

data.table. Survey data to summarize.

id_cols

character vector. Columns to identify unique units (hierarchy).

weight_col

character(1). Column with weights.

strata_col

character(1). Column with strata. Default NULL.

group_col

character vector. Columns to group by. Default NULL.

data_cols

character vector. Columns to calculate totals and CIs for. Default: all targets.

ci_level

numeric. Confidence interval level (fraction, not percent). Default 0.9.

use_survey_package

logical. Use survey package (TRUE) or srvyr (FALSE). Default TRUE.

Value

data.table. Survey totals and confidence intervals by group.

Details

  • Aggregates data by group and computes totals and CIs for specified columns.

  • Supports both survey and srvyr packages for design and estimation.

  • Returns a copy; does not modify by reference.

  • Assumes valid survey design and columns; errors if missing.

Settings

None.

See also

  • scripts/reporting/summarize_survey.R

Other reporting utilities: find_level_idx(), get_target_map(), summarize_pums(), summarize_survey(), tabulate_target()

Examples

## Not run:
summarize_data(data, id_cols = "hh_id", weight_col = "final_weight")
#> Error in `:=`(group_name, 1): Check that is.data.table(DT) == TRUE. Otherwise, :=, `:=`(...) and let(...) are defined for use in j, once only and in particular ways. Note that namespace-qualification like data.table::`:=`(...) is not supported. See help(":=").
## End(Not run)