Data analysis

Perceptions of Inequality and Meritocracy: Their Interplay in Shaping Preferences for Market Justice in Chile (2016-2023)

Author

Researcher

Published

April 11, 2025

1 Presentation

This is the analysis code for the paper “Perceptions of Inequality and Meritocracy: Their Interplay in Shaping Preferences for Market Justice in Chile (2016-2023)”. The dataset used is df_study1_long_t7.RData.

2 Libraries

Show the code
if (! require("pacman")) install.packages("pacman")

pacman::p_load(tidyverse, 
               sjmisc, 
               sjPlot, 
               lme4, 
               here, 
               performance,
               influence.ME, 
               marginaleffects,
               MLMusingR,
               texreg, 
               ggdist,
               misty,
               kableExtra,
               ggalluvial, 
               shadowtext,
               MetBrewer,
               patchwork,
               sjlabelled)


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

3 Data

Show the code
load(file = here("input/data/proc/df_study1_long_t7.RData"))

glimpse(df_study1_long_t7)
Rows: 10,422
Columns: 29
$ idencuesta            <dbl> 1101011, 1101012, 1101021, 1101023, 1101032, 110…
$ muestra               <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, …
$ ola                   <fct> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, …
$ comuna                <chr> "Iquique", "Iquique", "Iquique", "Iquique", "Iqu…
$ comunacod             <dbl> 1101, 1101, 1101, 1101, 1101, 1101, 1101, 1101, …
$ ponderador_long_total <dbl> 0.11821742, 0.11821742, 0.05633656, 0.07703080, …
$ segmento              <dbl> 110101, 110101, 110102, 110102, 110103, 110103, …
$ estrato               <dbl> 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, …
$ just_educ             <fct> Strongly desagree, Desagree, Agree, Strongly des…
$ just_pension          <fct> Strongly desagree, Desagree, Neither agree nor d…
$ just_health           <fct> Desagree, Desagree, Agree, Strongly desagree, St…
$ mjp                   <dbl> 1.333333, 2.000000, 3.666667, 1.333333, 1.000000…
$ merit_effort          <fct> Agree, Agree, Agree, Neither agree nor desagree,…
$ merit_talent          <fct> Agree, Agree, Agree, Strongly agree, Strongly de…
$ perc_sal_gerente      <dbl> 2000000, 500000, 20000000, 6000000, 4000000, 500…
$ perc_sal_obrero       <dbl> 300000, 300000, 400000, 280000, 300000, 400000, …
$ just_sal_gerente      <dbl> 2000000, 3000000, 200000000, 1800000, 300000, 50…
$ just_sal_obrero       <dbl> 500000, 500000, 700000, 600000, 600000, 500000, …
$ perc_inequality       <dbl> 1.8971200, 0.5108256, 3.9120230, 3.0647251, 2.59…
$ just_inequality       <dbl> 1.3862944, 1.7917595, 5.6549923, 1.0986123, -0.6…
$ educ                  <fct> Less than Universitary, Less than Universitary, …
$ educyear              <dbl> 4.30, 9.80, 14.90, 9.80, 12.02, 13.90, 12.02, 7.…
$ sex                   <fct> Female, Female, Male, Male, Female, Female, Male…
$ age                   <fct> 50-64, 50-64, 50-64, 50-64, 50-64, 30-49, 30-49,…
$ ess                   <dbl> 5, 5, 6, 5, 6, 7, 4, 0, 4, 5, 4, 3, 6, 3, 5, 8, …
$ ideo                  <fct> Does not identify, Does not identify, Does not i…
$ quintil               <fct> Q1, Q4, Q5, Q5, Q4, Q4, Q2, Q2, Q2, Q5, Q4, Q2, …
$ quintil1              <fct> Q1, Q4, Q5, Q5, Q4, Q4, Q2, Q2, Q2, Q5, Q4, Q2, …
$ ing_pc                <dbl> 75000.0, 250000.0, 765000.0, 490000.0, 250000.0,…
Show the code
# Generate analytical sample

df_study1 <- df_study1_long_t7 %>%
  select(idencuesta,
         ola,
         ponderador_long_total,
         segmento,
         estrato,
         just_health,
         just_pension,
         just_educ,
         mjp, 
         perc_inequality, 
         just_inequality,
         merit_effort, 
         merit_talent, 
         educ,
         quintil1,
         sex,
         age,
         ess, 
         ideo) %>% 
  na.omit() %>% 
  mutate(ola = as.factor(ola),
         ola_num = as.numeric(ola),
         ola_2=as.numeric(ola)^2)

df_study1 <- df_study1 %>%
  group_by(idencuesta) %>%             # Agrupar por el identificador del participante
  mutate(n_participaciones = n()) %>%  # Contar el número de filas (participaciones) por participante
  ungroup()

df_study1 <- df_study1 %>% filter(n_participaciones>1)

# Corregir etiquetas

df_study1$just_health <- sjlabelled::set_label(df_study1$just_health, 
                        label = "Health distributive justice")

df_study1$just_pension <- sjlabelled::set_label(df_study1$just_pension, 
                        label = "Pension distributive justice")

df_study1$just_educ <- sjlabelled::set_label(df_study1$just_educ, 
                        label = "Education distributive justice")

df_study1$merit_effort <- sjlabelled::set_label(df_study1$merit_effort, 
                        label = "People are rewarded for their efforts")

df_study1$merit_talent <- sjlabelled::set_label(df_study1$merit_talent, 
                        label = "People are rewarded for their intelligence")

4 Analysis

4.1 Descriptives

Show the code
t1 <- df_study1 %>% 
  filter(ola == 1) %>% 
  select(just_health, just_pension, just_educ, mjp) 

print(summarytools::dfSummary(t1), method="render")
Table 1: Estadísticos descriptivos items justicia de mercado

Data Frame Summary

t1

Dimensions: 1477 x 4
Duplicates: 1407
No Variable Label Stats / Values Freqs (% of Valid) Graph Valid Missing
1 just_health [factor] Health distributive justice
1. Strongly desagree
2. Desagree
3. Neither agree nor desagre
4. Agree
5. Strongly agree
554 ( 37.5% )
713 ( 48.3% )
61 ( 4.1% )
131 ( 8.9% )
18 ( 1.2% )
1477 (100.0%) 0 (0.0%)
2 just_pension [factor] Pension distributive justice
1. Strongly desagree
2. Desagree
3. Neither agree nor desagre
4. Agree
5. Strongly agree
425 ( 28.8% )
704 ( 47.7% )
106 ( 7.2% )
219 ( 14.8% )
23 ( 1.6% )
1477 (100.0%) 0 (0.0%)
3 just_educ [factor] Education distributive justice
1. Strongly desagree
2. Desagree
3. Neither agree nor desagre
4. Agree
5. Strongly agree
519 ( 35.1% )
764 ( 51.7% )
71 ( 4.8% )
112 ( 7.6% )
11 ( 0.7% )
1477 (100.0%) 0 (0.0%)
4 mjp [numeric] Market justice preferences
Mean (sd) : 2 (0.8)
min ≤ med ≤ max:
1 ≤ 2 ≤ 5
IQR (CV) : 1 (0.4)
12 distinct values 1477 (100.0%) 0 (0.0%)

Generated by summarytools 1.0.1 (R version 4.2.2)
2025-04-11

Show the code
t2 <- df_study1 %>% 
  filter(ola == 1) %>% 
  select(perc_inequality, just_inequality, merit_effort, merit_talent) 

print(summarytools::dfSummary(t2), method="render")
Table 2: Estadísticos descriptivos variables independientes

Data Frame Summary

t2

Dimensions: 1477 x 4
Duplicates: 84
No Variable Label Stats / Values Freqs (% of Valid) Graph Valid Missing
1 perc_inequality [numeric] Inequality gap perception
Mean (sd) : 3.7 (1.1)
min ≤ med ≤ max:
0.4 ≤ 3.7 ≤ 6.9
IQR (CV) : 1.6 (0.3)
294 distinct values 1477 (100.0%) 0 (0.0%)
2 just_inequality [numeric] Inequality gap justification
Mean (sd) : 2.1 (1.4)
min ≤ med ≤ max:
-5.5 ≤ 2.1 ≤ 9.9
IQR (CV) : 1.7 (0.6)
215 distinct values 1477 (100.0%) 0 (0.0%)
3 merit_effort [factor] People are rewarded for their efforts
1. Strongly desagree
2. Desagree
3. Neither agree nor desagre
4. Agree
5. Strongly agree
168 ( 11.4% )
679 ( 46.0% )
258 ( 17.5% )
324 ( 21.9% )
48 ( 3.2% )
1477 (100.0%) 0 (0.0%)
4 merit_talent [factor] People are rewarded for their intelligence
1. Strongly desagree
2. Desagree
3. Neither agree nor desagre
4. Agree
5. Strongly agree
134 ( 9.1% )
604 ( 40.9% )
289 ( 19.6% )
395 ( 26.7% )
55 ( 3.7% )
1477 (100.0%) 0 (0.0%)

Generated by summarytools 1.0.1 (R version 4.2.2)
2025-04-11

Show the code
datos.health <- df_study1 %>% 
  mutate(just_health = factor(just_health, 
                              levels = c("Strongly agree",
                                         "Agree",
                                         "Neither agree nor desagree",
                                         "Desagree",
                                         "Strongly desagree"))) %>% 
  group_by(idencuesta, ola) %>% 
  count(just_health) %>% 
  group_by(ola) %>% 
  mutate(porcentaje=n/sum(n)) %>% 
  ungroup() %>% 
  na.omit() %>% 
  mutate(wave = case_when(ola == 1 ~ "2016",
                          ola == 2 ~ "2017",
                          ola == 3 ~ "2018",
                          ola == 4 ~ "2019",
                          ola == 6 ~ "2022",
                          ola == 7 ~ "2023"),
         wave = factor(wave, levels = c("2016",
                                      "2017",
                                      "2018",
                                      "2019",
                                      "2022",
                                      "2023")))



etiquetas.health <- df_study1 %>%
  mutate(just_health = factor(just_health, 
                              levels = c("Strongly agree",
                                         "Agree",
                                         "Neither agree nor desagree",
                                         "Desagree",
                                         "Strongly desagree"))) %>% 
  group_by(ola, just_health) %>%
  summarise(count = n(), .groups = "drop") %>%
  group_by(ola) %>%
  mutate(porcentaje = count / sum(count)) %>% 
  na.omit() %>% 
  mutate(idencuesta = 1,
         wave = case_when(ola == 1 ~ "2016",
                          ola == 2 ~ "2017",
                          ola == 3 ~ "2018",
                          ola == 4 ~ "2019",
                          ola == 6 ~ "2022",
                          ola == 7 ~ "2023"),
         wave = factor(wave, levels = c("2016",
                                        "2017",
                                        "2018",
                                        "2019",
                                        "2022",
                                        "2023")))




p1 <- datos.health %>% 
  ggplot(aes(x = wave, fill = just_health, stratum = just_health,
             alluvium = idencuesta, y = porcentaje)) +
  ggalluvial::geom_flow(alpha = .4) + 
  ggalluvial::geom_stratum(linetype = 0) +
  scale_y_continuous(labels = scales::percent) + 
  scale_fill_manual(values =  c("#0571B0","#92C5DE","#b3b3b3ff","#F4A582","#CA0020")) +
  geom_shadowtext(data = etiquetas.health,
                  aes(label = ifelse(porcentaje > 0 , scales::percent(porcentaje, accuracy = .1),"")),
                  position = position_stack(vjust = .5),
                  show.legend = FALSE,
                  size = 3,
                  color = rep('white'),
                  bg.colour='grey30')+
  labs(y = "%",
       x = NULL,
       fill = NULL,
       title = "a. Health") +
  theme_ggdist() +
  theme(legend.position = "none") 
  


datos.pension <- df_study1 %>% 
   mutate(just_pension = factor(just_pension, 
                              levels = c("Strongly agree",
                                         "Agree",
                                         "Neither agree nor desagree",
                                         "Desagree",
                                         "Strongly desagree"))) %>% 
  group_by(idencuesta, ola) %>% 
  count(just_pension) %>% 
  group_by(ola) %>% 
  mutate(porcentaje=n/sum(n)) %>% 
  ungroup() %>% 
  na.omit() %>% 
  mutate(wave = case_when(ola == 1 ~ "2016",
                          ola == 2 ~ "2017",
                          ola == 3 ~ "2018",
                          ola == 4 ~ "2019",
                          ola == 6 ~ "2022",
                          ola == 7 ~ "2023"),
         wave = factor(wave, levels = c("2016",
                                        "2017",
                                        "2018",
                                        "2019",
                                        "2022",
                                        "2023")))



etiquetas.pension <- df_study1 %>%
  mutate(just_pension = factor(just_pension, 
                              levels = c("Strongly agree",
                                         "Agree",
                                         "Neither agree nor desagree",
                                         "Desagree",
                                         "Strongly desagree"))) %>% 
  group_by(ola, just_pension) %>%
  summarise(count = n(), .groups = "drop") %>%
  group_by(ola) %>%
  mutate(porcentaje = count / sum(count)) %>% 
  na.omit() %>% 
  mutate(idencuesta = 1,
         wave = case_when(ola == 1 ~ "2016",
                          ola == 2 ~ "2017",
                          ola == 3 ~ "2018",
                          ola == 4 ~ "2019",
                          ola == 6 ~ "2022",
                          ola == 7 ~ "2023"),
         wave = factor(wave, levels = c("2016",
                                        "2017",
                                        "2018",
                                        "2019",
                                        "2022",
                                        "2023")))

p2 <- datos.pension %>% 
  ggplot(aes(x = wave, fill = just_pension, stratum = just_pension,
             alluvium = idencuesta, y = porcentaje)) +
  ggalluvial::geom_flow(alpha = .4) + 
  ggalluvial::geom_stratum(linetype = 0) +
  scale_y_continuous(labels = scales::percent) + 
  scale_fill_manual(values =  c("#0571B0","#92C5DE","#b3b3b3ff","#F4A582","#CA0020")) +
  geom_shadowtext(data = etiquetas.pension,
                  aes(label = ifelse(porcentaje > 0 , scales::percent(porcentaje, accuracy = .1),"")),
                  position = position_stack(vjust = .5),
                  show.legend = FALSE,
                  size = 3,
                  color = rep('white'),
                  bg.colour='grey30')+
  labs(y = "%",
       x = NULL,
       fill = NULL,
       title = "b. Pension") +
  theme_ggdist() +
  theme(legend.position = "none") 


datos.educ <- df_study1 %>% 
  mutate(just_educ = factor(just_educ, 
                              levels = c("Strongly agree",
                                         "Agree",
                                         "Neither agree nor desagree",
                                         "Desagree",
                                         "Strongly desagree"))) %>%
  group_by(idencuesta, ola) %>% 
  count(just_educ) %>% 
  group_by(ola) %>% 
  mutate(porcentaje=n/sum(n)) %>% 
  ungroup() %>% 
  na.omit() %>% 
  mutate(wave = case_when(ola == 1 ~ "2016",
                          ola == 2 ~ "2017",
                          ola == 3 ~ "2018",
                          ola == 4 ~ "2019",
                          ola == 6 ~ "2022",
                          ola == 7 ~ "2023"),
         wave = factor(wave, levels = c("2016",
                                        "2017",
                                        "2018",
                                        "2019",
                                        "2022",
                                        "2023")))



etiquetas.educ <- df_study1 %>%
    mutate(just_educ = factor(just_educ, 
                              levels = c("Strongly agree",
                                         "Agree",
                                         "Neither agree nor desagree",
                                         "Desagree",
                                         "Strongly desagree"))) %>%
  group_by(ola, just_educ) %>%
  summarise(count = n(), .groups = "drop") %>%
  group_by(ola) %>%
  mutate(porcentaje = count / sum(count)) %>% 
  na.omit() %>% 
  mutate(idencuesta = 1,
         wave = case_when(ola == 1 ~ "2016",
                          ola == 2 ~ "2017",
                          ola == 3 ~ "2018",
                          ola == 4 ~ "2019",
                          ola == 6 ~ "2022",
                          ola == 7 ~ "2023"),
         wave = factor(wave, levels = c("2016",
                                        "2017",
                                        "2018",
                                        "2019",
                                        "2022",
                                        "2023")))

p3 <- datos.educ %>% 
  ggplot(aes(x = wave, fill = just_educ, stratum = just_educ,
             alluvium = idencuesta, y = porcentaje)) +
  ggalluvial::geom_flow(alpha = .4) + 
  ggalluvial::geom_stratum(linetype = 0) +
  scale_y_continuous(labels = scales::percent) + 
  scale_fill_manual(values =  c("#0571B0","#92C5DE","#b3b3b3ff","#F4A582","#CA0020")) +
  geom_shadowtext(data = etiquetas.educ,
                  aes(label = ifelse(porcentaje > 0 , scales::percent(porcentaje, accuracy = .1),"")),
                  position = position_stack(vjust = .5),
                  show.legend = FALSE,
                  size = 3,
                  color = rep('white'),
                  bg.colour='grey30')+
  labs(y = "%",
       x = NULL,
       fill = NULL,
       title = "c. Education",
       caption = "Source: own elaboration with pooled data from ELSOC 2016-2023 (n = 8460)") +
  theme_ggdist() +
  theme(legend.position = "bottom") 

(p1 / p2 / p3)
Figure 1: Change in the justification of inequality in health, pensions and education over time (2016-2023)
Show the code
library(srvyr)

df_pond <- df_study1 %>% 
  mutate(ola = case_when(ola == 1 ~ "2016",
                         ola == 2 ~ "2017",
                         ola == 3 ~ "2018",
                         ola == 4 ~ "2019",
                         ola == 6 ~ "2022",
                         ola == 7 ~ "2023"),
         ola = factor(ola, levels = c("2016",
                                        "2017",
                                        "2018",
                                        "2019",
                                        "2022",
                                        "2023"))) %>% 
  as_survey_design(.data = .,
                   ids = segmento, 
                   strata = estrato, 
                   weights = ponderador_long_total)

df_pond %>% 
  select(ola, mjp, perc_inequality, just_inequality, merit_effort, merit_talent) %>% 
  mutate_at(.vars = 2:6, .funs = ~ as.numeric(.)) %>% 
  group_by(ola) %>% 
  summarise_all(~survey_mean(., vartype = "ci")) %>% 
  pivot_longer(cols = -ola,
               names_to = "temp",
               values_to = "valor") %>%
  mutate(
    ci = case_when(
      str_ends(temp, "_low") ~ "ic_low",
      str_ends(temp, "_upp") ~ "ic_upp",
      TRUE                   ~ "mean"
    ),
    variable = str_remove(temp, "_low|_upp")
  ) %>%
  select(ola, variable, ci, valor) %>%
  pivot_wider(
    names_from  = ci,
    values_from = valor
  ) %>% 
  mutate(variable = case_when(variable == "mjp" ~ "Market justice preferences",
                              variable == "perc_inequality"  ~ "Inequality perception",
                              variable == "just_inequality"  ~ "Inequality justification",
                              variable == "merit_effort"     ~ "Merit: Effort",
                              variable == "merit_talent"     ~ "Merit: Talent"),
         variable = factor(variable, levels = c("Market justice preferences",
                                      "Inequality perception",
                                      "Inequality justification",
                                      "Merit: Effort",
                                      "Merit: Talent")
                              )) %>% 
  ggplot(aes(x = ola, y = mean, group = variable)) +
  geom_point(aes(shape=variable, color=variable), size = 3.5) +
  geom_line(aes(color = variable), linewidth = 0.8) +
  geom_errorbar(aes(ymin = ic_low, ymax = ic_upp, color = variable),
                width = 0.1) +
  scale_color_manual(values = c("#023858", # azul oscuro
                                "#0571B0", # azul intermedio
                                "#b3b3b3", # gris
                                "#F4A582", # salmón
                                "#B2182B"  # rojo profundo
                                         )) +
  scale_shape_manual(values=c(15, 16, 17, 18, 4)) +
  labs(y = "Mean",
       x = NULL,
       color = NULL,
       shape = NULL,
       caption = "Source: own elaboration with pooled data from ELSOC 2016-2023 (n = 8460)") +
  theme_ggdist() +
  theme(legend.position = "top",
        text = element_text(size = 12))

4.2 Correlations

Show the code
M <- df_study1 %>%
  filter(ola == 7) %>% 
  select(mjp, perc_inequality, just_inequality, merit_effort, merit_talent) %>% 
  mutate_all(.funs = ~ as.numeric(.)) %>% 
  cor(use = "pairwise.complete.obs", method = "pearson")

diag(M) <- NA

rownames(M) <- c("A. Marjet justice preferences",
                 "B. Inequality gap perception",
                 "C. Inequality gap justification",
                 "D. People are rewarded for their efforts ",
                 "E. People are rewarded for their intelligence")

#set Column names of the matrix
colnames(M) <-c("(A)", "(B)","(C)","(D)","(E)")

#Plot the matrix using corrplot
corrplot::corrplot(M,
                   method = "color",
                   addCoef.col = "black",
                   type = "upper",
                   tl.col = "black",
                   col = colorRampPalette(c("#E16462", "white", "#0D0887"))(12),
                   bg = "white",
                   na.label = "-")

4.3 Longitudinal multilevel models

Show the code
df_study1$merit_effort <- as_numeric(df_study1$merit_effort)
df_study1$merit_talent <- as_numeric(df_study1$merit_talent)

df_study1 <- df_study1 %>% 
  mutate(ola = case_when(ola == 1 ~ "2016",
                         ola == 2 ~ "2017",
                         ola == 3 ~ "2018",
                         ola == 4 ~ "2019",
                         ola == 6 ~ "2022",
                         ola == 7 ~ "2023"),
         ola = factor(ola, levels = c("2016",
                                        "2017",
                                        "2018",
                                        "2019",
                                        "2022",
                                        "2023")))


df_study1 <- df_study1 %>% 
  group_by(idencuesta) %>% 
  mutate(perc_inequality_mean = mean(perc_inequality, na.rm = T),
         perc_inequality_cwc = perc_inequality - perc_inequality_mean,
         just_inequality_mean = mean(just_inequality, na.rm = T),
         just_inequality_cwc = just_inequality - just_inequality_mean,
         merit_effort_mean = mean(merit_effort, na.rm = T),
         merit_effort_cwc = merit_effort - merit_effort_mean,
         merit_talent_mean = mean(merit_talent, na.rm = T),
         merit_talent_cwc = merit_talent - merit_talent_mean,
         ) %>% 
  ungroup()
Show the code
m0 <- lmer(mjp ~ 1 + (1 | idencuesta), 
                data = df_study1)

performance::icc(m0, by_group = T)
# ICC by Group

Group      |   ICC
------------------
idencuesta | 0.231
Show the code
m1 <- lmer(mjp ~ 1 + ola + (1 | idencuesta),
                data = df_study1)

m1.1 <- lmer(mjp ~ 1 + ola_num + ola_2 + (1 | idencuesta),
                data = df_study1)

m2 <- lmer(mjp ~ 1 + ola_num + ola_2 + (1 + ola_num | idencuesta),
                data = df_study1)

anova(m1.1, m2)
Data: df_study1
Models:
m1.1: mjp ~ 1 + ola_num + ola_2 + (1 | idencuesta)
m2: mjp ~ 1 + ola_num + ola_2 + (1 + ola_num | idencuesta)
     npar   AIC   BIC logLik deviance  Chisq Df            Pr(>Chisq)    
m1.1    5 20299 20334 -10145    20289                                    
m2      7 20234 20283 -10110    20220 69.409  2 0.0000000000000008471 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Show the code
m3 <- lmer(mjp ~ 1 + ola_num + ola_2 + perc_inequality + (1 + ola_num | idencuesta), 
           data = df_study1)

m4 <- lmer(mjp ~ 1 + ola_num + ola_2 + perc_inequality + just_inequality + (1 + ola_num | idencuesta),
                data = df_study1)

m5 <- lmer(mjp ~ 1 + ola_num + ola_2 + perc_inequality + just_inequality + merit_effort + (1 + ola_num | idencuesta),
                data = df_study1)

m6 <- lmer(mjp ~ 1 + ola_num + ola_2 + perc_inequality + just_inequality + merit_effort + merit_talent + (1 + ola_num | idencuesta),
                data = df_study1)


m7 <- lmer(mjp ~ 1 + ola_num + ola_2 + perc_inequality + just_inequality + merit_effort + merit_talent + perc_inequality_mean + just_inequality_mean + merit_effort_mean + merit_talent_mean + (1 + ola_num | idencuesta),
                data = df_study1)

m8 <- lmer(mjp ~ 1 + ola_num + ola_2 + perc_inequality + just_inequality + merit_effort + merit_talent + perc_inequality_mean + just_inequality_mean + merit_effort_mean + merit_talent_mean + educ + quintil1 + ess + ideo + sex + age + (1 + ola_num | idencuesta),
                data = df_study1)

m9 <- lmer(mjp ~ 1 + ola_num + ola_2 + perc_inequality*ola_num + just_inequality + merit_effort + merit_talent + perc_inequality_mean + just_inequality_mean + merit_effort_mean + merit_talent_mean + educ + quintil1 + ess + ideo + sex + age + (1 + perc_inequality | idencuesta),
                data = df_study1)

m10 <- lmer(mjp ~ 1 + ola_num + ola_2 + perc_inequality + just_inequality*ola_num + merit_effort + merit_talent + perc_inequality_mean + just_inequality_mean + merit_effort_mean + merit_talent_mean + educ + quintil1 + ess + ideo + sex + age + (1 + just_inequality| idencuesta),
                data = df_study1)

m11 <- lmer(mjp ~ 1 + ola_num + ola_2 + perc_inequality + just_inequality + merit_effort*ola_num + merit_talent + perc_inequality_mean + just_inequality_mean + merit_effort_mean + merit_talent_mean + educ + quintil1 + ess + ideo + sex + age + (1 + merit_effort | idencuesta),
                data = df_study1)

m12 <- lmer(mjp ~ 1 + ola_num + ola_2 + perc_inequality + just_inequality + merit_effort + merit_talent*ola_num + perc_inequality_mean + just_inequality_mean + merit_effort_mean + merit_talent_mean + educ + quintil1 + ess + ideo + sex + age + (1 + merit_talent | idencuesta),
                data = df_study1)
Show the code
ccoef <- list(
  "(Intercept)" = "Intercept",
  "ola2017" = "Wave 2017",
  "ola2018" = "Wave 2018",
  "ola2019" = "Wave 2019",
  "ola2022" = "Wave 2022",
  "ola2023" = "Wave 2023",
  ola_num = "Wave",
  ola_2 = "Wave^2",
  perc_inequality = "Perception inequality (WE)",
  just_inequality = "Justification inequality (WE)",
  merit_effort = "Merit: Effort (WE)",
  merit_talent = "Merit: Talent (WE)",
  perc_inequality_mean = "Perception inequality (BE)",
  just_inequality_mean = "Justification inequality (BE)",
  merit_effort_mean = "Merit: Effort (BE)",
  merit_talent_mean = "Merit: Talent (BE)",
  educUniversitary = "Universitary (Ref.= Less than universitary)",
  quintil1Q2 = "Quintile 2",
  quintil1Q3 = "Quintile 3",
  quintil1Q4 = "Quintile 4",
  quintil1Q5 = "Quintile 5",
  quintil1QNA = "Quintile NA",
  ess = "Subjective social status",
  ideoCenter = "Center",
  ideoRight = "Right",
  "ideoDoes not identify" = "Does not identify",
  sexFemale = "Female (Ref.= Male)",
  "age30-49" = "30-49",
  "age50-64" = "50-64",
  "age65 or more" = "65 or more"
  )

texreg::htmlreg(list(m1, m2, m3, m4, m5, m6, m7, m8),
               custom.model.names = c(paste0("Model ", seq(1:8))),
               caption = NULL,
               stars = c(0.05, 0.01, 0.001),
               custom.coef.map = ccoef,
               digits = 3,
               groups = list("Wave (Ref.= 2016)" = 2:6, "Income (Ref.= Quintile 1)" = 18:22, "Political identification (Ref.= Left)" = 24:26, "Age (Ref. = 18-29)" = 28:30),
               custom.note = "Note: Cells contain regression coefficients with standard errors in parentheses. %stars.",
               threeparttable = T,
               leading.zero = T,
               float.pos = "h!",
               use.packages = F,
               booktabs = T,
               scalebox = 1)
Table 3: Longitudinal multilevel models for market justice preferences
  Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8
Intercept 1.964*** 1.966*** 2.103*** 2.059*** 1.752*** 1.721*** 1.541*** 1.574***
  (0.022) (0.033) (0.046) (0.045) (0.052) (0.053) (0.095) (0.112)
Wave (Ref.= 2016)                
                 
     Wave 2017 -0.162***              
  (0.028)              
     Wave 2018 -0.015              
  (0.027)              
     Wave 2019 -0.041              
  (0.027)              
     Wave 2022 0.289***              
  (0.028)              
     Wave 2023 0.286***              
  (0.027)              
Wave   -0.069*** -0.072*** -0.057** -0.057** -0.058** -0.064*** -0.064***
    (0.018) (0.018) (0.018) (0.018) (0.018) (0.018) (0.018)
Wave^2   0.017*** 0.017*** 0.015*** 0.015*** 0.015*** 0.016*** 0.016***
    (0.002) (0.002) (0.002) (0.002) (0.002) (0.002) (0.002)
Perception inequality (WE)     -0.037*** -0.086*** -0.075*** -0.074*** -0.056*** -0.056***
      (0.009) (0.010) (0.010) (0.010) (0.011) (0.011)
Justification inequality (WE)       0.100*** 0.095*** 0.094*** 0.055*** 0.055***
        (0.009) (0.009) (0.009) (0.010) (0.010)
Merit: Effort (WE)         0.107*** 0.088*** 0.071*** 0.071***
          (0.009) (0.011) (0.012) (0.012)
Merit: Talent (WE)           0.029* 0.025* 0.025*
            (0.011) (0.012) (0.012)
Perception inequality (BE)             -0.077*** -0.078***
              (0.023) (0.023)
Justification inequality (BE)             0.156*** 0.131***
              (0.021) (0.021)
Merit: Effort (BE)             0.092** 0.088**
              (0.033) (0.033)
Merit: Talent (BE)             -0.006 -0.016
              (0.033) (0.033)
Universitary (Ref.= Less than universitary)               -0.006
                (0.035)
Income (Ref.= Quintile 1)                
                 
     Quintile 2               0.000
                (0.038)
     Quintile 3               0.034
                (0.038)
     Quintile 4               0.093*
                (0.039)
     Quintile 5               0.147***
                (0.042)
     Quintile NA               0.172**
                (0.066)
Subjective social status               -0.007
                (0.008)
Political identification (Ref.= Left)                
                 
     Center               0.058
                (0.037)
     Right               0.246***
                (0.042)
     Does not identify               0.092**
                (0.032)
Female (Ref.= Male)               -0.080**
                (0.026)
Age (Ref. = 18-29)                
                 
     30-49               -0.010
                (0.037)
     50-64               0.002
                (0.038)
     65 or more               0.004
                (0.046)
AIC 20273.132 20260.158 20251.351 20140.808 20002.562 20005.214 19957.283 19986.833
BIC 20329.477 20309.460 20307.696 20204.196 20072.993 20082.688 20062.929 20191.083
Log Likelihood -10128.566 -10123.079 -10117.676 -10061.404 -9991.281 -9991.607 -9963.641 -9964.417
Num. obs. 8460 8460 8460 8460 8460 8460 8460 8460
Num. groups: idencuesta 1681 1681 1681 1681 1681 1681 1681 1681
Var: idencuesta (Intercept) 0.174 0.214 0.210 0.204 0.179 0.178 0.180 0.172
Var: Residual 0.528 0.494 0.494 0.493 0.490 0.490 0.488 0.488
Var: idencuesta ola_num   0.007 0.007 0.007 0.006 0.006 0.006 0.006
Cov: idencuesta (Intercept) ola_num   -0.018 -0.017 -0.018 -0.015 -0.015 -0.016 -0.016
Note: Cells contain regression coefficients with standard errors in parentheses. ***p < 0.001; **p < 0.01; *p < 0.05.
Show the code
ccoef <- list(
  "(Intercept)" = "Intercept",
  "ola_num:perc_inequality" = "Perception inequality (WE) x Wave",
  "ola_num:just_inequality" = "Justification inequality (WE) x Wave",
  "ola_num:merit_effort" = "Merit: Effort (WE) x Wave",
  "ola_num:merit_talent" = "Merit: Talent (WE) x Wave"
  )

texreg::htmlreg(list(m9,m10,m11,m12),
               custom.model.names = c(paste0("Model ", seq(9:12))),
               caption = NULL,
               stars = c(0.05, 0.01, 0.001),
               custom.coef.map = ccoef,
               digits = 3,
               custom.note = "Note: Cells contain regression coefficients with standard errors in parentheses. %stars.",
               threeparttable = T,
               leading.zero = T,
               float.pos = "h!",
               use.packages = F,
               booktabs = T,
               scalebox = 1)
Table 4: Time interactions within effects for market justice preferences
  Model 1 Model 2 Model 3 Model 4
Intercept 1.611*** 1.716*** 1.461*** 1.474***
  (0.124) (0.112) (0.118) (0.119)
Perception inequality (WE) x Wave 0.001      
  (0.004)      
Justification inequality (WE) x Wave   0.015***    
    (0.003)    
Merit: Effort (WE) x Wave     -0.012**  
      (0.004)  
Merit: Talent (WE) x Wave       -0.010*
        (0.004)
AIC 20053.598 19988.291 20003.746 19999.606
BIC 20264.891 20199.584 20215.039 20210.899
Log Likelihood -9996.799 -9964.145 -9971.873 -9969.803
Num. obs. 8460 8460 8460 8460
Num. groups: idencuesta 1681 1681 1681 1681
Var: idencuesta (Intercept) 0.152 0.078 0.219 0.214
Var: idencuesta perc_inequality 0.002      
Cov: idencuesta (Intercept) perc_inequality -0.007      
Var: Residual 0.520 0.512 0.503 0.502
Var: idencuesta just_inequality   0.012    
Cov: idencuesta (Intercept) just_inequality   -0.000    
Var: idencuesta merit_effort     0.024  
Cov: idencuesta (Intercept) merit_effort     -0.049  
Var: idencuesta merit_talent       0.023
Cov: idencuesta (Intercept) merit_talent       -0.048
Note: Cells contain regression coefficients with standard errors in parentheses. ***p < 0.001; **p < 0.01; *p < 0.05.