Package 'vizsurvey'

Title: Visualisation And Analysis During a Survey Field
Description: vizsurvey is an R package designed to streamline the quality assessment of survey data by providing intuitive visual diagnostics through an interactive dashboard. vizsurvey is especially useful for institutions or researchers conducting large-scale surveys with multiple interviewers, enabling a fast and systematic overview of data quality over time.
Authors: Thomas Delclite [aut, cre], Adrien Mierop [aut]
Maintainer: Thomas Delclite <[email protected]>
License: GPL (>= 3)
Version: 0.4.5
Built: 2026-10-28 13:54:58 UTC
Source: https://github.com/tdelc/vizsurvey

Help Index

Build the df_timer of interviewer

Description

Build the df_timer of interviewer

Usage

build_df_timer_intvwr(df_timer_intv, cfg)

Arguments

df_timer_intv

df_timer_intv
cfg

config list

Value

data.frame

Classify all variable of a data.frame

Description

Classify all variable of a data.frame

Usage

classify_df(df, threhold = 15)

Arguments

df

A data frame
threhold

Maximum number of modalities to classify variable as modal

Value

a data frame

Examples

classify_df(iris)

Combine several variables in one key (multi-level wave or filter)

Description

Used when var_wave (or var_filter) contains more than one variable in the configuration file, for example the year and the quarter. The complete key of a row is then "2024 / T1".

Usage

combine_vars(df, vars, sep = " / ")

Arguments

df

data.frame
vars

vector of the variable of each level
sep

separator between the levels

Value

vector of the complete keys

Examples

head(combine_vars(mtcars,c("cyl","gear")))

corrections of each df of a list

Description

corrections of each df of a list

Usage

correct_list_df(list_df)

Arguments

list_df

list of df

Value

list

Create a template of configuration file

Description

Create a template of configuration file

Usage

create_config(
  folder_path,
  file_name = "config.txt",
  name_survey = NULL,
  vars_discretes = NULL,
  vars_continuous = NULL,
  vars_ignore = NULL,
  prefix_discretes = NULL,
  prefix_continuous = NULL,
  prefix_ignore = NULL,
  var_wave = NULL,
  var_filter = NULL,
  var_intvwr = NULL,
  var_intv = NULL,
  var_date = NULL,
  var_info_geo = NULL,
  var_timer = NULL,
  var_itm_duration = NULL,
  var_session = NULL,
  duration_min_during = NULL,
  duration_min_inter = NULL,
  night_start = NULL,
  night_end = NULL
)

Arguments

folder_path

folder where create the file
file_name

Name of the config file (config.txt by default)
name_survey

Name of the survey (not used)
vars_discretes

(optional) preset discretes variables name (VAR1,VAR2,...)
vars_continuous

(optional) preset continuous variables name (VAR1,VAR2,...)
prefix_discretes

(optional) preset prefix for discretes variables name
prefix_continuous

(optional) preset prefix for continuous variables name
var_wave

(optional) variable name of wave. Two variables can be given (c("YEAR","QUARTER")) : the second one is then a second level of wave
var_filter

(optional) variable name of filter. Two variables can be given : the second one is then a second level of filter
var_intvwr

(optional) variable name of interviewer id
var_intv

(optional) variable name of interview id
var_date

(optional) variable name of the date of the interview
var_timer

(optional, audit trail) variable name of the beginning of a item
var_itm_duration

(optional, audit trail) variable name of the duration of a item
var_session

(optional, audit trail) variable name of the session of a item
duration_min_during

(optional, audit trail) threshold of a interview duration
duration_min_inter

(optional, audit trail) threshold of the duration between two interviews
night_start

(optional, audit trail) hour of beginning of the night
night_end

(optional, audit trail) hour of end of the night

Examples

create_config(".") # creation of config.txt in working directory

Create statistics from database

Description

Create statistics from database

Usage

create_df_stats(
  df_,
  configs_,
  var_calculs,
  mod_filter = NULL,
  na.rm = FALSE,
  force_cat = FALSE
)

Arguments

df_

database
var_calculs

variable to create stats
mod_filter

(optional) modality to filter data
na.rm

include or not missing values as modality
configs

configs

Value

df data.frame

Simulate EU-SILC dataset with injected errors

Description

Simulate EU-SILC dataset with injected errors

Usage

create_eusilc_sim()

Value

data.frame from laeken::eusilc with ids and errors

Examples

# create_eusilc_sim()

Use PUF files for SILC example

Description

Use PUF files for SILC example

Usage

create_fake_silc(
  vec_country = c("BE", "RO"),
  path_out = "inst/shiny/complete/data/SILC/"
)

Arguments

vec_country

Vector of country to import
path_out

Path to export csv

Value

nothing, but creation of csv datasets

Examples

# create_fake_silc()

Cut a continuous variable in n categories.

Description

Cut a continuous variable in n categories.

Usage

cut_safe(x, n_breaks = 5)

Arguments

x

vector to cut
n_breaks

number of breaks

Value

vector

Description d'une variable, globale ou par groupe

Description

Remplace summarytools::dfSummary. Renvoie un data.frame prêt à afficher : une ligne par modalité (variables catégorielles) ou par statistique (variables continues), une colonne par groupe.

Usage

describe_variable(
  x,
  by = NULL,
  type = c("auto", "categorical", "continuous"),
  max_levels = 20,
  digits = 1
)

Arguments

x

vecteur à décrire
by

(optionnel) vecteur de même longueur, un bloc de colonnes par valeur
type

"auto" (numérique et plus de 15 valeurs distinctes -> continue), "categorical" ou "continuous"
max_levels

nombre maximum de modalités affichées, le reste est regroupé dans "(other)"
digits

nombre de décimales

Details

Les pourcentages des modalités sont calculés sur les valeurs valides, celui de la ligne "(missing)" sur l'ensemble des lignes.

Value

data.frame

Description d'une variable rendue en DT

Description

Description d'une variable rendue en DT

Usage

describe_variable_dt(x, by = NULL, ..., caption = NULL)

Arguments

x

vecteur à décrire
by

(optionnel) vecteur de même longueur, un bloc de colonnes par valeur
caption

(optionnel) titre du tableau

Value

datatable

Replace empty by na

Description

Replace empty by na

Usage

empty_as_na(vec)

Arguments

vec

Vector of values

Value

Vector

Examples

airquality[which(is.na(airquality$Ozone)), "Ozone"] <- ""
empty_as_na(airquality$Ozone)

extract a config from key (config from load_config)

Description

extract a config from key (config from load_config)

Usage

extract_config(config, key_)

Arguments

config

df of configuration
key_

key to extract

Value

string

Extract timer rds files from folder

Description

Extract timer rds files from folder

Usage

find_timer_rds(folder)

Arguments

folder

folder where check the file

Value

path of files

tranform data from folder to config and df

Description

tranform data from folder to config and df

Usage

folder_to_df(folder, file_pattern = "*.csv", file_config = "config.txt")

Arguments

folder

folder of databases
file_pattern

pattern of the databases (*.csv by default)
file_config

name of the configuration file (config.txt by default)

Value

list(df,configs)

Examples

## Not run: 
folder_to_df("ESS10")

## End(Not run)

Vecteur de groupe "VAR = valeur" / "VAR != valeur" pour les comparaisons

Description

Vecteur de groupe "VAR = valeur" / "VAR != valeur" pour les comparaisons

Usage

group_vs_others(x, value, label)

Arguments

x

Vecteur a regrouper
value

valeur discrétionnaire
label

label à utiliser

Value

vector

Create a heatmap

Description

Create a heatmap

Usage

heatmap_group(df_stats, threshold = 5, color = "red2")

Arguments

df_stats

data frame from prepa_stats function
threshold

threshold to show difference
color

color of the cells

Value

heatmap (ggplot)

Examples

library(laeken)
data(eusilc)

df_stats <- prepa_stats(eusilc, "db040")
heatmap_group(df_stats, 5)

index_homogeneity

Description

index_homogeneity

Usage

index_homogeneity(df, var_itw, threahold = 30)

Arguments

df

dataset
var_itw

groupe variable
threahold

threahold

Value

dataset

Check if value is integer64

Description

Check if value is integer64

Usage

is.integer64(x)

Arguments

x

Value

Value

Boolean

Examples

is.integer64(c(1:100)) # FALSE

First level of a multi-level key

Description

First level of a multi-level key

Usage

key_level1(x, sep = " / ")

Arguments

x

vector of keys
sep

separator between the levels

Value

vector of the first level values

Examples

key_level1(c("2024 / T1","2024 / T2","2023 / T4"))

Other levels of a multi-level key (used as label in the interface)

Description

Other levels of a multi-level key (used as label in the interface)

Usage

key_level2(x, sep = " / ")

Arguments

x

vector of keys
sep

separator between the levels

Value

vector of the values after the first level

Examples

key_level2(c("2024 / T1","2024 / T2","2023 / T4"))

Keys (modalities) of a wave or filter variable

Description

Keys (modalities) of a wave or filter variable

Usage

keys_vars(df, vars, var_key, level = "all", sep = " / ")

Arguments

df

data.frame
vars

variables of each level (see vars_levels)
var_key

name of the variable containing the complete key
level

1 (first level only), 2 (complete key) or "all" (both)
sep

separator between the levels

Value

vector of keys

Examples

keys_vars(mtcars,"cyl","cyl")

List distribution of discrete variables

Description

List distribution of discrete variables

Usage

list_dist(df, vars_vd)

Arguments

df

data.frame
vars_vd

vector of discrete variables

Value

list

Examples

list_dist(mtcars,c("cyl","vs","gear"))

List of proportions for categorical variables

Description

List of proportions for categorical variables

Usage

list_dist_dt(df, vars_discretes, na.rm = FALSE)

Arguments

df

data.frame
vars_discretes

vector of categorical variable to calculate proportion
na.rm

Remove missing values or not ?

Value

list

load a config file for prepare data

Description

load a config file for prepare data

Usage

load_config(file_path)

Arguments

file_path

path of the configuration file

Value

df

Loop of stats creation by filter

Description

Loop of stats creation by filter

Usage

loop_stats(df, configs, var_calculs, na.rm = FALSE, force_cat = FALSE)

Arguments

df

database
configs

configs
var_calculs

variable to create stats

Value

df

Rows of a data.frame matching one or several keys

Description

A row matches a key when its complete key is one of the keys, or when its first level is one of the keys (selection of a year without the quarter).

Usage

match_keys(df, vars, var_key, keys, sep = " / ")

Arguments

df

data.frame
vars

variables of each level (see vars_levels)
var_key

name of the variable containing the complete key
keys

vector of keys to keep
sep

separator between the levels

Value

logical vector

Examples

match_keys(mtcars,"cyl","cyl",4)

Specific chisq test to NA and Other modality

Description

Specific chisq test to NA and Other modality

Usage

my_chisq_test(x, varname, ldist, useNA = "ifany")

Arguments

x

value to procede chisq test
varname

name of the variable
ldist

named list of expected probability
useNA

useNA parameter for the table function

Value

chisq value

Examples

ldist <- list_dist(mtcars,c("cyl","gear"))
sub_mtcars <- subset(mtcars,vs == 1)
my_chisq_test(sub_mtcars$cyl,"cyl",ldist)

Create a summarise of all the difference

Description

Create a summarise of all the difference

Usage

prepa_stats(df, var_group, vars_vd = NULL, vars_vc = NULL)

Arguments

df

data frame for the summary
var_group

Name of group variable
vars_vd

(optional) Vector of discrete variables
vars_vc

(optional) Vector of continuous variables

Value

data frame

Examples

library(laeken)
data(eusilc)

info_vars <- classify_df(eusilc)
vars_vd <- info_vars[info_vars$type == "Modal", ]$variable
vars_vc <- info_vars[info_vars$type == "Continuous", ]$variable
prepa_stats(eusilc, "db040", vars_vd, vars_vc)

prepa_stats_dt

Description

prepa_stats_dt

Usage

prepa_stats_dt(df, var_group, configs, na.rm = FALSE)

Arguments

df
var_group
configs
na.rm

Value

data.frame

Preparation of a survey

Description

Preparation of a survey

Usage

prepa_survey(
  folder_path,
  file_pattern = "*.csv",
  file_config = "config.txt",
  na.rm = FALSE
)

Arguments

folder_path

folder of survey
file_pattern

pattern of the databases (*.csv by default)
file_config

name of the configuration file (config.txt by default)

Value

NULL (creation of rds)

Examples

## Not run: 
prepa_survey("shiny/complete/ESS10")

## End(Not run)

Preparation of all surveys from a folder

Description

Preparation of all surveys from a folder

Usage

prepa_surveys(
  folder_path,
  depth_folder = 1,
  file_pattern = "*.csv",
  file_config = "config.txt"
)

Arguments

folder_path

folder of the folders of survey
depth_folder

level of depth for the tree structure
file_pattern

pattern of the databases (*.csv by default)
file_config

name of the configuration file (config.txt by default)

Value

NULL (creation of rds)

Examples

## Not run: 
prepa_surveys("inst/extdata/SILC/HFILE")

## End(Not run)

Shiny vizsurvey

Description

Shiny vizsurvey

Usage

runVizsurvey()

Value

shinyapp

Examples

## Not run: runVizsurvey()

Shiny vizsurvey with already prepared data

Description

Shiny vizsurvey with already prepared data

Usage

runVizsurvey_from_folder(
  path,
  path_dict = NULL,
  path_nomen = NULL,
  path_archive = NULL,
  data_rds_pattern = "global",
  depth_folder = 1
)

Arguments

data_rds_pattern

name of the rds file contains all the data
depth_folder

level of depth for the tree structure
link

link to directory of data

Value

shinyapp

Examples

# We assume that config.txt, and prepa_surveys are already done here.
## Not run: runVizsurvey_from_folder("inst/extdata",depth_folder = 3)

Shiny vizsurvey from a R data.frame

Description

Shiny vizsurvey from a R data.frame

Usage

runVizsurvey_from_r(
  df,
  vars_discretes = NULL,
  vars_continuous = NULL,
  var_wave = NULL,
  var_filter = NULL,
  var_intvwr = NULL
)

Arguments

df

data.frame
vars_discretes

(optional) preset of discretes variables
vars_continuous

(optional) preset of continuous variables
var_wave

(optional) name of wave variable. Two variables can be given (c("YEAR","QUARTER")) : the second one is then a second level of wave
var_filter

(optional) name of filter variable. Two variables can be given : the second one is then a second level of filter
var_intvwr

(optional) name of interviewer variable

Value

shinyapp

Examples

library(laeken)
data(eusilc)
set.seed(123)
eusilc$NR_ITW <- paste(eusilc$db040,sample(1:5,nrow(eusilc),replace = TRUE),sep="-")
## Not run: runVizsurvey_from_r(eusilc,var_intvwr = "NR_ITW",var_filter = "db040")

Robust Scale of a varible with IQR

Description

Robust Scale of a varible with IQR

Usage

scale_IQR(x)

Arguments

x

vector

Value

vector

Examples

head(scale_IQR(iris$Sepal.Length))

calculate isoforest score from df

Description

calculate isoforest score from df

Usage

score_isoforest(df)

Arguments

df

database

Value

vector

Examples

score_isoforest(iris[sapply(iris, is.numeric)])

Variables of each level of the wave or of the filter

Description

Kept compatible with the global.rds prepared before the multi-level option : vars_wave (or vars_filter) is then missing and var_wave is used.

Usage

vars_levels(configs, type = "wave")

Arguments

configs

list of configuration
type

"wave" or "filter"

Value

vector of the variables of each level

Examples

vars_levels(list(var_wave = "YEAR"),"wave")