Skip to contents

Calculate and report geo-naming match statistics

Source: R/harmonize_admin_names.R
calculate_match_stats.Rd

Compares entries in a dataset against a lookup across specified admin levels (e.g., country, province/state/region, district, subdistrict, settlement) and reports match statistics to the console.

Usage

calculate_match_stats(
  data,
  lookup_data,
  level0 = NULL,
  level1 = NULL,
  level2 = NULL,
  level3 = NULL,
  level4 = NULL
)

Arguments

data

A data frame containing the target data to be matched.

lookup_data

A data frame serving as the reference for matching.

level0

Column name (country) present in both data and lookup_data.

level1

Column name (province/state/region) present in both datasets.

level2

Column name (district) present in both datasets.

level3

Column name (subdistrict) present in both datasets.

level4

Column name (settlement) present in both datasets.

Value

Invisibly returns NULL. Output is produced via cli messages.

Details

  • Input columns supplied via level* are normalized to lower case before matching.

  • Base N for each side is the count of unique hierarchical names formed from the supplied levels (e.g., level0_level1_...). Per-level rows show matches out of each side's Base N.

  • Console output includes:

    1. a two-column summary (Target vs Lookup as base N),

    2. side-level completeness messages (success/info/warning),

    3. a per-level report of missing names (NA or empty strings) on either side. These missing names are not included in N only if the implementation drops them before counting (see Note below).

Note

If you want missing names (NA/empty) excluded from Base N, ensure the key-building step drops them before counting (see example patch below).

Examples

# minimal runnable example (toy data)
data <- data.frame(
  country = c("Kenya", "Kenya", "Uganda"),
  district = c("Nairobi", "Kisumu", "Kampala")
)
lookup <- data.frame(
  country = c("Kenya", "Uganda"),
  district = c("Nairobi", "Kampala")
)
calculate_match_stats(
  data, lookup, level0 = "country", level2 = "district"
)
#> 
#> ── ℹ Match Summary ─────────────────────────────────────────────────────────────
#> 
#> ℹ Data has names not in lookup.
#> 
#> Target data as base N                                                       
#> • country (level0): 2 out of 2 matched                                      
#> • district (level2): 2 out of 3 matched                                     
#> Lookup data as base N                                                       
#> • country (level0): 2 out of 2 matched                                      
#> • district (level2): 2 out of 2 matched