c08 aggregate Ruhela: non-majority religion x gender -> RC

Usage

morie_otis_analyze_c08_ruhela_aggregate(data, out_dir = NULL)

morie_otis_analyze_c08_mrm_aggregate(data, out_dir = NULL)

Arguments

data: c08 data.frame.
out_dir: Optional.

Value

morie_otis_analysis_result.

Examples

# \donttest{
otis_c08 <- morie_synth_otis("c08", n = 120L, seed = 1L)
morie_otis_analyze_c08_ruhela_aggregate(otis_c08)
#> OTIS c08 -- Aggregate RF: non-majority religion x Gender -> Individuals in RC
#> =============================================================================
#>   Source file                            c08_individuals_in_segregation_and_restrictive_confinement_religion_by_gender.csv
#>   Dataset id                             c08
#>   Treatment column                       T_minority_religion
#>   Outcome (count)                        NumberIndividuals_RestrictiveConfinement
#>   Year FE                                EndFiscalYear
#>   Covariate FEs                          Gender
#>   Cluster group (GEE)                    --
#>   Cells                                  120
#>   Total outcome count                    4440
#>   Overdispersion (Poisson AIC - NB AIC)  1559
#> 
#> Aggregate Ruhela formulation -- Poisson + NB IRR for treatment effect on count outcome:
#>   Family   Coefficient                   IRR     95% CI          p-value   AIC     Notes
#>   ----------------------------------------------------------------------------------------------------
#>   Poisson  factor(T_minority_religion)1  0.9317  [0.870, 0.998]  4.34e-02  2671.8  n=120, total_y=4440
#>   NB       factor(T_minority_religion)1  0.9002  [0.636, 1.273]  5.52e-01  1112.8  n=120, total_y=4440
#> 
#> Aggregate RF on c08, parallel to c05 with gender control.
#> 
#> 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. 
# }