Data preparation

1 Presentation

This is the data preparation code for the paper “Changes in the Justification of Pension Inequality in Chile (2016–2023) and its Relationship to Social Class and Beliefs in Meritocracy”. The prepared dataset is ELSOC_Long_2016_2023_1.00.RData.

2 Libraries

if (! require("pacman")) install.packages("pacman")

pacman::p_load(tidyverse,
               car,
               sjmisc, 
               here,
               sjlabelled,
               SciViews,
               naniar,
               readxl,
               sjPlot)


options(scipen=999)
rm(list = ls())

3 Data

load(url("https://dataverse.harvard.edu/api/access/datafile/10797987"))

glimpse(elsoc_long_2016_2023)

4 Processing

elsoc_long_2016_2023[elsoc_long_2016_2023 ==-999] <- NA
elsoc_long_2016_2023[elsoc_long_2016_2023 ==-888] <- NA
elsoc_long_2016_2023[elsoc_long_2016_2023 ==-777] <- NA
elsoc_long_2016_2023[elsoc_long_2016_2023 ==-666] <- NA

elsoc_long_2016_2023 <- elsoc_long_2016_2023 %>% 
  mutate(just_pension = d02_01, 
         merit_effort = c18_09,
         merit_talent = c18_10, 
         age = m0_edad, m01, 
         sex = m0_sexo, 
         ideo = c15) %>% 
  as_tibble() %>% 
  sjlabelled::drop_labels(., drop.na = FALSE)

# Market Justice Preferences

frq(elsoc_long_2016_2023$just_pension)

Grado de acuerdo: Justicia distributiva en pensiones (x) <numeric> 
# total N=20761 valid N=17966 mean=2.24 sd=1.11

Value |                          Label |    N | Raw % | Valid % | Cum. %
------------------------------------------------------------------------
    1 |       Totalmente en desacuerdo | 4889 | 23.55 |   27.21 |  27.21
    2 |                  En desacuerdo | 7802 | 37.58 |   43.43 |  70.64
    3 | Ni de acuerdo ni en desacuerdo | 1740 |  8.38 |    9.68 |  80.32
    4 |                     De acuerdo | 3087 | 14.87 |   17.18 |  97.51
    5 |          Totalmente de acuerdo |  448 |  2.16 |    2.49 | 100.00
 <NA> |                           <NA> | 2795 | 13.46 |    <NA> |   <NA>

elsoc_long_2016_2023$just_pension <- car::recode(elsoc_long_2016_2023$just_pension, 
recodes = c("1='Strongly disagree'; 2='Disagree'; 3='Neither agree nor disagree'; 4='Agree'; 5='Strongly agree'"), 
levels = c("Strongly disagree", "Disagree", "Neither agree nor disagree", "Agree", "Strongly agree"),
as.factor = T)
                                                 
elsoc_long_2016_2023$just_pension <- sjlabelled::set_label(elsoc_long_2016_2023$just_pension, 
                        label = "Pension distributive justice")

# Social class scheme EOW

frq(elsoc_long_2016_2023$m07)

Relacion de empleo (x) <numeric> 
# total N=20761 valid N=7288 mean=2.35 sd=1.90

Value
-----
    1
    2
    3
    4
    5
    6
    7
 <NA>

                                                                              Label
-----------------------------------------------------------------------------------
                                               Empleado u obrero en empresa privada
     Empleado u obrero del sector publico (incluso empresa publica o municipalidad)
                                          Miembro de las Fuerzas Armadas y de Orden
Patron/a o empleador/a (contrata o paga a honorarios a uno/o o mas trabajadores/as)
                                                Trabaja solo, no tiene empleados/as
                                                             Familiar no remunerado
                                                                 Servicio domestico
                                                                               <NA>

    N | Raw % | Valid % | Cum. %
--------------------------------
 4160 | 20.04 |   57.08 |  57.08
  964 |  4.64 |   13.23 |  70.31
   80 |  0.39 |    1.10 |  71.41
  329 |  1.58 |    4.51 |  75.92
 1379 |  6.64 |   18.92 |  94.84
   15 |  0.07 |    0.21 |  95.05
  361 |  1.74 |    4.95 | 100.00
13473 | 64.90 |    <NA> |   <NA>

elsoc_long_2016_2023 <- elsoc_long_2016_2023 %>% 
  mutate(rel_empleo = factor(m07,
                             levels = 1:7,
                             labels= c("Empleado u obrero en empresa privada",
                                       "Empleado u obrero del sector público",
                                       "Miembro de las Fuerzas Armadas y de Orden",
                                       "Patrón/a o empleador/a",
                                       "Trabaja solo, no tiene empleados",
                                       "Familiar no remunerado",
                                       "Servicio doméstico"
                                       )))


# Definir los niveles y etiquetas originales
niveles_rel_empleo <- 1:7
etiquetas_rel_empleo <- c("Empleado u obrero en empresa privada",
                          "Empleado u obrero del sector público",
                          "Miembro de las Fuerzas Armadas y de Orden",
                          "Patrón/a o empleador/a",
                          "Trabaja solo, no tiene empleados",
                          "Familiar no remunerado",
                          "Servicio doméstico")

# Crear columna desplazada y rellenar valores NA
elsoc_long_2016_2023 <- elsoc_long_2016_2023 %>%
  group_by(idencuesta) %>%                                              # Agrupa por id para trabajar en cada individuo
  mutate(rel_empleo_lagged = lag(as.character(rel_empleo), n = 1)) %>%  # Desplaza rel_empleo a la siguiente ola
  ungroup() %>%
  mutate(rel_empleo = if_else(!is.na(rel_empleo), 
                              as.character(rel_empleo), 
                              rel_empleo_lagged),
         rel_empleo = factor(rel_empleo,                                # Convertir de nuevo a factor
                             levels = etiquetas_rel_empleo)) %>%
  select(-rel_empleo_lagged)                                            # Elimina la columna temporal

# Tabla de frecuencias y porcentajes
sjt.xtab(elsoc_long_2016_2023$rel_empleo,elsoc_long_2016_2023$ola,
         show.col.prc=TRUE,
         var.labels=c("Relación de empleo","Ola"),
         show.summary=FALSE,         title="Frecuencias y porcentajes de Relación de Empleo, por ola")

Frecuencias y porcentajes de Relación de Empleo, por ola

Relación de empleo	Ola							Total
	2016	2017	2018	2019	2021	2022	2023
Empleado u obrero en empresa privada	1091 61.1 %	898 60.3 %	1349 58.3 %	1147 57.8 %	897 54.1 %	731 52.8 %	866 54 %	6979 57.1 %
Empleado u obrero del sector público	186 10.4 %	155 10.4 %	291 12.6 %	262 13.2 %	229 13.8 %	195 14.1 %	264 16.5 %	1582 12.9 %
Miembro de las Fuerzas Armadas y de Orden	25 1.4 %	17 1.1 %	25 1.1 %	21 1.1 %	19 1.1 %	11 0.8 %	12 0.7 %	130 1.1 %
Patrón/a o empleador/a	85 4.8 %	72 4.8 %	109 4.7 %	92 4.6 %	86 5.2 %	67 4.8 %	55 3.4 %	566 4.6 %
Trabaja solo, no tiene empleados	321 18 %	277 18.6 %	415 17.9 %	354 17.8 %	347 20.9 %	307 22.2 %	309 19.3 %	2330 19.1 %
Familiar no remunerado	4 0.2 %	3 0.2 %	6 0.3 %	4 0.2 %	1 0.1 %	2 0.1 %	4 0.2 %	24 0.2 %
Servicio doméstico	75 4.2 %	67 4.5 %	118 5.1 %	106 5.3 %	78 4.7 %	71 5.1 %	94 5.9 %	609 5 %
Total	1787 100 %	1489 100 %	2313 100 %	1986 100 %	1657 100 %	1384 100 %	1604 100 %	12220 100 %

elsoc_long_2016_2023$rel_empleo2 <- car::recode(elsoc_long_2016_2023$rel_empleo,
                                recodes=c("'Patrón/a o empleador/a'='Empleador';
                                           'Trabaja solo, no tiene empleados'='Autoempleado';
                                           'Empleado u obrero en empresa privada'='Asalariado';
                                           'Empleado u obrero del sector público'='Asalariado';
                                           'Miembro de las Fuerzas Armadas y de Orden'='Asalariado';
                                           'Familiar no remunerado'='Asalariado';
                                           'Servicio doméstico'='Asalariado'"),
                                as.factor=TRUE, # convertir a factor
                                levels=c("Empleador",
                                         "Autoempleado",
                                         "Asalariado")) # ordenar niveles

# Tabla de frecuencias y porcentajes
sjt.xtab(elsoc_long_2016_2023$rel_empleo2,elsoc_long_2016_2023$ola,
         show.col.prc=TRUE,
         var.labels=c("Relación de empleo 2","Ola"),
         show.summary=FALSE,
         title="Frecuencias y porcentajes de Relación de Empleo 2, por ola")

Frecuencias y porcentajes de Relación de Empleo 2, por ola

Relación de empleo 2	Ola							Total
	2016	2017	2018	2019	2021	2022	2023
Empleador	85 4.8 %	72 4.8 %	109 4.7 %	92 4.6 %	86 5.2 %	67 4.8 %	55 3.4 %	566 4.6 %
Autoempleado	321 18 %	277 18.6 %	415 17.9 %	354 17.8 %	347 20.9 %	307 22.2 %	309 19.3 %	2330 19.1 %
Asalariado	1381 77.3 %	1140 76.6 %	1789 77.3 %	1540 77.5 %	1224 73.9 %	1010 73 %	1240 77.3 %	9324 76.3 %
Total	1787 100 %	1489 100 %	2313 100 %	1986 100 %	1657 100 %	1384 100 %	1604 100 %	12220 100 %

# Carga bbdd con comparacion 88 y 08
isco08_88 <- read_excel(here::here("input/data/original/isco08-88.xls"))

isco08_88 <- isco08_88 %>%
  rename(isco08=`ISCO 08 Code`,isco88=`ISCO-88 code`)

# Exploración
valores_repetidos <- table(isco08_88$isco88) # Contar los valores repetidos en la variable
valores_repetidos <- valores_repetidos[valores_repetidos > 1]
valores_repetidos <- table(isco08_88$isco08) # Contar los valores repetidos en la variable
valores_repetidos <- valores_repetidos[valores_repetidos > 1]

# CIUO - ISCO: Recodificación 88->08 4 digitos en datos elsoc
indices <- match(elsoc_long_2016_2023$ciuo88_m03,isco08_88$isco88)
elsoc_long_2016_2023$ciuo08_rec <- isco08_88$isco08[indices]
elsoc_long_2016_2023$ciuo08_rec <- as.numeric(elsoc_long_2016_2023$ciuo08_rec) # es solo la recodificación de las que eran 88 en data, es decir, año 2016

# Creación variable isco08 con ciuo08_m03 y ciuo08_rec (ex ciuo88_m03) 4 digitos
elsoc_long_2016_2023$isco08 <- ifelse(!is.na(elsoc_long_2016_2023$ciuo08_rec),elsoc_long_2016_2023$ciuo08_rec,elsoc_long_2016_2023$ciuo08_m03) # es la unión en una variable de todo en 08

# Crear una columna con la variable "isco08" adelantada una ola
elsoc_long_2016_2023 <- elsoc_long_2016_2023 %>%
  group_by(idencuesta) %>%        # Agrupa por id para trabajar en cada individuo
  mutate(isco08_lagged=lag(isco08,n=1)) %>%  # Desplaza isco08 a la siguiente ola
  ungroup()

# Rellenar los valores NA en la variable original
elsoc_long_2016_2023 <- elsoc_long_2016_2023 %>%
  mutate(isco08=ifelse(!is.na(isco08),isco08,isco08_lagged)) %>%  # Si isco08 es NA, sustituir con el valor de la ola anterior 
  select(-isco08_lagged)                 # Elimina la columna temporal


# CIUO - ISCO: creación de variables con 1 dígito
elsoc_long_2016_2023$isco08_1d <- as.character(elsoc_long_2016_2023$isco08)
elsoc_long_2016_2023$isco08_1d <- substr(elsoc_long_2016_2023$isco08_1d,1,1)
elsoc_long_2016_2023$isco08_1d <- as.numeric(elsoc_long_2016_2023$isco08_1d)

# CIUO - ISCO: creación de variables con 2 dígitos
elsoc_long_2016_2023$isco08_2d <- as.character(elsoc_long_2016_2023$isco08)
elsoc_long_2016_2023$isco08_2d <- substr(elsoc_long_2016_2023$isco08_2d,1,2)
elsoc_long_2016_2023$isco08_2d <- as.numeric(elsoc_long_2016_2023$isco08_2d)

# Tabla de frecuencias y porcentajes 1 dígito
sjt.xtab(elsoc_long_2016_2023$isco08_1d,elsoc_long_2016_2023$ola,
         show.col.prc=TRUE, 
         var.labels=c("CIUO08","Ola"),
         show.summary=FALSE,
         title="Frecuencias y porcentajes de Ocupación con 1 dígito, por ola")

Frecuencias y porcentajes de Ocupación con 1 dígito, por ola

CIUO08	Ola							Total
	2016	2017	2018	2019	2021	2022	2023
1	89 5 %	75 5 %	59 2.5 %	47 2.3 %	41 2.5 %	32 2.3 %	48 3 %	391 3.2 %
2	257 14.4 %	198 13.3 %	312 13.3 %	265 13.2 %	250 15.1 %	205 14.8 %	238 14.8 %	1725 14.1 %
3	362 20.3 %	304 20.4 %	180 7.7 %	150 7.5 %	148 8.9 %	116 8.4 %	202 12.6 %	1462 11.9 %
4	26 1.5 %	18 1.2 %	187 8 %	164 8.2 %	129 7.8 %	107 7.7 %	92 5.7 %	723 5.9 %
5	348 19.5 %	279 18.8 %	579 24.8 %	501 25 %	383 23.1 %	325 23.4 %	361 22.5 %	2776 22.6 %
6	41 2.3 %	37 2.5 %	27 1.2 %	22 1.1 %	30 1.8 %	26 1.9 %	23 1.4 %	206 1.7 %
7	319 17.9 %	280 18.8 %	346 14.8 %	301 15 %	297 17.9 %	259 18.7 %	243 15.1 %	2045 16.7 %
8	120 6.7 %	96 6.5 %	202 8.6 %	166 8.3 %	127 7.6 %	103 7.4 %	127 7.9 %	941 7.7 %
9	220 12.3 %	200 13.4 %	446 19.1 %	391 19.5 %	256 15.4 %	214 15.4 %	272 16.9 %	1999 16.3 %
Total	1782 100 %	1487 100 %	2338 100 %	2007 100 %	1661 100 %	1387 100 %	1606 100 %	12268 100 %

# Creación variable Nivel de cualificación
elsoc_long_2016_2023 <- mutate(elsoc_long_2016_2023,
               cualificacion=factor(case_when(isco08_2d<=34&m01>=9~"Experts",
                                              isco08_2d<=34&m01<9~"Skilled",
                                              isco08_2d%in%c(35,41:44,53)~"Skilled",
                                              isco08_2d%in%c(51,54,61,62,71:75,81)&m01>5~"Skilled",
                                              isco08_2d%in%c(51,54,61,62,71:75,81)&m01<=5~"Unskilled",
                                              isco08_2d%in%c(52,63,82,83,91:99)~"Unskilled"),
                                    levels=c("Experts","Skilled","Unskilled")))

# Tabla de frecuencias y porcentajes 
sjt.xtab(elsoc_long_2016_2023$cualificacion,elsoc_long_2016_2023$ola,
         show.col.prc=TRUE, 
         var.labels=c("Nivel de cualificación","Ola"),
         show.summary=FALSE,
         title="Frecuencias y porcentajes del Nivel de Cualificación, por ola")

Frecuencias y porcentajes del Nivel de Cualificación, por ola

Nivel de cualificación	Ola							Total
	2016	2017	2018	2019	2021	2022	2023
Experts	240 13.5 %	181 12.2 %	293 12.5 %	234 11.7 %	244 14.7 %	194 14 %	270 17 %	1656 13.5 %
Skilled	650 36.5 %	554 37.3 %	679 29.1 %	613 30.6 %	516 31.1 %	419 30.3 %	441 27.8 %	3872 31.6 %
Unskilled	892 50.1 %	752 50.6 %	1364 58.4 %	1158 57.8 %	900 54.2 %	772 55.7 %	873 55.1 %	6711 54.8 %
Total	1782 100 %	1487 100 %	2336 100 %	2005 100 %	1660 100 %	1385 100 %	1584 100 %	12239 100 %

# Crear variable supervisa==1, no supervisa==0
elsoc_long_2016_2023 <- elsoc_long_2016_2023 %>%
  mutate(supervisa=case_when(m06==0~0,
                             m06>=1~1))


# Crear una columna con la variable "m06" adelantada una ola
elsoc_long_2016_2023 <- elsoc_long_2016_2023 %>%
  arrange(ola) %>%  # Ordena por ola, por si acaso no está ordenado
  group_by(idencuesta) %>%        # Agrupa por id para trabajar en cada individuo
  mutate(m06_lagged=lag(m06,n=1)) %>%  # Desplaza "m06" a la siguiente ola
  ungroup() 

# Rellenar los valores NA en la variable original
elsoc_long_2016_2023 <- elsoc_long_2016_2023 %>%
  mutate(m06_full=ifelse(!is.na(m06),m06,m06_lagged)) %>%  # Si m06 es NA, sustituir con el valor de la ola anterior
  select(-m06_lagged) 

# Crear variable supervisa==1, no supervisa==0
elsoc_long_2016_2023 <- elsoc_long_2016_2023 %>%
  mutate(supervisa=case_when(m06_full==0~0,
                             m06_full>=1~1))

# Tabla de frecuencias y porcentajes 
sjt.xtab(elsoc_long_2016_2023$supervisa,elsoc_long_2016_2023$ola,
         show.col.prc=TRUE, 
         var.labels=c("Supervisa","Ola"),
         show.summary=FALSE,
         title="Frecuencias y porcentajes de variable Supervisa, por ola")

Frecuencias y porcentajes de variable Supervisa, por ola

Supervisa	Ola							Total
	2016	2017	2018	2019	2021	2022	2023
0	1339 74.3 %	1124 74.8 %	1788 77 %	1532 76.9 %	778 62.2 %	643 62.7 %	1112 71.1 %	8316 72.6 %
1	463 25.7 %	378 25.2 %	534 23 %	461 23.1 %	473 37.8 %	383 37.3 %	452 28.9 %	3144 27.4 %
Total	1802 100 %	1502 100 %	2322 100 %	1993 100 %	1251 100 %	1026 100 %	1564 100 %	11460 100 %

# Creación de variable posición de clase
elsoc_long_2016_2023 <- mutate(elsoc_long_2016_2023,
               class1=case_when(
                 rel_empleo2=="Empleador"~"Petit bourgeoisie",
                 rel_empleo2=="Autoempleado"&m01>=9~"Petit bourgeoisie",
                 rel_empleo2=="Autoempleado"&m01<9~"Informal",
                 rel_empleo2=="Asalariado"&cualificacion=="Experts"&supervisa==1~"Expert managers",
                 rel_empleo2=="Asalariado"&cualificacion=="Experts"&supervisa==0~"Experts",
                 rel_empleo2=="Asalariado"&cualificacion=="Skilled"&supervisa==1~"Supervisors",
                 rel_empleo2=="Asalariado"&cualificacion=="Unskilled"&supervisa==1~"Supervisors",
                 rel_empleo2=="Asalariado"&cualificacion=="Skilled"&supervisa==0~"Workers",
                 rel_empleo2=="Asalariado"&cualificacion=="Unskilled"&supervisa==0~"Workers"))

elsoc_long_2016_2023$class1 <- factor(elsoc_long_2016_2023$class1,
                      levels=c("Petit bourgeoisie",
                               "Informal",
                               "Expert managers",
                               "Experts",
                               "Supervisors",
                               "Workers"))

# Tabla de frecuencias y porcentajes 
sjt.xtab(elsoc_long_2016_2023$class1, elsoc_long_2016_2023$ola,
         show.col.prc = TRUE, 
         var.labels = c("Posición de clase 1","Ola"),
         show.summary = FALSE,
         title = "Frecuencias y porcentajes de la Posición de clase 1, por ola")

Frecuencias y porcentajes de la Posición de clase 1, por ola

Posición de clase 1	Ola							Total
	2016	2017	2018	2019	2021	2022	2023
Petit bourgeoisie	105 5.9 %	89 6 %	143 6.2 %	121 6.1 %	118 7.6 %	96 7.5 %	84 5.4 %	756 6.4 %
Informal	301 17 %	260 17.6 %	381 16.6 %	323 16.4 %	315 20.4 %	278 21.7 %	280 18 %	2138 18 %
Expert managers	93 5.3 %	70 4.7 %	119 5.2 %	95 4.8 %	115 7.4 %	95 7.4 %	138 8.9 %	725 6.1 %
Experts	112 6.3 %	84 5.7 %	129 5.6 %	103 5.2 %	94 6.1 %	71 5.5 %	99 6.4 %	692 5.8 %
Supervisors	275 15.5 %	229 15.5 %	327 14.2 %	291 14.8 %	275 17.8 %	222 17.3 %	253 16.3 %	1872 15.7 %
Workers	884 49.9 %	743 50.4 %	1199 52.2 %	1039 52.7 %	630 40.7 %	520 40.6 %	700 45 %	5715 48 %
Total	1770 100 %	1475 100 %	2298 100 %	1972 100 %	1547 100 %	1282 100 %	1554 100 %	11898 100 %

# Meritocracy

frq(elsoc_long_2016_2023$merit_effort)

Grado de acuerdo: Las personas son recompensadas por sus esfuerzos (x) <numeric> 
# total N=20761 valid N=20629 mean=2.60 sd=1.02

Value |                          Label |    N | Raw % | Valid % | Cum. %
------------------------------------------------------------------------
    1 |       Totalmente en desacuerdo | 2173 | 10.47 |   10.53 |  10.53
    2 |                  En desacuerdo | 9442 | 45.48 |   45.77 |  56.30
    3 | Ni en desacuerdo ni de acuerdo | 4037 | 19.45 |   19.57 |  75.87
    4 |                     De acuerdo | 4379 | 21.09 |   21.23 |  97.10
    5 |          Totalmente de acuerdo |  598 |  2.88 |    2.90 | 100.00
 <NA> |                           <NA> |  132 |  0.64 |    <NA> |   <NA>

frq(elsoc_long_2016_2023$merit_talent)

Grado de acuerdo: Las personas son recompensada por su inteligencia (x) <numeric> 
# total N=20761 valid N=20631 mean=2.77 sd=1.03

Value |                          Label |    N | Raw % | Valid % | Cum. %
------------------------------------------------------------------------
    1 |       Totalmente en desacuerdo | 1766 |  8.51 |    8.56 |   8.56
    2 |                  En desacuerdo | 8108 | 39.05 |   39.30 |  47.86
    3 | Ni en desacuerdo ni de acuerdo | 4571 | 22.02 |   22.16 |  70.02
    4 |                     De acuerdo | 5577 | 26.86 |   27.03 |  97.05
    5 |          Totalmente de acuerdo |  609 |  2.93 |    2.95 | 100.00
 <NA> |                           <NA> |  130 |  0.63 |    <NA> |   <NA>

elsoc_long_2016_2023 <- elsoc_long_2016_2023 %>% 
  mutate(
    across(
      .cols = c(merit_effort, merit_talent),
      .fns = ~ car::recode(., recodes = c("1='Strongly disagree'; 2='Disagree';
                                          3='Neither agree nor disagree'; 4='Agree';
                                          5='Strongly agree'"), 
                           levels = c("Strongly disagree", "Disagree", "Neither agree nor disagree", "Agree", "Strongly agree"),
                           as.factor = T)
    )
  )


elsoc_long_2016_2023$merit_effort <- sjlabelled::set_label(elsoc_long_2016_2023$merit_effort, 
                        label = "People are rewarded for their efforts")

elsoc_long_2016_2023$merit_talent <- sjlabelled::set_label(elsoc_long_2016_2023$merit_talent, 
                        label = "People are rewarded for their intelligence")

# Controls

# sex
elsoc_long_2016_2023$sex <- car::recode(elsoc_long_2016_2023$sex, 
                           recodes = c("1='Male'; 2='Female'"), 
                           levels = c("Male", "Female"),
                           as.factor = T)

elsoc_long_2016_2023$sex <- sjlabelled::set_label(elsoc_long_2016_2023$sex, 
                        label = "Gender")

# age
frq(elsoc_long_2016_2023$age)

Edad del entrevistado (x) <numeric> 
# total N=20761 valid N=20761 mean=48.84 sd=15.43

Value |   N | Raw % | Valid % | Cum. %
--------------------------------------
   18 |  34 |  0.16 |    0.16 |   0.16
   19 |  72 |  0.35 |    0.35 |   0.51
   20 | 113 |  0.54 |    0.54 |   1.05
   21 | 156 |  0.75 |    0.75 |   1.81
   22 | 187 |  0.90 |    0.90 |   2.71
   23 | 231 |  1.11 |    1.11 |   3.82
   24 | 262 |  1.26 |    1.26 |   5.08
   25 | 324 |  1.56 |    1.56 |   6.64
   26 | 318 |  1.53 |    1.53 |   8.17
   27 | 312 |  1.50 |    1.50 |   9.68
   28 | 361 |  1.74 |    1.74 |  11.42
   29 | 345 |  1.66 |    1.66 |  13.08
   30 | 365 |  1.76 |    1.76 |  14.84
   31 | 364 |  1.75 |    1.75 |  16.59
   32 | 369 |  1.78 |    1.78 |  18.37
   33 | 377 |  1.82 |    1.82 |  20.18
   34 | 384 |  1.85 |    1.85 |  22.03
   35 | 373 |  1.80 |    1.80 |  23.83
   36 | 444 |  2.14 |    2.14 |  25.97
   37 | 373 |  1.80 |    1.80 |  27.76
   38 | 411 |  1.98 |    1.98 |  29.74
   39 | 365 |  1.76 |    1.76 |  31.50
   40 | 386 |  1.86 |    1.86 |  33.36
   41 | 391 |  1.88 |    1.88 |  35.24
   42 | 395 |  1.90 |    1.90 |  37.15
   43 | 379 |  1.83 |    1.83 |  38.97
   44 | 361 |  1.74 |    1.74 |  40.71
   45 | 370 |  1.78 |    1.78 |  42.49
   46 | 410 |  1.97 |    1.97 |  44.47
   47 | 411 |  1.98 |    1.98 |  46.45
   48 | 405 |  1.95 |    1.95 |  48.40
   49 | 403 |  1.94 |    1.94 |  50.34
   50 | 451 |  2.17 |    2.17 |  52.51
   51 | 417 |  2.01 |    2.01 |  54.52
   52 | 462 |  2.23 |    2.23 |  56.75
   53 | 413 |  1.99 |    1.99 |  58.74
   54 | 463 |  2.23 |    2.23 |  60.97
   55 | 495 |  2.38 |    2.38 |  63.35
   56 | 498 |  2.40 |    2.40 |  65.75
   57 | 449 |  2.16 |    2.16 |  67.91
   58 | 468 |  2.25 |    2.25 |  70.17
   59 | 461 |  2.22 |    2.22 |  72.39
   60 | 481 |  2.32 |    2.32 |  74.70
   61 | 397 |  1.91 |    1.91 |  76.61
   62 | 393 |  1.89 |    1.89 |  78.51
   63 | 386 |  1.86 |    1.86 |  80.37
   64 | 355 |  1.71 |    1.71 |  82.08
   65 | 362 |  1.74 |    1.74 |  83.82
   66 | 316 |  1.52 |    1.52 |  85.34
   67 | 335 |  1.61 |    1.61 |  86.96
   68 | 253 |  1.22 |    1.22 |  88.17
   69 | 260 |  1.25 |    1.25 |  89.43
   70 | 282 |  1.36 |    1.36 |  90.79
   71 | 268 |  1.29 |    1.29 |  92.08
   72 | 230 |  1.11 |    1.11 |  93.18
   73 | 235 |  1.13 |    1.13 |  94.32
   74 | 241 |  1.16 |    1.16 |  95.48
   75 | 245 |  1.18 |    1.18 |  96.66
   76 | 180 |  0.87 |    0.87 |  97.52
   77 | 152 |  0.73 |    0.73 |  98.26
   78 | 105 |  0.51 |    0.51 |  98.76
   79 |  98 |  0.47 |    0.47 |  99.23
   80 |  62 |  0.30 |    0.30 |  99.53
   81 |  40 |  0.19 |    0.19 |  99.73
   82 |  19 |  0.09 |    0.09 |  99.82
   83 |   9 |  0.04 |    0.04 |  99.86
   84 |  11 |  0.05 |    0.05 |  99.91
   85 |   4 |  0.02 |    0.02 |  99.93
   86 |   2 |  0.01 |    0.01 |  99.94
   87 |   2 |  0.01 |    0.01 |  99.95
   88 |   3 |  0.01 |    0.01 |  99.97
   89 |   3 |  0.01 |    0.01 |  99.98
   90 |   2 |  0.01 |    0.01 |  99.99
   91 |   1 |  0.00 |    0.00 | 100.00
   92 |   1 |  0.00 |    0.00 | 100.00
 <NA> |   0 |  0.00 |    <NA> |   <NA>

elsoc_long_2016_2023$age <- 
  factor(car::recode(elsoc_long_2016_2023$age, 
                     "18:29=1;30:49=2;50:64=3;65:150=4"),
         labels = c('18-29', '30-49', '50-64', '65 or more'))
elsoc_long_2016_2023$age <-
  sjlabelled::set_label(elsoc_long_2016_2023$age, 
                        label = c("Age groups")) 


# political indentification

frq(elsoc_long_2016_2023$ideo)

Autoubicacion escala izquierda-derecha (x) <numeric> 
# total N=20761 valid N=20443 mean=7.39 sd=3.96

Value |            Label |    N | Raw % | Valid % | Cum. %
----------------------------------------------------------
    0 |      0 Izquierda | 1146 |  5.52 |    5.61 |   5.61
    1 |                1 |  386 |  1.86 |    1.89 |   7.49
    2 |                2 |  625 |  3.01 |    3.06 |  10.55
    3 |                3 | 1007 |  4.85 |    4.93 |  15.48
    4 |                4 | 1193 |  5.75 |    5.84 |  21.31
    5 |         5 Centro | 5360 | 25.82 |   26.22 |  47.53
    6 |                6 |  749 |  3.61 |    3.66 |  51.20
    7 |                7 |  707 |  3.41 |    3.46 |  54.65
    8 |                8 |  601 |  2.89 |    2.94 |  57.59
    9 |                9 |  177 |  0.85 |    0.87 |  58.46
   10 |       10 Derecha | 1135 |  5.47 |    5.55 |  64.01
   11 | 11 Independiente |  727 |  3.50 |    3.56 |  67.57
   12 |       12 Ninguno | 6630 | 31.93 |   32.43 | 100.00
 <NA> |             <NA> |  318 |  1.53 |    <NA> |   <NA>

elsoc_long_2016_2023$ideo<-
factor(
  car::recode(
    elsoc_long_2016_2023$ideo,
    "c(11,12,-888,-999)='Does not identify';c(0,1,2,3,4)='Left';
     c(5)='Center';c(6,7,8,9,10)='Right'"
  ),
  levels = c('Left', 'Center', 'Right', 'Does not identify')
)

elsoc_long_2016_2023$ideo<- factor(elsoc_long_2016_2023$ideo,levels = levels(elsoc_long_2016_2023$ideo))

elsoc_long_2016_2023$ideo <- 
sjlabelled::set_label(x = elsoc_long_2016_2023$ideo, 
                      label = "Political identification") 

frq(elsoc_long_2016_2023$ideo)

Political identification (x) <categorical> 
# total N=20761 valid N=20443 mean=2.67 sd=1.17

Value             |    N | Raw % | Valid % | Cum. %
---------------------------------------------------
Left              | 4357 | 20.99 |   21.31 |  21.31
Center            | 5360 | 25.82 |   26.22 |  47.53
Right             | 3369 | 16.23 |   16.48 |  64.01
Does not identify | 7357 | 35.44 |   35.99 | 100.00
<NA>              |  318 |  1.53 |    <NA> |   <NA>

# Socioeconomic_________________________________________________________________

# Education_______________________________________
elsoc_long_2016_2023$educ <- 
  car::recode(elsoc_long_2016_2023$m01,
              "c(1,2,3,4,5,6,7)=1;c(8,9,10)=2; c(-888,-999)=NA")
elsoc_long_2016_2023$educ <-
  factor(elsoc_long_2016_2023$educ,
         labels = c("Less than Universitary","Universitary"))

#reverse education, reference level is the highest level
#elsoc_long_2016_2023$educ <- forcats::fct_rev(elsoc_long_2016_2023$educ)

elsoc_long_2016_2023$educ <- 
sjlabelled::set_label(x = elsoc_long_2016_2023$educ,
                      label = "Education")
sjmisc::frq(elsoc_long_2016_2023$educ)

Education (x) <categorical> 
# total N=20761 valid N=20746 mean=1.19 sd=0.39

Value                  |     N | Raw % | Valid % | Cum. %
---------------------------------------------------------
Less than Universitary | 16786 | 80.85 |   80.91 |  80.91
Universitary           |  3960 | 19.07 |   19.09 | 100.00
<NA>                   |    15 |  0.07 |    <NA> |   <NA>

#Recoding of education to years based on casen 2017.
elsoc_long_2016_2023$educyear<- as.numeric(
  car::recode(elsoc_long_2016_2023$m01, 
              "1=0;2=4.3;3=7.5;4=9.8;5=12.02;6=13.9;
               7=14.8;8=14.9;9=16.9;10=19.07;c(-888,-999)=NA", 
              as.numeric = T))

elsoc_long_2016_2023$educyear <- 
sjlabelled::set_label(x = elsoc_long_2016_2023$educyear,
                      label = "Education in years")

class(elsoc_long_2016_2023$educyear)

[1] "numeric"

sjmisc::frq(elsoc_long_2016_2023$educyear)

Education in years (x) <numeric> 
# total N=20761 valid N=20746 mean=11.56 sd=3.99

Value |    N | Raw % | Valid % | Cum. %
---------------------------------------
 0.00 |  200 |  0.96 |    0.96 |   0.96
 4.30 | 2412 | 11.62 |   11.63 |  12.59
 7.50 | 1992 |  9.59 |    9.60 |  22.19
 9.80 | 2692 | 12.97 |   12.98 |  35.17
12.02 | 6122 | 29.49 |   29.51 |  64.68
13.90 |  749 |  3.61 |    3.61 |  68.29
14.80 | 2619 | 12.61 |   12.62 |  80.91
14.90 | 1197 |  5.77 |    5.77 |  86.68
16.90 | 2413 | 11.62 |   11.63 |  98.31
19.07 |  350 |  1.69 |    1.69 | 100.00
 <NA> |   15 |  0.07 |    <NA> |   <NA>

# Reshape long to wide

df_study1_long <- elsoc_long_2016_2023 %>% 
  select(idencuesta,
         ola,
         muestra,
         ponderador_long_total, 
         segmento, 
         estrato,
         just_pension,
         class1,
         merit_effort,
         merit_talent,
         educ,
         educyear,
         sex,
         age,
         ideo)

df_study1_wide <- df_study1_long %>% 
  tidyr::pivot_wider(id_cols = c("idencuesta","muestra"),
                     names_from = "ola",
                     values_from = names(select(df_study1_long,ponderador_long_total,segmento, estrato, just_pension:ideo)))

# fix data to w01 values
df_study1_wide$class1_2 <-df_study1_wide$class1_1 #class
df_study1_wide$class1_3 <-df_study1_wide$class1_1
df_study1_wide$class1_4 <-df_study1_wide$class1_1
df_study1_wide$class1_5 <-df_study1_wide$class1_1
df_study1_wide$class1_6 <-df_study1_wide$class1_1
df_study1_wide$class1_7 <-df_study1_wide$class1_1


df_study1_wide$age_2 <-df_study1_wide$age_1 #age
df_study1_wide$age_3 <-df_study1_wide$age_1
df_study1_wide$age_4 <-df_study1_wide$age_1
df_study1_wide$age_5 <-df_study1_wide$age_1
df_study1_wide$age_6 <-df_study1_wide$age_1
df_study1_wide$age_7 <-df_study1_wide$age_1

df_study1_wide$sex_2 <-df_study1_wide$sex_1 #sex
df_study1_wide$sex_3 <-df_study1_wide$sex_1
df_study1_wide$sex_4 <-df_study1_wide$sex_1
df_study1_wide$sex_5 <-df_study1_wide$sex_1
df_study1_wide$sex_6 <-df_study1_wide$sex_1
df_study1_wide$sex_7 <-df_study1_wide$sex_1

df_study1_wide$educ_2 <-df_study1_wide$educ_1 #education
df_study1_wide$educ_3 <-df_study1_wide$educ_1
df_study1_wide$educ_4 <-df_study1_wide$educ_1
df_study1_wide$educ_5 <-df_study1_wide$educ_1
df_study1_wide$educ_6 <-df_study1_wide$educ_1
df_study1_wide$educ_7 <-df_study1_wide$educ_1

df_study1_wide$educyear_2 <-df_study1_wide$educyear_1 #education years
df_study1_wide$educyear_3 <-df_study1_wide$educyear_1
df_study1_wide$educyear_4 <-df_study1_wide$educyear_1
df_study1_wide$educyear_5 <-df_study1_wide$educyear_1
df_study1_wide$educyear_6 <-df_study1_wide$educyear_1
df_study1_wide$educyear_7 <-df_study1_wide$educyear_1


df_study1_wide$ideo_2 <-df_study1_wide$ideo_1 # political position
df_study1_wide$ideo_3 <-df_study1_wide$ideo_1
df_study1_wide$ideo_4 <-df_study1_wide$ideo_1
df_study1_wide$ideo_5 <-df_study1_wide$ideo_1
df_study1_wide$ideo_6 <-df_study1_wide$ideo_1
df_study1_wide$ideo_7 <-df_study1_wide$ideo_1


dim(df_study1_wide)

[1] 4447   86

sjPlot::view_df(df_study1_wide,
                show.frq = T,show.values = T,show.na = T,show.prc = T, show.type = T)

just_pension_5 [28]

Data frame: df_study1_wide

ID	Name	Type	Label	missings	Values	Value Labels	Freq.	%
1	idencuesta	numeric	Folio identificador de participante	0 (0.00%)	range: 1101011-NA
2	muestra	numeric	Identificador de muestra de encuesta	0 (0.00%)	1 2	Muestra Original Muestra Refresco	2928 1519	65.84 34.16
3	ponderador_long_total_1	numeric	Ponderador longitudinal (2016-2022). Muestra completa	1520 (34.18%)	range: 0.0-6.6
4	ponderador_long_total_2	numeric	Ponderador longitudinal (2016-2022). Muestra completa	1974 (44.39%)	range: 0.0-7.7
5	ponderador_long_total_3	numeric	Ponderador longitudinal (2016-2022). Muestra completa	699 (15.72%)	range: 0.0-9.0
6	ponderador_long_total_4	numeric	Ponderador longitudinal (2016-2022). Muestra completa	1030 (23.16%)	range: 0.0-11.8
7	ponderador_long_total_5	numeric	Ponderador longitudinal (2016-2022). Muestra completa	1707 (38.39%)	range: 0.0-13.8
8	ponderador_long_total_6	numeric	Ponderador longitudinal (2016-2022). Muestra completa	1717 (38.61%)	range: 0.0-15.0
9	ponderador_long_total_7	numeric	Ponderador longitudinal (2016-2022). Muestra completa	1721 (38.70%)	range: 0.0-16.4
10	segmento_1	numeric	Segmento	1520 (34.18%)	range: 110101-1420411
11	segmento_2	numeric	Segmento	1974 (44.39%)	range: 110101-1420411
12	segmento_3	numeric	Segmento	699 (15.72%)	range: 110101-1420411
13	segmento_4	numeric	Segmento	1030 (23.16%)	range: 110101-1420411
14	segmento_5	numeric	Segmento	1707 (38.39%)	range: 110101-1420411
15	segmento_6	numeric	Segmento	1717 (38.61%)	range: 110101-1420411
16	segmento_7	numeric	Segmento	1721 (38.70%)	range: 110101-1420411
17	estrato_1	numeric	Estrato muestral	1520 (34.18%)	1 2 3 4 5 6	Gran Santiago Gran Valparaiso Gran Concepcion Ciudades Grandes Ciudades Medianas Ciudades Pequennias	720 375 391 408 567 466	24.60 12.81 13.36 13.94 19.37 15.92
18	estrato_2	numeric	Estrato muestral	1974 (44.39%)	1 2 3 4 5 6	Gran Santiago Gran Valparaiso Gran Concepcion Ciudades Grandes Ciudades Medianas Ciudades Pequennias	597 314 335 363 457 407	24.14 12.70 13.55 14.68 18.48 16.46
19	estrato_3	numeric	Estrato muestral	699 (15.72%)	1 2 3 4 5 6	Gran Santiago Gran Valparaiso Gran Concepcion Ciudades Grandes Ciudades Medianas Ciudades Pequennias	934 437 514 623 636 604	24.92 11.66 13.71 16.62 16.97 16.12
20	estrato_4	numeric	Estrato muestral	1030 (23.16%)	1 2 3 4 5 6	Gran Santiago Gran Valparaiso Gran Concepcion Ciudades Grandes Ciudades Medianas Ciudades Pequennias	833 395 467 568 593 561	24.38 11.56 13.67 16.62 17.35 16.42
21	estrato_5	numeric	Estrato muestral	1707 (38.39%)	1 2 3 4 5 6	Gran Santiago Gran Valparaiso Gran Concepcion Ciudades Grandes Ciudades Medianas Ciudades Pequennias	671 316 394 456 470 433	24.49 11.53 14.38 16.64 17.15 15.80
22	estrato_6	numeric	Estrato muestral	1717 (38.61%)	1 2 3 4 5 6	Gran Santiago Gran Valparaiso Gran Concepcion Ciudades Grandes Ciudades Medianas Ciudades Pequennias	688 298 369 449 440 486	25.20 10.92 13.52 16.45 16.12 17.80
23	estrato_7	numeric	Estrato muestral	1721 (38.70%)	1 2 3 4 5 6	Gran Santiago Gran Valparaiso Gran Concepcion Ciudades Grandes Ciudades Medianas Ciudades Pequennias	665 308 359 441 467 486	24.39 11.30 13.17 16.18 17.13 17.83
24	just_pension_1	categorical	Pension distributive justice	1524 (34.27%)		Strongly disagree Disagree Neither agree nor disagree Agree Strongly agree	862 1337 228 443 53	29.49 45.74 7.80 15.16 1.81
25	just_pension_2	categorical	Pension distributive justice	1982 (44.57%)		Strongly disagree Disagree Neither agree nor disagree Agree Strongly agree	906 1022 172 314 51	36.75 41.46 6.98 12.74 2.07
26	just_pension_3	categorical	Pension distributive justice	718 (16.15%)		Strongly disagree Disagree Neither agree nor disagree Agree Strongly agree	1253 1517 283 573 103	33.60 40.68 7.59 15.37 2.76
27	just_pension_4	categorical	Pension distributive justice	1038 (23.34%)		Strongly disagree Disagree Neither agree nor disagree Agree Strongly agree	963 1598 321 467 60	28.25 46.88 9.42 13.70 1.76
29	just_pension_6	categorical	Pension distributive justice	1726 (38.81%)		Strongly disagree Disagree Neither agree nor disagree Agree Strongly agree	415 1207 348 661 90	15.25 44.36 12.79 24.29 3.31
30	just_pension_7	categorical	Pension distributive justice	1728 (38.86%)		Strongly disagree Disagree Neither agree nor disagree Agree Strongly agree	490 1121 388 629 91	18.02 41.23 14.27 23.13 3.35
31	class1_1	categorical		2677 (60.20%)		Petit bourgeoisie Informal Expert managers Experts Supervisors Workers	105 301 93 112 275 884	5.93 17.01 5.25 6.33 15.54 49.94
32	class1_2	categorical		2677 (60.20%)		Petit bourgeoisie Informal Expert managers Experts Supervisors Workers	105 301 93 112 275 884	5.93 17.01 5.25 6.33 15.54 49.94
33	class1_3	categorical		2677 (60.20%)		Petit bourgeoisie Informal Expert managers Experts Supervisors Workers	105 301 93 112 275 884	5.93 17.01 5.25 6.33 15.54 49.94
34	class1_4	categorical		2677 (60.20%)		Petit bourgeoisie Informal Expert managers Experts Supervisors Workers	105 301 93 112 275 884	5.93 17.01 5.25 6.33 15.54 49.94
35	class1_5	categorical		2677 (60.20%)		Petit bourgeoisie Informal Expert managers Experts Supervisors Workers	105 301 93 112 275 884	5.93 17.01 5.25 6.33 15.54 49.94
36	class1_6	categorical		2677 (60.20%)		Petit bourgeoisie Informal Expert managers Experts Supervisors Workers	105 301 93 112 275 884	5.93 17.01 5.25 6.33 15.54 49.94
37	class1_7	categorical		2677 (60.20%)		Petit bourgeoisie Informal Expert managers Experts Supervisors Workers	105 301 93 112 275 884	5.93 17.01 5.25 6.33 15.54 49.94
38	merit_effort_1	categorical	People are rewarded for their efforts	1538 (34.59%)		Strongly disagree Disagree Neither agree nor disagree Agree Strongly agree	357 1331 497 646 78	12.27 45.75 17.08 22.21 2.68
39	merit_effort_2	categorical	People are rewarded for their efforts	1988 (44.70%)		Strongly disagree Disagree Neither agree nor disagree Agree Strongly agree	282 1057 478 556 86	11.47 42.98 19.44 22.61 3.50
40	merit_effort_3	categorical	People are rewarded for their efforts	737 (16.57%)		Strongly disagree Disagree Neither agree nor disagree Agree Strongly agree	347 1482 797 925 159	9.35 39.95 21.48 24.93 4.29
41	merit_effort_4	categorical	People are rewarded for their efforts	1046 (23.52%)		Strongly disagree Disagree Neither agree nor disagree Agree Strongly agree	415 1604 631 653 98	12.20 47.16 18.55 19.20 2.88
42	merit_effort_5	categorical	People are rewarded for their efforts	1735 (39.02%)		Strongly disagree Disagree Neither agree nor disagree Agree Strongly agree	237 1415 422 567 71	8.74 52.18 15.56 20.91 2.62
43	merit_effort_6	categorical	People are rewarded for their efforts	1729 (38.88%)		Strongly disagree Disagree Neither agree nor disagree Agree Strongly agree	265 1323 587 491 52	9.75 48.68 21.60 18.06 1.91
44	merit_effort_7	categorical	People are rewarded for their efforts	1727 (38.84%)		Strongly disagree Disagree Neither agree nor disagree Agree Strongly agree	270 1230 625 541 54	9.93 45.22 22.98 19.89 1.99
45	merit_talent_1	categorical	People are rewarded for their intelligence	1540 (34.63%)		Strongly disagree Disagree Neither agree nor disagree Agree Strongly agree	288 1163 559 814 83	9.91 40.01 19.23 28.00 2.86
46	merit_talent_2	categorical	People are rewarded for their intelligence	1987 (44.68%)		Strongly disagree Disagree Neither agree nor disagree Agree Strongly agree	228 901 555 675 101	9.27 36.63 22.56 27.44 4.11
47	merit_talent_3	categorical	People are rewarded for their intelligence	739 (16.62%)		Strongly disagree Disagree Neither agree nor disagree Agree Strongly agree	277 1257 838 1177 159	7.47 33.90 22.60 31.74 4.29
48	merit_talent_4	categorical	People are rewarded for their intelligence	1047 (23.54%)		Strongly disagree Disagree Neither agree nor disagree Agree Strongly agree	340 1329 775 862 94	10.00 39.09 22.79 25.35 2.76
49	merit_talent_5	categorical	People are rewarded for their intelligence	1733 (38.97%)		Strongly disagree Disagree Neither agree nor disagree Agree Strongly agree	168 1290 495 694 67	6.19 47.53 18.24 25.57 2.47
50	merit_talent_6	categorical	People are rewarded for their intelligence	1726 (38.81%)		Strongly disagree Disagree Neither agree nor disagree Agree Strongly agree	238 1108 667 658 50	8.75 40.72 24.51 24.18 1.84
51	merit_talent_7	categorical	People are rewarded for their intelligence	1726 (38.81%)		Strongly disagree Disagree Neither agree nor disagree Agree Strongly agree	227 1060 682 697 55	8.34 38.96 25.06 25.62 2.02
52	educ_1	categorical	Education	1522 (34.23%)		Less than Universitary Universitary	2390 535	81.71 18.29
53	educ_2	categorical	Education	1522 (34.23%)		Less than Universitary Universitary	2390 535	81.71 18.29
54	educ_3	categorical	Education	1522 (34.23%)		Less than Universitary Universitary	2390 535	81.71 18.29
55	educ_4	categorical	Education	1522 (34.23%)		Less than Universitary Universitary	2390 535	81.71 18.29
56	educ_5	categorical	Education	1522 (34.23%)		Less than Universitary Universitary	2390 535	81.71 18.29
57	educ_6	categorical	Education	1522 (34.23%)		Less than Universitary Universitary	2390 535	81.71 18.29
58	educ_7	categorical	Education	1522 (34.23%)		Less than Universitary Universitary	2390 535	81.71 18.29
59	educyear_1	numeric	Education in years	1522 (34.23%)	range: 0.0-19.1
60	educyear_2	numeric	Education in years	1522 (34.23%)	range: 0.0-19.1
61	educyear_3	numeric	Education in years	1522 (34.23%)	range: 0.0-19.1
62	educyear_4	numeric	Education in years	1522 (34.23%)	range: 0.0-19.1
63	educyear_5	numeric	Education in years	1522 (34.23%)	range: 0.0-19.1
64	educyear_6	numeric	Education in years	1522 (34.23%)	range: 0.0-19.1
65	educyear_7	numeric	Education in years	1522 (34.23%)	range: 0.0-19.1
66	sex_1	categorical	Gender	1520 (34.18%)		Male Female	1163 1764	39.73 60.27
67	sex_2	categorical	Gender	1520 (34.18%)		Male Female	1163 1764	39.73 60.27
68	sex_3	categorical	Gender	1520 (34.18%)		Male Female	1163 1764	39.73 60.27
69	sex_4	categorical	Gender	1520 (34.18%)		Male Female	1163 1764	39.73 60.27
70	sex_5	categorical	Gender	1520 (34.18%)		Male Female	1163 1764	39.73 60.27
71	sex_6	categorical	Gender	1520 (34.18%)		Male Female	1163 1764	39.73 60.27
72	sex_7	categorical	Gender	1520 (34.18%)		Male Female	1163 1764	39.73 60.27
73	age_1	categorical	Age groups	1520 (34.18%)		18-29 30-49 50-64 65 or more	506 1157 839 425	17.29 39.53 28.66 14.52
74	age_2	categorical	Age groups	1520 (34.18%)		18-29 30-49 50-64 65 or more	506 1157 839 425	17.29 39.53 28.66 14.52
75	age_3	categorical	Age groups	1520 (34.18%)		18-29 30-49 50-64 65 or more	506 1157 839 425	17.29 39.53 28.66 14.52
76	age_4	categorical	Age groups	1520 (34.18%)		18-29 30-49 50-64 65 or more	506 1157 839 425	17.29 39.53 28.66 14.52
77	age_5	categorical	Age groups	1520 (34.18%)		18-29 30-49 50-64 65 or more	506 1157 839 425	17.29 39.53 28.66 14.52
78	age_6	categorical	Age groups	1520 (34.18%)		18-29 30-49 50-64 65 or more	506 1157 839 425	17.29 39.53 28.66 14.52
79	age_7	categorical	Age groups	1520 (34.18%)		18-29 30-49 50-64 65 or more	506 1157 839 425	17.29 39.53 28.66 14.52
80	ideo_1	categorical	Political identification	1586 (35.66%)		Left Center Right Does not identify	583 604 409 1265	20.38 21.11 14.30 44.22
81	ideo_2	categorical	Political identification	1586 (35.66%)		Left Center Right Does not identify	583 604 409 1265	20.38 21.11 14.30 44.22
82	ideo_3	categorical	Political identification	1586 (35.66%)		Left Center Right Does not identify	583 604 409 1265	20.38 21.11 14.30 44.22
83	ideo_4	categorical	Political identification	1586 (35.66%)		Left Center Right Does not identify	583 604 409 1265	20.38 21.11 14.30 44.22
84	ideo_5	categorical	Political identification	1586 (35.66%)		Left Center Right Does not identify	583 604 409 1265	20.38 21.11 14.30 44.22
85	ideo_6	categorical	Political identification	1586 (35.66%)		Left Center Right Does not identify	583 604 409 1265	20.38 21.11 14.30 44.22
86	ideo_7	categorical	Political identification	1586 (35.66%)		Left Center Right Does not identify	583 604 409 1265	20.38 21.11 14.30 44.22

# reshape from long to wide
pacman::p_load(datasets,data.table)
df_study1_long <- data.table::melt.data.table(data.table::setDT(df_study1_wide),
              id.vars = c("idencuesta","muestra"),
              variable.name = c("ola"),
              measure = 
patterns("^ponderador_long_total_", "^segmento_", "^estrato_", "^just_pension_","^class1_","^merit_effort_","^merit_talent_", "^educ_","^educyear_","^sex_", "^age_", "^ideo_"),
              value.name = c("ponderador_long_total","segmento","estrato", "just_pension","class","merit_effort","merit_talent",  "educ","educyear","sex", "age", "ideo"),
              na.rm = F,value.factor = TRUE
              )

names(df_study1_long) #check names of long dataset

 [1] "idencuesta"            "muestra"               "ola"                  
 [4] "ponderador_long_total" "segmento"              "estrato"              
 [7] "just_pension"          "class"                 "merit_effort"         
[10] "merit_talent"          "educ"                  "educyear"             
[13] "sex"                   "age"                   "ideo"                 

dim(df_study1_long) #check dimensions of the dataframe

[1] 31129    15

# Original dataset with 7 waves
df_study2_long <- df_study1_long

# filter the dataset for the waves 1 to 4 and 6 to 7
df_study1_long <-
df_study1_long %>% 
  filter(ola %in% c(1,2,3,4,6,7)) %>% 
  mutate(ola=factor(ola,levels = 1:7,labels = 1:7))
dim(df_study1_long) #check, now is OK

[1] 26682    15

# df_study1_long <- 
# set_label(x = df_study1_long,
#           label = get_label(select(df_study1,names(df_study1_long))))
  

#______________________________________________________________________________
# obtain the idencuesta for wave 7
ids <- 
  elsoc_long_2016_2023 %>% 
  select(idencuesta,ola) %>% 
  filter(ola==7) %>% 
  sjmisc::frq(idencuesta,show.na = F) %>% as.data.frame()


# filter data by the idencuesta of t7
df_study1_long_t7 <- 
  df_study1_long %>%
  filter(idencuesta %in% ids$val)

names(df_study1_long_t7)

 [1] "idencuesta"            "muestra"               "ola"                  
 [4] "ponderador_long_total" "segmento"              "estrato"              
 [7] "just_pension"          "class"                 "merit_effort"         
[10] "merit_talent"          "educ"                  "educyear"             
[13] "sex"                   "age"                   "ideo"                 

dim(df_study1_long_t7)

[1] 16356    15

sjmisc::frq(df_study1_long_t7$ola)

x <categorical> 
# total N=16356 valid N=16356 mean=3.83 sd=2.11

Value |    N | Raw % | Valid % | Cum. %
---------------------------------------
    1 | 2726 | 16.67 |   16.67 |  16.67
    2 | 2726 | 16.67 |   16.67 |  33.33
    3 | 2726 | 16.67 |   16.67 |  50.00
    4 | 2726 | 16.67 |   16.67 |  66.67
    5 |    0 |  0.00 |    0.00 |  66.67
    6 | 2726 | 16.67 |   16.67 |  83.33
    7 | 2726 | 16.67 |   16.67 | 100.00
 <NA> |    0 |  0.00 |    <NA> |   <NA>

save(df_study1_long,file = here::here("input/data/proc/df_study1_long.RData"))
save(df_study1_long_t7,file = here::here("input/data/proc/df_study1_long_t7.RData"))
save(df_study2_long,file = here::here("input/data/proc/df_study2_long.RData"))