Skip to contents

Simulate the effect of collapsing strata on precision

Source: R/strata-audit.R
simulate_strata_collapse.Rd

Compares per-stratum RSE and DEFF before and after merging a set of strata into a single combined stratum.

Usage

simulate_strata_collapse(audit, merge_strata)

Arguments

audit

A creel_strata_audit object returned by audit_strata().

merge_strata

Character vector. Names of strata to merge (must all appear in audit$strata$stratum).

Value

A plain tibble (no S3 class) with columns: state ("before" or "after"), stratum, N_h, n_h, RSE, DEFF, meets_target.

Details

Merged strata are pooled using population-weighted means:

  • N_merged = sum(N_h[merge_strata])

  • n_merged = sum(n_h[merge_strata])

  • ybar_merged = sum(N_h * ybar_h) / N_merged (population-weighted)

  • s2_merged = sum(n_h * s2_h) / n_merged (sample-size-weighted pooled within-stratum variance)

RSE and DEFF for the merged stratum are computed using the same FPC-corrected formulas as audit_strata(). Unmerged strata appear identically in both "before" and "after" rows.

References

Cochran, W.G. 1977. Sampling Techniques, 3rd ed. Wiley, New York.

See also

Other "Planning & Sample Size": audit_strata(), compare_designs(), creel_n_camera(), creel_n_cpue(), creel_n_effort(), creel_power(), cv_from_n(), power_creel(), reallocate_strata()

Examples

audit <- audit_strata(
  c(early_season = 30, mid_season = 30, late_season = 20),
  n_h    = c(early_season = 10, mid_season = 12, late_season = 8),
  ybar_h = c(40, 45, 38),
  s2_h   = c(300, 320, 280)
)
simulate_strata_collapse(audit, merge_strata = c("early_season", "mid_season"))
#> # A tibble: 5 × 7
#>   state  stratum                   N_h   n_h    RSE  DEFF meets_target
#>   <chr>  <chr>                   <int> <int>  <dbl> <dbl> <lgl>       
#> 1 before early_season               30    10 0.112   3.17 TRUE        
#> 2 before mid_season                 30    12 0.0889  2.54 TRUE        
#> 3 before late_season                20     8 0.121   3.33 TRUE        
#> 4 after  early_season+mid_season    60    22 0.0704  1.42 TRUE        
#> 5 after  late_season                20     8 0.121   3.33 TRUE