Decompose Aggregate Change into Within and Between Components

Decomposes aggregate change between two periods into within-occupation, between-occupation, interaction, and unclassifiable components. The within term is restricted to rows marked comparable.

Usage

onet_decompose_change(
  from_scores,
  to_scores,
  from_weights,
  to_weights,
  occupation_code = "reference_soc_code",
  score = "measure_score",
  weight = "employment",
  comparable = "safely_comparable"
)

Arguments

from_scores

Initial-period occupation scores.

to_scores

Final-period occupation scores.

from_weights

Initial-period weights.

to_weights

Final-period weights.

occupation_code

Join column for scores and weights.

score

Score column in score tables.

weight

Weight column in weight tables.

comparable

Optional logical column in to_scores or from_scores marking rows safe for within-change attribution.

Value

A tibble with decomposition components and coverage list-column metadata readable with onet_coverage().

Examples

from_scores <- tibble::tibble(
  reference_soc_code = c("15-1252", "29-1141"),
  measure_score = c(0.6, 0.3),
  safely_comparable = c(TRUE, TRUE)
)
to_scores <- tibble::tibble(
  reference_soc_code = c("15-1252", "29-1141"),
  measure_score = c(0.7, 0.2),
  safely_comparable = c(TRUE, FALSE)
)
from_weights <- tibble::tibble(
  reference_soc_code = c("15-1252", "29-1141"),
  employment = c(100, 300)
)
to_weights <- tibble::tibble(
  reference_soc_code = c("15-1252", "29-1141"),
  employment = c(150, 250)
)
onet_decompose_change(from_scores, to_scores, from_weights, to_weights)
#> # A tibble: 5 × 3
#>   component        value coverage        
#>   <chr>            <dbl> <list>          
#> 1 within          0.025  <tibble [1 × 3]>
#> 2 between         0.0375 <tibble [1 × 3]>
#> 3 interaction     0.025  <tibble [1 × 3]>
#> 4 unclassifiable -0.075  <tibble [1 × 3]>
#> 5 total_change    0.0125 <tibble [1 × 3]>