Package 'SocialFacts'

Title: Average Attributable Fraction and Other Relevant Social Sciences Indicators
Description: Compute Average Attributable Fraction (AAF) and produce a GT table with Odds.ratio, Average Marginal Effects and AAF, with confidence interval and p.value. Also compute other metrics such as Yule's Q and Cramer's V. For more details, see Fergusion and al. (2024) <doi:10.1007/s10654-024-01129-1>.
Authors: Alioscha Massein [aut, cre, cph] (ORCID: <https://orcid.org/0009-0005-5791-1045>), Brice Lefèvre [aut] (ORCID: <https://orcid.org/0000-0002-2482-7104>)
Maintainer: Alioscha Massein <[email protected]>
License: MIT + file LICENSE
Version: 1.3.0
Built: 2026-09-24 09:44:20 UTC
Source: https://github.com/geminy3/socialfacts

Help Index

get_cramer

Description

Get Cramer or YuleQ's tab

Usage

cramer_tab(var = NULL, metric = "cramer", data = NULL, weight = NULL)

Arguments

var

A list of variable name you want to compute "metric" on
metric

Choose between 'Cramer' or 'YuleQ' parameter to select one of this metrics
data

The data table you want to compute 'metric' parameter on
weight

The weight vector if there is one in the dataframe

Value

A list of gt information

Examples

data <- graphPAF::Hordaland_data
cramer_tab(c("y", "urban.rural"), metric = 'cramer', data = data)

get_AAF

Description

Get the AAF

Usage

get_AAF(
  model = NULL,
  nvar = NULL,
  name_var = NULL,
  vars_dep = NULL,
  data = NULL,
  nbootstrap = 50,
  nperm = 100
)

Arguments

model

The model you want to compute AAF on
nvar

You can choose this integer parameter if you prefer to add explanatory variables in the order they appear in the model. You don't have to use name_var parameter.
name_var

A list of explanatory variables you want to compute AAF on
vars_dep

A string with the name of the variable you want to explain
data

The data you used to fit the model (with the variable in name_var and vars_dep)
nbootstrap

An integer with the n bootsrap replication you want. By default, 50.
nperm

Number of permutation between modalities and variable to compute the AAF

Value

A sf table with the result of the AAF computation

Examples

set.seed(9876)
data <- data.frame(
   "var_dep" = as.factor(sample(c(rep_len(c(0, 1), length.out = 1000)), size = 100)),
   "var_1" = as.factor(sample(c(rep_len(c(1, 2, 3), length.out = 1000)), size = 100)),
   "var_2" = as.factor(sample(c(rep_len(c(3, 1, 2), length.out = 100)), size = 100))
   )
glmmodel <- glm(var_dep ~ var_1 + var_2,
                data = data, family = binomial("logit"))

res <- get_AAF(model = glmmodel, nvar = 3, vars_dep = "var_dep", data = data,
               nbootstrap = 5)
res <- get_AAF(model = glmmodel,
               name_var = c("var_1", "var_2"),
              vars_dep = "var_dep", data = data, nbootstrap = 5)

get_list

Description

Get a list with the names of the variables pass as parameter

Usage

get_list(name_var = NULL)

Arguments

name_var

A list of variable you want to pass a name for a list

Value

a list of zero named after the variable in name_var parameter

Examples

vec <- get_list(name_var = c("var1", "var2"))

Get Yule's Q

Description

Get Yule's Q

Usage

get_yuleq(
  vars_dep = NULL,
  name_var = c(),
  data = NULL,
  source = "",
  alpha = 0.01,
  weight = NULL
)

Arguments

vars_dep

A string with name of the dependant variable to use as one variable for the 2x2 table
name_var

A vector of explanatory variable you want to compute Yule's Q on
data

A dataframe with the variable named before
source

A string with the name of the data source
alpha

A integer with the alpha value for fisher.test and IC
weight

A vector with weigths corresponding to the dataframe individuals weights

Value

A gt table with the Yule's computed for the variable in name_var

Examples

data <- graphPAF::Hordaland_data
get_yuleq(vars_dep = "y", name_var = c("y",  "urban.rural"),
                    data = data, source = "dataSource - Year",
                    alpha = 0.01)

result_tab

Description

Get the result table of OR, AME and AAF

Usage

result_tab(
  model = NULL,
  var_ref = NULL,
  var_names = NULL,
  res_AAF = NULL,
  source = "",
  data = data.frame()
)

Arguments

model

The model used to perform statistical analysis
var_ref

The variable to explain in the statistical model
var_names

A list with the name of the variable you want to show on the contingency tables
res_AAF

the result of the AAF computation with get_AAF() function (done outside of this function because of the length of the computation)
source

Data source information to add to the plots
data

The data used for the statistical analysis

Value

A gt table

Examples

set.seed(9876)
data <- data.frame(
   "var_dep" = as.factor(sample(c(rep_len(c(0, 1), length.out = 1000)), size = 100)),
   "var_1" = as.factor(sample(c(rep_len(c(1, 2, 3), length.out = 1000)), size = 100)),
   "var_2" = as.factor(sample(c(rep_len(c(3, 1, 2), length.out = 100)), size = 100))
   )
glmmodel <- glm(var_dep ~ var_1 + var_2, data = data, family = binomial("logit"))


res <- get_AAF(model = glmmodel, nvar = 3, vars_dep = "var_dep", data = data,
               nbootstrap = 2)
res_tab <- result_tab(model = glmmodel, var_ref = "var_dep",
                      var_names = c("var_1", "var_2"),
                      res_AAF = res$res, source = "TESTDATA", data = data)