Estimate a survey proportion using ENEMDU complex survey design

Estimates a weighted proportion for a binary 0/1 variable using ENEMDU design variables. This function is appropriate for variables such as poverty flags once they have been built from validated inputs.

Usage

enemdu_survey_proportion(
  data,
  value,
  group_vars = NULL,
  ids = "upm",
  strata = "estrato",
  weight = "fexp",
  survey_type = NULL,
  indicator_id = NULL,
  measure = NULL,
  na_rm = TRUE,
  conf_level = 0.95,
  lonely_psu = "adjust",
  sample_n_min = 60,
  strict_binary = TRUE
)

Arguments

data

A data frame.

value

Binary 0/1 variable.

group_vars

Optional grouping variables.

ids

Primary sampling unit variable. Defaults to "upm".

strata

Strata variable. Defaults to "estrato".

weight

Expansion factor variable. Defaults to "fexp".

survey_type

Optional ENEMDU survey type.

indicator_id

Optional stable indicator identifier.

measure

Optional human-readable measure name.

na_rm

Logical. If TRUE, missing values in value are excluded.

conf_level

Confidence level for confidence intervals.

lonely_psu

Option passed to survey.lonely.psu.

sample_n_min

Minimum unweighted sample size for preliminary quality flag.

strict_binary

Logical. If TRUE, errors when non-missing values other than 0 or 1 are found.

Value

A tibble with survey proportion estimates and precision metadata.