b03 aggregate Ruhela: Alert presence -> seg placements

Usage

morie_otis_analyze_b03_ruhela_aggregate(data, out_dir = NULL)

morie_otis_analyze_b03_mrm_aggregate(data, out_dir = NULL)

Arguments

data: b03 data.frame.
out_dir: Optional.

Value

morie_otis_analysis_result.

Examples

# \donttest{
otis_b03 <- morie_synth_otis("b03", n = 120L, seed = 1L)
morie_otis_analyze_b03_ruhela_aggregate(otis_b03)
#> OTIS b03 -- Aggregate Ruhela formulation: Alert presence -> Number of segregation placements
#> ============================================================================================
#>   Source file                            b03_segregation_placements_alerts_and_hold_flags_by_institution.csv
#>   Dataset id                             b03
#>   Treatment column                       T_alert
#>   Outcome (count)                        Number_SegregationPlacements
#>   Year FE                                EndFiscalYear
#>   Covariate FEs                          Alert_Type, Region_AtTimeOfPlacement
#>   Cluster group (GEE)                    Institution_AtTimeOfPlacement
#>   Cells                                  120
#>   Total outcome count                    4895
#>   Overdispersion (Poisson AIC - NB AIC)  1216.6
#> 
#> Aggregate Ruhela formulation -- Poisson + NB IRR + GEE clustered for treatment effect on count outcome:
#>   Family                                                     Coefficient       IRR     95% CI          p-value   AIC     Notes
#>   -------------------------------------------------------------------------------------------------------------------------------------------
#>   Poisson                                                    factor(T_alert)1  0.9585  [0.898, 1.023]  2.00e-01  2353.9  n=120, total_y=4895
#>   NB                                                         factor(T_alert)1  0.9315  [0.697, 1.246]  6.32e-01  1137.4  n=120, total_y=4895
#>   GEE-Poisson (cluster:Institution_AtTimeOfPlacement, Exch)  factor(T_alert)1  0.9486  [0.723, 1.244]  7.03e-01  --      groups=25, n_obs=120
#> 
#> Aggregate Ruhela formulation on b03 testing whether alert-flagged person-cells receive more segregation placements than no-alert cells.
#> 
#> IRR > 1 ==> treatment increases the count rate; IRR < 1 ==> treatment decreases it. Concordance between Poisson and NB indicates equidispersion; large gap (Poisson AIC > NB AIC) indicates overdispersion -- trust NB. 
# }