Skip to contents

Multi-indicator precision for cluster designs

Source: R/prec_multi.R
prec_multi_cluster.Rd

Compute achieved precision for several indicators under a two- or three-stage cluster allocation. This is the inverse of n_multi_cluster().

Usage

prec_multi_cluster(indicators, ...)

# Default S3 method
prec_multi_cluster(
  indicators,
  ...,
  domains = NULL,
  stage_cost = NULL,
  resp_rate_psu = 1,
  resp_rate_ssu = 1,
  resp_rate = 1,
  plan = NULL
)

# S3 method for class 'svyplan_cluster'
prec_multi_cluster(indicators, ...)

Arguments

indicators

For the default method, a non-empty data frame with one row per indicator. It must contain n and n_per_psu. Include n_per_ssu for a three-stage design. Cluster homogeneity and indicator columns follow the schema used by n_multi_cluster(), including the derived three-stage var_ratio_ssu. For svyplan_cluster methods, an allocation returned by n_multi_cluster().

...

Additional arguments passed to methods. Unused arguments are rejected.

domains

Optional character vector naming domain columns in indicators.

stage_cost

Optional per-stage costs to retain for a later round trip to n_multi_cluster(). Costs do not enter the precision calculation.

resp_rate_psu

Default expected PSU response rate, in (0, 1]. Used where the indicator column is absent or NA.

resp_rate_ssu

Default expected SSU response rate for a three-stage design, in (0, 1]. It is not applicable to a two-stage design.

resp_rate

Default expected ultimate-unit response rate, in (0, 1]. Non-missing indicator columns override these three defaults row by row.

plan

Optional svyplan() profile providing design metadata.

Value

A svyplan_prec object with per-indicator cluster precision in $detail: .se, .moe, .rmoe, and .cv, with .rmoe measured against the row's p or abs(mu).

See also

n_multi_cluster() for the inverse, prec_multi() for the single-stage counterpart, and prec_cluster() for one indicator.

Other precision functions: prec_alloc(), prec_change(), prec_cluster(), prec_mean(), prec_multi(), prec_panel(), prec_pooled(), prec_prop(), prec_twophase()

Examples

indicators <- data.frame(
  name = c("stunting", "anemia"),
  p = c(0.30, 0.10),
  n = c(60, 60),
  n_per_psu = c(12, 12),
  icc_psu = c(0.02, 0.05)
)
prec_multi_cluster(indicators)
#> Multi-indicator sampling precision
#> 
#>  name     .se .moe .rmoe .cv       
#>  stunting NA  NA   NA    0.06287848
#>  anemia   NA  NA   NA    0.13919411