Summarise simulated DiD data as change-score statistics

Computes within-individual change scores (post minus pre) matched on subject_id, then returns means and SDs for each arm. Produces data in the format expected by meta_did() for design "did_change".

Usage

as_summary_did_change(sim)

Arguments

sim

A data frame produced by simulate_meta_did().

Value

A data frame with one row per study in the format required by validate_summary_data() for design "did_change".

Examples

sim <- simulate_meta_did(n_studies = 3, seed = 1)
as_summary_did_change(sim)
#> # A tibble: 3 × 8
#>   study_id design    n_control n_treatment mean_change_control sd_change_control
#>   <chr>    <chr>         <int>       <int>               <dbl>             <dbl>
#> 1 study_1  did_chan…       100         100             -0.0250             0.117
#> 2 study_2  did_chan…       100         100             -0.0168             0.128
#> 3 study_3  did_chan…       100         100             -0.0181             0.118
#> # ℹ 2 more variables: mean_change_treatment <dbl>, sd_change_treatment <dbl>