Data analysis

Changes in the Justification of Pension Inequality in Chile (2016–2023) and its Relationship to Social Class and Beliefs in Meritocracy

Author

Andreas Laffert, Research Assistant

Published

October 16, 2025

1 Presentation

This is the analysis 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 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, 
               srvyr,
               ordinal,
               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: 16,356
Columns: 17
$ 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, …
$ 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_pension          <fct> Strongly disagree, Disagree, Neither agree nor d…
$ egp                   <fct> NA, NA, Working class (V+VI+VII), Intermediate c…
$ merit_effort          <fct> Agree, Agree, Agree, Neither agree nor disagree,…
$ merit_talent          <fct> Agree, Agree, Agree, Strongly agree, Strongly di…
$ 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                   <dbl> 64, 60, 51, 62, 54, 32, 38, 69, 56, 65, 35, 26, …
$ aget                  <fct> 50-64, 50-64, 50-64, 50-64, 50-64, 30-49, 30-49,…
$ ideo                  <fct> Does not identify, Does not identify, Does not i…
$ n_participaciones     <int> 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, …
Show the code
# Generate analytical sample

df_study1 <- df_study1_long_t7 %>%
  select(-muestra) %>% 
  na.omit() %>% 
  mutate(ola = case_when(ola == 1 ~ 1,
                         ola == 2 ~ 2, 
                         ola == 3 ~ 3,
                         ola == 4 ~ 4,
                         ola == 6 ~ 5,
                         ola == 7 ~ 6)) %>% 
  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$just_pension <- sjlabelled::set_label(df_study1$just_pension, 
                        label = "Pension distributive justice")

df_study1$egp <- sjlabelled::set_label(df_study1$egp, 
                        label = "Social class")

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
datos.pension <- df_study1 %>% 
   mutate(just_pension = factor(just_pension, 
                              levels = c("Strongly agree",
                                         "Agree",
                                         "Neither agree nor disagree",
                                         "Disagree",
                                         "Strongly disagree"))) %>% 
  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 == 5 ~ "2022",
                          ola == 6 ~ "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 disagree",
                                         "Disagree",
                                         "Strongly disagree"))) %>% 
  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 == 5 ~ "2022",
                          ola == 6 ~ "2023"),
         wave = factor(wave, levels = c("2016",
                                        "2017",
                                        "2018",
                                        "2019",
                                        "2022",
                                        "2023")))


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 = NULL) +
  theme_ggdist() +
  theme(legend.position = "bottom")
Figure 1: Changes in the justification of inequality in pensions over time (2016-2023)
Show the code
t2 <- df_study1 %>% 
  filter(ola == last(ola)) %>% 
  select(just_pension, egp, merit_effort, merit_talent) 

print(summarytools::dfSummary(t2), method="render")
Table 1: Estadísticos descriptivos variables independientes

Data Frame Summary

t2

Dimensions: 1026 x 4
Duplicates: 855
No Variable Label Stats / Values Freqs (% of Valid) Graph Valid Missing
1 just_pension [factor] Pension distributive justice
1. Strongly disagree
2. Disagree
3. Neither agree nor disagre
4. Agree
5. Strongly agree
181 ( 17.6% )
403 ( 39.3% )
140 ( 13.6% )
266 ( 25.9% )
36 ( 3.5% )
1026 (100.0%) 0 (0.0%)
2 egp [factor] Social class
1. Service class (I+II)
2. Intermediate class (III+I
3. Working class (V+VI+VII)
146 ( 14.2% )
459 ( 44.7% )
421 ( 41.0% )
1026 (100.0%) 0 (0.0%)
3 merit_effort [factor] People are rewarded for their efforts
1. Strongly disagree
2. Disagree
3. Neither agree nor disagre
4. Agree
5. Strongly agree
116 ( 11.3% )
473 ( 46.1% )
230 ( 22.4% )
190 ( 18.5% )
17 ( 1.7% )
1026 (100.0%) 0 (0.0%)
4 merit_talent [factor] People are rewarded for their intelligence
1. Strongly disagree
2. Disagree
3. Neither agree nor disagre
4. Agree
5. Strongly agree
97 ( 9.5% )
424 ( 41.3% )
256 ( 25.0% )
234 ( 22.8% )
15 ( 1.5% )
1026 (100.0%) 0 (0.0%)

Generated by summarytools 1.1.4 (R version 4.2.3)
2025-10-16

Figure 2: Changes in the standarized mean of market justice preferences in pensions and meritocracy by social class (2016-2023)

4.2 Longitudinal multilevel models

4.3 ICC

Show the code
m0 <- clmm(just_pension ~ 1 + (1 | idencuesta), 
           link = "logit",
  Hess = TRUE, # calcula explícitamente la matriz varianza-covarianza de estimadores
                data = df_study1)

performance::icc(m0, by_group = T) # 0.23 es between, 0.77 within
# ICC by Group

Group      |   ICC
------------------
idencuesta | 0.230

4.4 Time effects

Show the code
#m1.1 <- clmm(just_pension ~ 1 + ola + (1 | idencuesta),
#                link = "logit",
#  Hess = TRUE,
#  data = df_study1)
#
#m1.2 <- clmm(just_pension ~ 1 + ola_num + (1 | idencuesta),
#                link = "logit",
#  Hess = TRUE, data = df_study1)
#
#m1.3 <- clmm(just_pension ~ 1 + ola_num + ola_2 + (1| idencuesta),
#                link = "logit",
#  Hess = TRUE, data = df_study1)
#
#m1.4 <- clmm(just_pension ~ 1 + ola_num + (1 + ola_num | idencuesta),
#                link = "logit",
#  Hess = TRUE, data = df_study1)
#
#save(m1.1,m1.2,m1.3,m1.4, file = here("output/time_effects.RData"))

load(file = here("output/time_effects.RData"))

ccoef <- list(
  "Strongly disagree|Disagree" = "Strongly disagree|Disagree",
  "Disagree|Neither agree nor disagree" = "Disagree|Neither agree nor disagree", 
  "Neither agree nor disagree|Agree" = "Neither agree nor disagree|Agree", 
  "Agree|Strongly agree" = "Agree|Strongly agree", 
  "ola2017" = "Wave 2017",
  "ola2018" = "Wave 2018",
  "ola2019" = "Wave 2019",
  "ola2022" = "Wave 2022",
  "ola2023" = "Wave 2023",
  ola_num = "Wave",
  ola_2 = "Wave^2")

texreg::htmlreg(list(m1.1,m1.2,m1.3,m1.4),
               caption.above = T,
               caption = NULL,
               stars = c(0.05, 0.01, 0.001),
               custom.coef.map = ccoef,
               digits = 3,
               groups = list("Wave (Ref.= 2016)" = 5:9),
               custom.note = "Note: Cells contain regression coefficients with standard errors in parentheses. %stars.",
               leading.zero = T,
               use.packages = F,
               booktabs = F,
               scalebox = 0.80,
               include.loglik = FALSE,
               include.aic = FALSE,
               center = T)
  Model 1 Model 2 Model 3 Model 4
Strongly disagree|Disagree -1.104*** -0.560*** -1.115*** -0.575***
  (0.072) (0.067) (0.115) (0.069)
Disagree|Neither agree nor disagree 1.254*** 1.764*** 1.221*** 1.791***
  (0.072) (0.072) (0.115) (0.075)
Neither agree nor disagree|Agree 1.904*** 2.406*** 1.867*** 2.447***
  (0.076) (0.076) (0.117) (0.080)
Agree|Strongly agree 4.521*** 5.003*** 4.473*** 5.088***
  (0.114) (0.115) (0.145) (0.122)
Wave (Ref.= 2016)        
         
     Wave 2017 -0.370***      
  (0.087)      
     Wave 2018 0.028      
  (0.087)      
     Wave 2019 0.051      
  (0.086)      
     Wave 2022 0.850***      
  (0.088)      
     Wave 2023 0.830***      
  (0.085)      
Wave   0.217*** -0.198** 0.220***
    (0.015) (0.071) (0.016)
Wave^2     0.059***  
      (0.010)  
BIC 14781.285 14838.135 14811.280 14848.231
Num. obs. 5755 5755 5755 5755
Groups (idencuesta) 1027 1027 1027 1027
Variance: idencuesta: (Intercept) 1.109 1.071 1.081 1.238
Variance: idencuesta: ola_num       0.025
Note: Cells contain regression coefficients with standard errors in parentheses. ***p < 0.001; **p < 0.01; *p < 0.05.

4.4.1 Anova

Show the code
anova(m1.2, m1.4) # quedarse con tiempo continua y con pendiente aleatoria
Likelihood ratio tests of cumulative link models:
 
     formula:                                                link: threshold:
m1.2 just_pension ~ 1 + ola_num + (1 | idencuesta)           logit flexible  
m1.4 just_pension ~ 1 + ola_num + (1 + ola_num | idencuesta) logit flexible  

     no.par   AIC  logLik LR.stat df Pr(>Chisq)  
m1.2      6 14798 -7393.1                        
m1.4      8 14795 -7389.5  7.2199  2    0.02705 *
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

4.5 WE and BE main effects

Show the code
## WE and BE main effects
#m3 <- clmm(just_pension ~ 1 + ola_num + egp + (1 + ola_num | idencuesta), 
#           link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#m4 <- clmm(just_pension ~ 1 + ola_num + egp +  merit_effort_cwc + #merit_talent_cwc + (1 + ola_num | idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#m5 <- clmm(just_pension ~ 1 + ola_num + egp +  merit_effort_cwc + #merit_talent_cwc + merit_effort_mean + merit_talent_mean + (1 + ola_num| #idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#m6 <- clmm(just_pension ~ 1 + ola_num + egp + merit_effort_cwc + merit_talent_cwc #+ merit_effort_mean + merit_talent_mean + educ + ideo + sex + age + (1 + ola_num| #idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#save(m3,m4,m5,m6, file = here("output/main_effects.RData"))

load(file = here("output/main_effects.RData"))

ccoef <- list(
  "Strongly disagree|Disagree" = "Strongly disagree|Disagree",
  "Disagree|Neither agree nor disagree" = "Disagree|Neither agree nor disagree", 
  "Neither agree nor disagree|Agree" = "Neither agree nor disagree|Agree", 
  "Agree|Strongly agree" = "Agree|Strongly agree", 
  ola_num = "Wave",
  "egpIntermediate class (III+IV)" = "Intermediate class (III+IV)",
  "egpService class (I+II)" = "Service class (I+II)",
  merit_effort_cwc = "Merit: Effort (WE)",
  merit_talent_cwc = "Merit: Talent (WE)",
  merit_effort_mean = "Merit: Effort (BE)",
  merit_talent_mean = "Merit: Talent (BE)",
  "educUniversitary" = "Universitary education (Ref.= Less than Universitary)",
  ideoCenter = "Center",
  ideoRight = "Right",
  "ideoDoes not identify" = "Does not identify",
  sexFemale = "Female (Ref.= Male)",
  age = "Age"
  )

texreg::htmlreg(list(m3,m4,m5,m6),
               caption.above = T,
               caption = NULL,
               stars = c(0.05, 0.01, 0.001),
               custom.coef.map = ccoef,
               digits = 3,
               groups = list("Social Class (Ref.= Working class (V+VI+VII))" = 6:7,
                       "Political identification (Ref.= Left)" =      13:15),
               custom.note = "Note: Cells contain regression coefficients with standard errors in parentheses. %stars.",
               leading.zero = T,
               use.packages = F,
               booktabs = F,
               scalebox = 0.80,
               include.loglik = FALSE,
               include.aic = FALSE,
               center = T)
  Model 1 Model 2 Model 3 Model 4
Strongly disagree|Disagree -0.482*** -0.469*** 0.976*** 0.779**
  (0.085) (0.085) (0.212) (0.259)
Disagree|Neither agree nor disagree 1.884*** 1.902*** 3.347*** 3.150***
  (0.090) (0.090) (0.218) (0.264)
Neither agree nor disagree|Agree 2.540*** 2.558*** 4.003*** 3.808***
  (0.095) (0.095) (0.221) (0.266)
Agree|Strongly agree 5.181*** 5.199*** 6.645*** 6.455***
  (0.133) (0.133) (0.243) (0.285)
Wave 0.220*** 0.224*** 0.223*** 0.224***
  (0.016) (0.016) (0.016) (0.016)
Social Class (Ref.= Working class (V+VI+VII))        
         
     Intermediate class (III+IV) 0.123 0.123 0.153 0.196*
  (0.090) (0.090) (0.088) (0.088)
     Service class (I+II) 0.267* 0.268* 0.260* 0.085
  (0.128) (0.128) (0.125) (0.144)
Merit: Effort (WE)   0.170*** 0.172*** 0.172***
    (0.041) (0.041) (0.041)
Merit: Talent (WE)   0.037 0.037 0.038
    (0.040) (0.040) (0.040)
Merit: Effort (BE)     0.449*** 0.425***
      (0.113) (0.110)
Merit: Talent (BE)     0.102 0.062
      (0.111) (0.108)
Universitary education (Ref.= Less than Universitary)       0.343**
        (0.122)
Political identification (Ref.= Left)        
         
     Center       0.251*
        (0.122)
     Right       0.669***
        (0.141)
     Does not identify       0.167
        (0.105)
Female (Ref.= Male)       -0.421***
        (0.082)
Age       -0.001
        (0.003)
BIC 14860.778 14839.799 14799.102 14790.003
Num. obs. 5755 5755 5755 5755
Groups (idencuesta) 1027 1027 1027 1027
Variance: idencuesta: (Intercept) 1.222 1.179 0.987 0.902
Variance: idencuesta: ola_num 0.025 0.020 0.019 0.020
Note: Cells contain regression coefficients with standard errors in parentheses. ***p < 0.001; **p < 0.01; *p < 0.05.

4.6 Interactions without controls (total effect)

Show the code
# ## WE and BE Interactions without controls

#m7 <- clmm(just_pension ~ 1 + ola_num + egp + merit_effort_cwc + merit_talent_cwc #+ merit_effort_mean + merit_talent_mean + egp*merit_effort_cwc + (1 + ola_num + #merit_effort_cwc| idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#m8 <- clmm(just_pension ~ 1 + ola_num + egp + merit_effort_cwc + merit_talent_cwc #+ merit_effort_mean + merit_talent_mean + egp*merit_talent_cwc + (1 + ola_num + #merit_talent_cwc| idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#
#m9 <- clmm(just_pension ~ 1 + ola_num + egp + merit_effort_cwc + merit_talent_cwc #+ merit_effort_mean + merit_talent_mean + egp*merit_effort_mean + (1 + ola_num + #merit_effort_mean| idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#m10 <- clmm(just_pension ~ 1 + ola_num + egp + merit_effort_cwc + #merit_talent_cwc + merit_effort_mean + merit_talent_mean + egp*merit_talent_mean #+ (1 + ola_num + merit_talent_mean| idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#save(m7,m8,m9,m10, file = here("output/interactions_total.RData"))

load(file = here("output/interactions_total.RData"))

ccoef <- list(
  "Strongly disagree|Disagree" = "Strongly disagree|Disagree",
  "Disagree|Neither agree nor disagree" = "Disagree|Neither agree nor disagree", 
  "Neither agree nor disagree|Agree" = "Neither agree nor disagree|Agree",
  "Agree|Strongly agree" = "Agree|Strongly agree", 
  "egpIntermediate class (III+IV)" = "Intermediate class (III+IV)",
  "egpService class (I+II)" = "Service class (I+II)",
  merit_effort_cwc = "Merit: Effort (WE)",
  merit_talent_cwc = "Merit: Talent (WE)",
  merit_effort_mean = "Merit: Effort (BE)",
  merit_talent_mean = "Merit: Talent (BE)",
  "egpIntermediate class (III+IV):merit_effort_cwc" =  "Intermediate class (III+IV) x Merit: Effort (WE)",
  "egpService class (I+II):merit_effort_cwc" = "Service class (I+II) x Merit: Effort (WE)",
  "egpIntermediate class (III+IV):merit_talent_cwc" = "Intermediate class (III+IV) x Merit: Talent (WE)",
  "egpService class (I+II):merit_talent_cwc" = "Service class (I+II) x Merit: Talent (WE)",
  "egpIntermediate class (III+IV):merit_effort_mean" =  "Intermediate class (III+IV) x Merit: Effort (BE)",
"egpService class (I+II):merit_effort_mean" = "Service class (I+II) x Merit: Effort (BE)",
"egpIntermediate class (III+IV):merit_talent_mean" = "Intermediate class (III+IV) x Merit: Talent (BE)",
"egpService class (I+II):merit_talent_mean" = "Service class (I+II) x Merit: Talent (BE)")

texreg::htmlreg(list(m7,m8,m9,m10),
               caption.above = T,
               caption = NULL,
               stars = c(0.05, 0.01, 0.001),
               custom.coef.map = ccoef,
               digits = 3,
               groups = list("Social Class (Ref.= Working class (V+VI+VII))" = 5:6,
                       "Social Class (Ref.= Working class (V+VI+VII)) x Meritocracy (WE)" =      11:14,
                       "Social Class (Ref.= Working class (V+VI+VII)) x Meritocracy (BE)" =      15:18),
               custom.note = "Note: Cells contain regression coefficients with standard errors in parentheses. %stars.",
               leading.zero = T,
               use.packages = F,
               booktabs = F,
               scalebox = 0.80,
               include.loglik = FALSE,
               include.aic = FALSE,
               center = T,
               custom.gof.rows = list("Controls"=c(rep("No",4))))
  Model 1 Model 2 Model 3 Model 4
Strongly disagree|Disagree 0.952*** 0.974*** 0.735* 0.750*
  (0.214) (0.214) (0.314) (0.320)
Disagree|Neither agree nor disagree 3.350*** 3.380*** 3.104*** 3.120***
  (0.222) (0.222) (0.318) (0.324)
Neither agree nor disagree|Agree 4.012*** 4.043*** 3.759*** 3.775***
  (0.225) (0.225) (0.320) (0.326)
Agree|Strongly agree 6.666*** 6.703*** 6.403*** 6.417***
  (0.248) (0.248) (0.335) (0.341)
Social Class (Ref.= Working class (V+VI+VII))        
         
     Intermediate class (III+IV) 0.154 0.155 -0.437 -0.263
  (0.089) (0.089) (0.420) (0.432)
     Service class (I+II) 0.265* 0.266* 0.481 0.251
  (0.126) (0.126) (0.589) (0.600)
Merit: Effort (WE) 0.134* 0.174*** 0.172*** 0.172***
  (0.058) (0.042) (0.041) (0.041)
Merit: Talent (WE) 0.038 -0.006 0.037 0.038
  (0.041) (0.057) (0.040) (0.040)
Merit: Effort (BE) 0.444*** 0.462*** 0.372** 0.443***
  (0.114) (0.114) (0.142) (0.112)
Merit: Talent (BE) 0.097 0.090 0.088 0.026
  (0.112) (0.113) (0.110) (0.141)
Social Class (Ref.= Working class (V+VI+VII)) x Meritocracy (WE)        
         
     Intermediate class (III+IV) x Merit: Effort (WE) 0.074      
  (0.071)      
     Service class (I+II) x Merit: Effort (WE) 0.018      
  (0.107)      
     Intermediate class (III+IV) x Merit: Talent (WE)   0.080    
    (0.071)    
     Service class (I+II) x Merit: Talent (WE)   0.071    
    (0.107)    
Social Class (Ref.= Working class (V+VI+VII)) x Meritocracy (BE)        
         
     Intermediate class (III+IV) x Merit: Effort (BE)     0.229  
      (0.159)  
     Service class (I+II) x Merit: Effort (BE)     -0.091  
      (0.220)  
     Intermediate class (III+IV) x Merit: Talent (BE)       0.151
        (0.155)
     Service class (I+II) x Merit: Talent (BE)       -0.002
        (0.211)
Controls No No No No
BIC 14832.108 14830.095 14831.984 14836.288
Num. obs. 5755 5755 5755 5755
Groups (idencuesta) 1027 1027 1027 1027
Variance: idencuesta: (Intercept) 1.003 1.003 3.889 3.287
Variance: idencuesta: ola_num 0.017 0.017 0.019 0.019
Variance: idencuesta: merit_effort_cwc 0.079      
Variance: idencuesta: merit_talent_cwc   0.092    
Variance: idencuesta: merit_effort_mean     0.407  
Variance: idencuesta: merit_talent_mean       0.330
Note: Cells contain regression coefficients with standard errors in parentheses. ***p < 0.001; **p < 0.01; *p < 0.05.

4.7 Interactions with controls (direct effect)

Show the code
# ## WE and BE Interactions with controls

#m11 <- clmm(just_pension ~ 1 + ola_num + egp + merit_effort_cwc + #merit_talent_cwc + merit_effort_mean + merit_talent_mean + ideo + sex + age + #egp*merit_effort_cwc + (1 + ola_num + merit_effort_cwc| idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#m12 <- clmm(just_pension ~ 1 + ola_num + egp + merit_effort_cwc + #merit_talent_cwc + merit_effort_mean + merit_talent_mean + ideo + sex + age + #egp*merit_talent_cwc + (1 + ola_num + merit_talent_cwc| idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#
#m13 <- clmm(just_pension ~ 1 + ola_num + egp + merit_effort_cwc + #merit_talent_cwc + merit_effort_mean + merit_talent_mean + ideo + sex + age + #egp*merit_effort_mean + (1 + ola_num + merit_effort_mean| idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#m14 <- clmm(just_pension ~ 1 + ola_num + egp + merit_effort_cwc + #merit_talent_cwc + merit_effort_mean + merit_talent_mean + ideo + sex + age + #egp*merit_talent_mean + (1 + ola_num + merit_talent_mean| idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#save(m11,m12,m13,m14, file = here("output/interactions_direct.RData"))

load(file = here("output/interactions_direct.RData"))

ccoef <- list(
  "Strongly disagree|Disagree" = "Strongly disagree|Disagree",
  "Disagree|Neither agree nor disagree" = "Disagree|Neither agree nor disagree", 
  "Neither agree nor disagree|Agree" = "Neither agree nor disagree|Agree",
  "Agree|Strongly agree" = "Agree|Strongly agree", 
  "egpIntermediate class (III+IV)" = "Intermediate class (III+IV)",
  "egpService class (I+II)" = "Service class (I+II)",
  merit_effort_cwc = "Merit: Effort (WE)",
  merit_talent_cwc = "Merit: Talent (WE)",
  merit_effort_mean = "Merit: Effort (BE)",
  merit_talent_mean = "Merit: Talent (BE)",
  "egpIntermediate class (III+IV):merit_effort_cwc" =  "Intermediate class (III+IV) x Merit: Effort (WE)",
  "egpService class (I+II):merit_effort_cwc" = "Service class (I+II) x Merit: Effort (WE)",
  "egpIntermediate class (III+IV):merit_talent_cwc" = "Intermediate class (III+IV) x Merit: Talent (WE)",
  "egpService class (I+II):merit_talent_cwc" = "Service class (I+II) x Merit: Talent (WE)",
  "egpIntermediate class (III+IV):merit_effort_mean" =  "Intermediate class (III+IV) x Merit: Effort (BE)",
"egpService class (I+II):merit_effort_mean" = "Service class (I+II) x Merit: Effort (BE)",
"egpIntermediate class (III+IV):merit_talent_mean" = "Intermediate class (III+IV) x Merit: Talent (BE)",
"egpService class (I+II):merit_talent_mean" = "Service class (I+II) x Merit: Talent (BE)")

texreg::htmlreg(list(m11,m12,m13,m14),
               caption.above = T,
               caption = NULL,
               stars = c(0.05, 0.01, 0.001),
               custom.coef.map = ccoef,
               digits = 3,
               groups = list("Social Class (Ref.= Working class (V+VI+VII))" = 5:6,
                       "Social Class (Ref.= Working class (V+VI+VII)) x Meritocracy (WE)" =      11:14,
                       "Social Class (Ref.= Working class (V+VI+VII)) x Meritocracy (BE)" =      15:18),
               custom.note = "Note: Cells contain regression coefficients with standard errors in parentheses. %stars.",
               leading.zero = T,
               use.packages = F,
               booktabs = F,
               scalebox = 0.80,
               include.loglik = FALSE,
               include.aic = FALSE,
               center = T,
               custom.gof.rows = list("Controls"=c(rep("Yes",4))))
  Model 1 Model 2 Model 3 Model 4
Strongly disagree|Disagree 0.631* 0.650* 0.299 0.502
  (0.259) (0.259) (0.348) (0.334)
Disagree|Neither agree nor disagree 3.030*** 3.056*** 2.669*** 2.854***
  (0.264) (0.264) (0.350) (0.337)
Neither agree nor disagree|Agree 3.693*** 3.722*** 3.326*** 3.512***
  (0.266) (0.267) (0.352) (0.339)
Agree|Strongly agree 6.352*** 6.386*** 5.971*** 6.173***
  (0.286) (0.286) (0.365) (0.353)
Social Class (Ref.= Working class (V+VI+VII))        
         
     Intermediate class (III+IV) 0.233** 0.234** -0.436 -0.270
  (0.088) (0.088) (0.412) (0.391)
     Service class (I+II) 0.300* 0.302* 0.323 0.264
  (0.125) (0.125) (0.577) (0.543)
Merit: Effort (WE) 0.136* 0.174*** 0.172*** 0.171***
  (0.058) (0.042) (0.041) (0.041)
Merit: Talent (WE) 0.039 -0.004 0.038 0.038
  (0.041) (0.057) (0.040) (0.040)
Merit: Effort (BE) 0.423*** 0.439*** 0.328* 0.442***
  (0.111) (0.112) (0.138) (0.111)
Merit: Talent (BE) 0.045 0.038 0.032 -0.025
  (0.109) (0.110) (0.108) (0.135)
Social Class (Ref.= Working class (V+VI+VII)) x Meritocracy (WE)        
         
     Intermediate class (III+IV) x Merit: Effort (WE) 0.072      
  (0.071)      
     Service class (I+II) x Merit: Effort (WE) 0.016      
  (0.107)      
     Intermediate class (III+IV) x Merit: Talent (WE)   0.078    
    (0.071)    
     Service class (I+II) x Merit: Talent (WE)   0.071    
    (0.107)    
Social Class (Ref.= Working class (V+VI+VII)) x Meritocracy (BE)        
         
     Intermediate class (III+IV) x Merit: Effort (BE)     0.259  
      (0.155)  
     Service class (I+II) x Merit: Effort (BE)     -0.018  
      (0.215)  
     Intermediate class (III+IV) x Merit: Talent (BE)       0.184
        (0.145)
     Service class (I+II) x Merit: Talent (BE)       0.010
        (0.199)
Controls Yes Yes Yes Yes
BIC 14823.107 14820.462 14820.991 14853.242
Num. obs. 5755 5755 5755 5755
Groups (idencuesta) 1027 1027 1027 1027
Variance: idencuesta: (Intercept) 0.937 0.938 3.839 0.000
Variance: idencuesta: ola_num 0.018 0.018 0.020 0.023
Variance: idencuesta: merit_effort_cwc 0.079      
Variance: idencuesta: merit_talent_cwc   0.094    
Variance: idencuesta: merit_effort_mean     0.414  
Variance: idencuesta: merit_talent_mean       0.089
Note: Cells contain regression coefficients with standard errors in parentheses. ***p < 0.001; **p < 0.01; *p < 0.05.

4.8 Interactions with meritocratic variables as dichotomized variables without controls

Show the code
## WE and BE Interactions merit dummy

#m15 <- clmm(just_pension ~ 1 + ola_num + egp + merit_effort_dic_cwc + #merit_talent_dic_cwc + merit_effort_dic_mean + merit_talent_dic_mean + #egp*merit_effort_dic_cwc + (1 + ola_num + merit_effort_dic_cwc| idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#m16 <- clmm(just_pension ~ 1 + ola_num + egp + merit_effort_dic_cwc + #merit_talent_dic_cwc + merit_effort_dic_mean + merit_talent_dic_mean + #egp*merit_talent_dic_cwc + (1 + ola_num + merit_talent_dic_cwc| idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#
#m17 <- clmm(just_pension ~ 1 + ola_num + egp + merit_effort_dic_cwc + #merit_talent_dic_cwc + merit_effort_dic_mean + merit_talent_dic_mean + #egp*merit_effort_dic_mean + (1 + ola_num + merit_effort_dic_mean| idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#m18 <- clmm(just_pension ~ 1 + ola_num + egp + merit_effort_dic_cwc + #merit_talent_dic_cwc + merit_effort_dic_mean + merit_talent_dic_mean + #egp*merit_talent_dic_mean + (1 + ola_num + merit_talent_dic_mean| idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#save(m15,m16,m17,m18, file = here("output/interactions_total_dico.RData"))

load(file = here("output/interactions_total_dico.RData"))
     
ccoef <- list(
  "Strongly disagree|Disagree" = "Strongly disagree|Disagree",
  "Disagree|Neither agree nor disagree" = "Disagree|Neither agree nor disagree", 
  "Neither agree nor disagree|Agree" = "Neither agree nor disagree|Agree",
  "Agree|Strongly agree" = "Agree|Strongly agree", 
  "egpIntermediate class (III+IV)" = "Intermediate class (III+IV)",
  "egpService class (I+II)" = "Service class (I+II)",
  merit_effort_dic_cwc = "Merit: Effort (WE)",
  merit_talent_dic_cwc = "Merit: Talent (WE)",
  merit_effort_dic_mean = "Merit: Effort (BE)",
  merit_talent_dic_mean = "Merit: Talent (BE)",
  "egpIntermediate class (III+IV):merit_effort_dic_cwc" =  "Intermediate class (III+IV) x Merit: Effort (WE)",
  "egpService class (I+II):merit_effort_dic_cwc" = "Service class (I+II) x Merit: Effort (WE)",
  "egpIntermediate class (III+IV):merit_talent_dic_cwc" = "Intermediate class (III+IV) x Merit: Talent (WE)",
  "egpService class (I+II):merit_talent_dic_cwc" = "Service class (I+II) x Merit: Talent (WE)",
  "egpIntermediate class (III+IV):merit_effort_dic_mean" =  "Intermediate class (III+IV) x Merit: Effort (BE)",
"egpService class (I+II):merit_effort_dic_mean" = "Service class (I+II) x Merit: Effort (BE)",
"egpIntermediate class (III+IV):merit_talent_dic_mean" = "Intermediate class (III+IV) x Merit: Talent (BE)",
"egpService class (I+II):merit_talent_dic_mean" = "Service class (I+II) x Merit: Talent (BE)")

texreg::htmlreg(list(m15,m16,m17,m18),
               caption.above = T,
               caption = NULL,
               stars = c(0.05, 0.01, 0.001),
               custom.coef.map = ccoef,
               digits = 3,
               groups = list("Social Class (Ref.= Working class (V+VI+VII))" = 5:6,
                       "Social Class (Ref.= Working class (V+VI+VII)) x Meritocracy (WE)" =      11:14,
                       "Social Class (Ref.= Working class (V+VI+VII)) x Meritocracy (BE)" =      15:18),
               custom.note = "Note: Cells contain regression coefficients with standard errors in parentheses. %stars.",
               leading.zero = T,
               use.packages = F,
               booktabs = F,
               scalebox = 0.80,
               include.loglik = FALSE,
               include.aic = FALSE,
               center = T,
               custom.gof.rows = list("Controls"=c(rep("No",4))))
  Model 1 Model 2 Model 3 Model 4
Strongly disagree|Disagree -0.257** -0.260** -0.315** -0.305**
  (0.098) (0.098) (0.109) (0.116)
Disagree|Neither agree nor disagree 2.114*** 2.111*** 2.048*** 2.058***
  (0.104) (0.104) (0.115) (0.121)
Neither agree nor disagree|Agree 2.772*** 2.769*** 2.705*** 2.714***
  (0.108) (0.108) (0.119) (0.125)
Agree|Strongly agree 5.420*** 5.418*** 5.351*** 5.358***
  (0.144) (0.144) (0.151) (0.157)
Social Class (Ref.= Working class (V+VI+VII))        
         
     Intermediate class (III+IV) 0.147 0.146 0.028 0.086
  (0.089) (0.089) (0.127) (0.140)
     Service class (I+II) 0.289* 0.288* 0.239 0.197
  (0.127) (0.127) (0.177) (0.194)
Merit: Effort (WE) 0.086 0.227* 0.222* 0.223*
  (0.128) (0.091) (0.091) (0.091)
Merit: Talent (WE) -0.035 -0.157 -0.037 -0.038
  (0.084) (0.116) (0.084) (0.084)
Merit: Effort (BE) 0.627* 0.624* 0.401 0.620*
  (0.260) (0.260) (0.331) (0.256)
Merit: Talent (BE) 0.215 0.219 0.191 0.055
  (0.243) (0.243) (0.241) (0.318)
Social Class (Ref.= Working class (V+VI+VII)) x Meritocracy (WE)        
         
     Intermediate class (III+IV) x Merit: Effort (WE) 0.303      
  (0.160)      
     Service class (I+II) x Merit: Effort (WE) -0.042      
  (0.234)      
     Intermediate class (III+IV) x Merit: Talent (WE)   0.250    
    (0.147)    
     Service class (I+II) x Merit: Talent (WE)   -0.005    
    (0.221)    
Social Class (Ref.= Working class (V+VI+VII)) x Meritocracy (BE)        
         
     Intermediate class (III+IV) x Merit: Effort (BE)     0.525  
      (0.389)  
     Service class (I+II) x Merit: Effort (BE)     0.179  
      (0.554)  
     Intermediate class (III+IV) x Merit: Talent (BE)       0.194
        (0.368)
     Service class (I+II) x Merit: Talent (BE)       0.293
        (0.502)
Controls No No No No
BIC 14904.006 14904.716 14904.839 14903.542
Num. obs. 5755 5755 5755 5755
Groups (idencuesta) 1027 1027 1027 1027
Variance: idencuesta: (Intercept) 1.128 1.128 1.158 1.296
Variance: idencuesta: ola_num 0.024 0.024 0.023 0.023
Variance: idencuesta: merit_effort_dic_cwc 0.041      
Variance: idencuesta: merit_talent_dic_cwc   0.038    
Variance: idencuesta: merit_effort_dic_mean     1.791  
Variance: idencuesta: merit_talent_dic_mean       2.522
Note: Cells contain regression coefficients with standard errors in parentheses. ***p < 0.001; **p < 0.01; *p < 0.05.

4.9 Interactions with meritocratic variables as dichotomized variables with controls

Show the code
## WE and BE Interactions merit dummy

#m19 <- clmm(just_pension ~ 1 + ola_num + egp + merit_effort_dic_cwc + #merit_talent_dic_cwc + merit_effort_dic_mean + merit_talent_dic_mean + ideo + sex #+ age + egp*merit_effort_dic_cwc + (1 + ola_num + merit_effort_dic_cwc| #idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#m20 <- clmm(just_pension ~ 1 + ola_num + egp + merit_effort_dic_cwc + #merit_talent_dic_cwc + merit_effort_dic_mean + merit_talent_dic_mean + ideo + sex #+ age + egp*merit_talent_dic_cwc + (1 + ola_num + merit_talent_dic_cwc| #idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#
#m21 <- clmm(just_pension ~ 1 + ola_num + egp + merit_effort_dic_cwc + #merit_talent_dic_cwc + merit_effort_dic_mean + merit_talent_dic_mean + ideo + sex #+ age + egp*merit_effort_dic_mean + (1 + ola_num + merit_effort_dic_mean| #idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#m22 <- clmm(just_pension ~ 1 + ola_num + egp + merit_effort_dic_cwc + #merit_talent_dic_cwc + merit_effort_dic_mean + merit_talent_dic_mean + ideo + sex #+ age + egp*merit_talent_dic_mean + (1 + ola_num + merit_talent_dic_mean| #idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#save(m19,m20,m21,m22, file = here("output/interactions_direct_dic.RData"))

load(file = here("output/interactions_direct_dic.RData"))
     
ccoef <- list(
  "Strongly disagree|Disagree" = "Strongly disagree|Disagree",
  "Disagree|Neither agree nor disagree" = "Disagree|Neither agree nor disagree", 
  "Neither agree nor disagree|Agree" = "Neither agree nor disagree|Agree",
  "Agree|Strongly agree" = "Agree|Strongly agree", 
  "egpIntermediate class (III+IV)" = "Intermediate class (III+IV)",
  "egpService class (I+II)" = "Service class (I+II)",
  merit_effort_dic_cwc = "Merit: Effort (WE)",
  merit_talent_dic_cwc = "Merit: Talent (WE)",
  merit_effort_dic_mean = "Merit: Effort (BE)",
  merit_talent_dic_mean = "Merit: Talent (BE)",
  "egpIntermediate class (III+IV):merit_effort_dic_cwc" =  "Intermediate class (III+IV) x Merit: Effort (WE)",
  "egpService class (I+II):merit_effort_dic_cwc" = "Service class (I+II) x Merit: Effort (WE)",
  "egpIntermediate class (III+IV):merit_talent_dic_cwc" = "Intermediate class (III+IV) x Merit: Talent (WE)",
  "egpService class (I+II):merit_talent_dic_cwc" = "Service class (I+II) x Merit: Talent (WE)",
  "egpIntermediate class (III+IV):merit_effort_dic_mean" =  "Intermediate class (III+IV) x Merit: Effort (BE)",
"egpService class (I+II):merit_effort_dic_mean" = "Service class (I+II) x Merit: Effort (BE)",
"egpIntermediate class (III+IV):merit_talent_dic_mean" = "Intermediate class (III+IV) x Merit: Talent (BE)",
"egpService class (I+II):merit_talent_dic_mean" = "Service class (I+II) x Merit: Talent (BE)")

texreg::htmlreg(list(m19,m20,m21,m22),
               caption.above = T,
               caption = NULL,
               stars = c(0.05, 0.01, 0.001),
               custom.coef.map = ccoef,
               digits = 3,
               groups = list("Social Class (Ref.= Working class (V+VI+VII))" = 5:6,
                       "Social Class (Ref.= Working class (V+VI+VII)) x Meritocracy (WE)" =      11:14,
                       "Social Class (Ref.= Working class (V+VI+VII)) x Meritocracy (BE)" =      15:18),
               custom.note = "Note: Cells contain regression coefficients with standard errors in parentheses. %stars.",
               leading.zero = T,
               use.packages = F,
               booktabs = F,
               scalebox = 0.80,
               include.loglik = FALSE,
               include.aic = FALSE,
               center = T,
               custom.gof.rows = list("Controls"=c(rep("Yes",4))))
  Model 1 Model 2 Model 3 Model 4
Strongly disagree|Disagree -0.383* -0.385* -0.482* -0.489*
  (0.191) (0.191) (0.200) (0.203)
Disagree|Neither agree nor disagree 1.988*** 1.986*** 1.883*** 1.875***
  (0.194) (0.194) (0.202) (0.205)
Neither agree nor disagree|Agree 2.648*** 2.646*** 2.541*** 2.533***
  (0.196) (0.196) (0.204) (0.207)
Agree|Strongly agree 5.299*** 5.298*** 5.189*** 5.180***
  (0.218) (0.218) (0.225) (0.228)
Social Class (Ref.= Working class (V+VI+VII))        
         
     Intermediate class (III+IV) 0.229** 0.229** 0.099 0.135
  (0.089) (0.089) (0.124) (0.137)
     Service class (I+II) 0.326** 0.325** 0.245 0.188
  (0.126) (0.126) (0.173) (0.189)
Merit: Effort (WE) 0.090 0.228* 0.224* 0.225*
  (0.128) (0.091) (0.091) (0.091)
Merit: Talent (WE) -0.035 -0.154 -0.037 -0.038
  (0.084) (0.117) (0.084) (0.084)
Merit: Effort (BE) 0.589* 0.587* 0.323 0.587*
  (0.253) (0.253) (0.322) (0.250)
Merit: Talent (BE) 0.065 0.069 0.041 -0.170
  (0.239) (0.239) (0.237) (0.310)
Social Class (Ref.= Working class (V+VI+VII)) x Meritocracy (WE)        
         
     Intermediate class (III+IV) x Merit: Effort (WE) 0.296      
  (0.160)      
     Service class (I+II) x Merit: Effort (WE) -0.046      
  (0.234)      
     Intermediate class (III+IV) x Merit: Talent (WE)   0.244    
    (0.147)    
     Service class (I+II) x Merit: Talent (WE)   -0.008    
    (0.220)    
Social Class (Ref.= Working class (V+VI+VII)) x Meritocracy (BE)        
         
     Intermediate class (III+IV) x Merit: Effort (BE)     0.572  
      (0.381)  
     Service class (I+II) x Merit: Effort (BE)     0.299  
      (0.540)  
     Intermediate class (III+IV) x Merit: Talent (BE)       0.318
        (0.358)
     Service class (I+II) x Merit: Talent (BE)       0.451
        (0.486)
Controls Yes Yes Yes Yes
BIC 14889.373 14890.219 14889.362 14889.076
Num. obs. 5755 5755 5755 5755
Groups (idencuesta) 1027 1027 1027 1027
Variance: idencuesta: (Intercept) 1.056 1.053 1.008 1.100
Variance: idencuesta: ola_num 0.025 0.025 0.024 0.024
Variance: idencuesta: merit_effort_dic_cwc 0.039      
Variance: idencuesta: merit_talent_dic_cwc   0.035    
Variance: idencuesta: merit_effort_dic_mean     1.535  
Variance: idencuesta: merit_talent_dic_mean       1.920
Note: Cells contain regression coefficients with standard errors in parentheses. ***p < 0.001; **p < 0.01; *p < 0.05.

4.10 Interactions meritocracy x time without controls

Show the code
## Interactions merit x time

#m23 <- clmm(just_pension ~ 1 + ola_num + egp + merit_effort_cwc + merit_talent_cwc + merit_effort_mean + #merit_talent_mean + merit_effort_cwc*ola_num + (1 + ola_num| idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#m24 <- clmm(just_pension ~ 1 + ola_num + egp + merit_effort_cwc + merit_talent_cwc + merit_effort_mean + #merit_talent_mean + merit_talent_cwc*ola_num + (1 + ola_num | idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#
#m25 <- clmm(just_pension ~ 1 + ola_num + egp + merit_effort_cwc + merit_talent_cwc + merit_effort_mean + #merit_talent_mean + merit_effort_mean*ola_num + (1 + ola_num | idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#m26 <- clmm(just_pension ~ 1 + ola_num + egp + merit_effort_cwc + merit_talent_cwc + merit_effort_mean + #merit_talent_mean  + merit_talent_mean*ola_num + (1 + ola_num | idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#save(m23,m24,m25,m26, file = here("output/interactions_merit_time_total.RData"))

load(file = here("output/interactions_merit_time_total.RData"))

ccoef <- list(
  "Strongly disagree|Disagree" = "Strongly disagree|Disagree",
  "Disagree|Neither agree nor disagree" = "Disagree|Neither agree nor disagree", 
  "Neither agree nor disagree|Agree" = "Neither agree nor disagree|Agree",
  "Agree|Strongly agree" = "Agree|Strongly agree", 
  merit_effort_cwc = "Merit: Effort (WE)",
  merit_talent_cwc = "Merit: Talent (WE)",
  merit_effort_mean = "Merit: Effort (BE)",
  merit_talent_mean = "Merit: Talent (BE)",
  "ola_num:merit_effort_cwc" =  "Time x Merit: Effort (WE)",
  "ola_num:merit_talent_cwc" = "Time x Merit: Talent (WE)",
  "ola_num:merit_effort_mean" = "Time x Merit: Effort (BE)",
  "ola_num:merit_talent_mean" = "Time x Merit: Talent (BE)")

texreg::htmlreg(list(m23,m24,m25,m26),
               caption.above = T,
               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.",
               leading.zero = T,
               use.packages = F,
               booktabs = F,
               scalebox = 0.80,
               include.loglik = FALSE,
               include.aic = FALSE,
               center = T,
               custom.gof.rows = list("Controls"=c(rep("No",4))))
  Model 1 Model 2 Model 3 Model 4
Strongly disagree|Disagree 0.976*** 0.976*** 1.588*** 1.733***
  (0.212) (0.212) (0.309) (0.316)
Disagree|Neither agree nor disagree 3.347*** 3.347*** 3.960*** 4.105***
  (0.218) (0.218) (0.315) (0.322)
Neither agree nor disagree|Agree 4.003*** 4.003*** 4.616*** 4.761***
  (0.221) (0.221) (0.317) (0.325)
Agree|Strongly agree 6.645*** 6.645*** 7.254*** 7.400***
  (0.243) (0.243) (0.333) (0.340)
Merit: Effort (WE) 0.212** 0.172*** 0.170*** 0.173***
  (0.077) (0.041) (0.041) (0.041)
Merit: Talent (WE) 0.038 0.032 0.038 0.034
  (0.040) (0.076) (0.040) (0.040)
Merit: Effort (BE) 0.448*** 0.449*** 0.686*** 0.450***
  (0.113) (0.113) (0.143) (0.113)
Merit: Talent (BE) 0.103 0.102 0.101 0.377**
  (0.111) (0.111) (0.111) (0.140)
Time x Merit: Effort (WE) -0.012      
  (0.020)      
Time x Merit: Talent (WE)   0.002    
    (0.019)    
Time x Merit: Effort (BE)     -0.070**  
      (0.026)  
Time x Merit: Talent (BE)       -0.083**
        (0.025)
Controls No No No No
BIC 14807.375 14807.753 14800.313 14797.237
Num. obs. 5755 5755 5755 5755
Groups (idencuesta) 1027 1027 1027 1027
Variance: idencuesta: (Intercept) 0.984 0.987 0.981 0.976
Variance: idencuesta: ola_num 0.019 0.019 0.018 0.017
Note: Cells contain regression coefficients with standard errors in parentheses. ***p < 0.001; **p < 0.01; *p < 0.05.

4.11 Interactions meritocracy x time with controls

Show the code
## Interactions merit x time

#m27 <- clmm(just_pension ~ 1 + ola_num + egp + merit_effort_cwc + merit_talent_cwc + merit_effort_mean + #merit_talent_mean + educ + ideo + sex + age + merit_effort_cwc*ola_num + (1 + ola_num| idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#m28 <- clmm(just_pension ~ 1 + ola_num + egp + merit_effort_cwc + merit_talent_cwc + merit_effort_mean + #merit_talent_mean + educ + ideo + sex + age + merit_talent_cwc*ola_num + (1 + ola_num | idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#
#m29 <- clmm(just_pension ~ 1 + ola_num + egp + merit_effort_cwc + merit_talent_cwc + merit_effort_mean + #merit_talent_mean + educ + ideo + sex + age + merit_effort_mean*ola_num + (1 + ola_num | idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#m30 <- clmm(just_pension ~ 1 + ola_num + egp + merit_effort_cwc + merit_talent_cwc + merit_effort_mean + #merit_talent_mean + educ + ideo + sex + age + merit_talent_mean*ola_num + (1 + ola_num | idencuesta), 
#            link = "logit",
#  Hess = TRUE,
#           data = df_study1)
#
#save(m27,m28,m29,m30, file = here("output/interactions_merit_time_direct.RData"))

load(file = here("output/interactions_merit_time_direct.RData"))

ccoef <- list(
  "Strongly disagree|Disagree" = "Strongly disagree|Disagree",
  "Disagree|Neither agree nor disagree" = "Disagree|Neither agree nor disagree", 
  "Neither agree nor disagree|Agree" = "Neither agree nor disagree|Agree",
  "Agree|Strongly agree" = "Agree|Strongly agree", 
  merit_effort_cwc = "Merit: Effort (WE)",
  merit_talent_cwc = "Merit: Talent (WE)",
  merit_effort_mean = "Merit: Effort (BE)",
  merit_talent_mean = "Merit: Talent (BE)",
  "ola_num:merit_effort_cwc" =  "Time x Merit: Effort (WE)",
  "ola_num:merit_talent_cwc" = "Time x Merit: Talent (WE)",
  "ola_num:merit_effort_mean" = "Time x Merit: Effort (BE)",
  "ola_num:merit_talent_mean" = "Time x Merit: Talent (BE)")

texreg::htmlreg(list(m27,m28,m29,m30),
               caption.above = T,
               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.",
               leading.zero = T,
               use.packages = F,
               booktabs = F,
               scalebox = 0.80,
               include.loglik = FALSE,
               include.aic = FALSE,
               center = T,
               custom.gof.rows = list("Controls"=c(rep("Yes",4))))
  Model 1 Model 2 Model 3 Model 4
Strongly disagree|Disagree 0.906*** 0.909 1.393*** 1.531***
  (0.220)   (0.343) (0.349)
Disagree|Neither agree nor disagree 3.171*** 3.172 3.765*** 3.905***
  (0.226)   (0.348) (0.354)
Neither agree nor disagree|Agree 3.813*** 3.813 4.423*** 4.563***
  (0.228)   (0.350) (0.356)
Agree|Strongly agree 6.470*** 6.471 7.066*** 7.207***
  (0.250)   (0.364) (0.370)
Merit: Effort (WE) 0.223** 0.163 0.170*** 0.172***
  (0.075)   (0.041) (0.041)
Merit: Talent (WE) 0.032 0.032 0.039 0.034
  (0.040)   (0.040) (0.040)
Merit: Effort (BE) 0.408*** 0.409 0.663*** 0.426***
  (0.093)   (0.141) (0.110)
Merit: Talent (BE) 0.095 0.094 0.061 0.336*
  (0.091)   (0.108) (0.137)
Time x Merit: Effort (WE) -0.018      
  (0.019)      
Time x Merit: Talent (WE)   -0.001    
         
Time x Merit: Effort (BE)     -0.071**  
      (0.026)  
Time x Merit: Talent (BE)       -0.082**
        (0.025)
Controls Yes Yes Yes Yes
BIC 14900.956 14901.875 14791.204 14788.267
Num. obs. 5755 5755 5755 5755
Groups (idencuesta) 1027 1027 1027 1027
Variance: idencuesta: (Intercept) 0.000 0.000 0.895 0.891
Variance: idencuesta: ola_num 0.049 0.049 0.019 0.019
Note: Cells contain regression coefficients with standard errors in parentheses. ***p < 0.001; **p < 0.01; *p < 0.05.