A importância do setor no salário na Suécia, uma comparação entre diferentes grupos ocupacionais, parte 3

Para concluir a análise sobre a significância do setor no salário para diferentes grupos ocupacionais na Suécia, examinarei neste post a correlação entre salário e setor usando estatísticas para educação.

O valor F da tabela Anova é usado como o valor único para discriminar o quanto a região e o salário se correlacionam. Para análise exploratória, o valor Anova parece bom o suficiente.

Primeiro, defina bibliotecas e funções.

library (tidyverse)

## -- Attaching packages --------------------------------------------------- tidyverse 1.3.0 --

## v ggplot2 3.2.1     v purrr   0.3.3
## v tibble  2.1.3     v dplyr   0.8.3
## v tidyr   1.0.2     v stringr 1.4.0
## v readr   1.3.1     v forcats 0.4.0

## -- Conflicts ------------------------------------------------------ tidyverse_conflicts() --
## x dplyr::filter() masks stats::filter()
## x dplyr::lag()    masks stats::lag()

library (broom)
library (car)

## Loading required package: carData

## 
## Attaching package: 'car'

## The following object is masked from 'package:dplyr':
## 
##     recode

## The following object is masked from 'package:purrr':
## 
##     some

library (sjPlot)

## Registered S3 methods overwritten by 'lme4':
##   method                          from
##   cooks.distance.influence.merMod car 
##   influence.merMod                car 
##   dfbeta.influence.merMod         car 
##   dfbetas.influence.merMod        car

## Learn more about sjPlot with 'browseVignettes("sjPlot")'.

readfile %
  gather (starts_with("19"), starts_with("20"), key = "year", value = salary) %>%
  drop_na() %>%
  mutate (year_n = parse_number (year))
}

A tabela de dados é baixada do Statistics Sweden. Ele é salvo como um arquivo delimitado por vírgula sem cabeçalho, 000000CY.csv, http://www.statistikdatabasen.scb.se/pxweb/en/ssd/.

Renomeei o arquivo para 000000CY_sector.csv porque o nome do arquivo 000000CY.csv foi usado em uma postagem anterior.

A tabela: Salário básico médio, salário mensal e salário das mulheres como porcentagem do salário de homens por setor, grupo ocupacional (SSYK 2012), sexo e nível de escolaridade (SUN). Ano 2014 – 2018 Salário mensal 1-3 setor público 4-5 setor privado

No gráfico e nas tabelas, também é possível encontrar informações sobre como o aumento dos salários por ano para cada grupo ocupacional é afetado quando as interações são levadas em consideração.

tb % 
  mutate(edulevel = `level of education`)

numedulevel %
  knitr::kable(
  booktabs = TRUE,
  caption = 'Initial approach, length of education') 

tbnum % 
  right_join(numedulevel, by = c("level of education" = "level.of.education")) %>%
  filter(!is.na(eduyears)) %>%
  mutate(eduyears = factor(eduyears))

## Warning: Column `level of education`/`level.of.education` joining character
## vector and factor, coercing into character vector

summary_table = vector()
anova_table = vector()

for (i in unique(tbnum$`occuptional  (SSYK 2012)`)){
  temp  90){
    model <- lm(log(salary) ~ edulevel + sex + year_n + sector, data = temp)
    summary_table <- rbind (summary_table, mutate (tidy (summary (model)), ssyk = i, interaction = "none"))
    anova_table <- rbind (anova_table, mutate (tidy (Anova (model, type = 2)), ssyk = i, interaction = "none"))  
  
    model <- lm(log(salary) ~ edulevel * sector + sex + year_n, data = temp)
    summary_table <- rbind (summary_table, mutate (tidy (summary (model)), ssyk = i, interaction = "sector and edulevel"))
    anova_table <- rbind (anova_table, mutate (tidy (Anova (model, type = 2)), ssyk = i, interaction = "sector and edulevel"))  
    
    model <- lm(log(salary) ~ edulevel + sector * sex + year_n, data = temp)
    summary_table <- rbind (summary_table, mutate (tidy (summary (model)), ssyk = i, interaction = "sector and sex"))
    anova_table <- rbind (anova_table, mutate (tidy (Anova (model, type = 2)), ssyk = i, interaction = "sector and sex"))  
    
    model <- lm(log(salary) ~ edulevel +  year_n * sector + sex, data = temp)
    summary_table <- rbind (summary_table, mutate (tidy (summary (model)), ssyk = i, interaction = "sector and year"))
    anova_table <- rbind (anova_table, mutate (tidy (Anova (model, type = 2)), ssyk = i, interaction = "sector and year"))  
    
    model <- lm(log(salary) ~ edulevel * sector * sex * year_n, data = temp)
    summary_table <- rbind (summary_table, mutate (tidy (summary (model)), ssyk = i, interaction = "sector, year, edulevel and sex"))
    anova_table <- rbind (anova_table, mutate (tidy (Anova (model, type = 2)), ssyk = i, interaction = "sector, year, edulevel and sex"))      
  }
}

## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison
## Note: model has aliased coefficients
##       sums of squares computed by model comparison

anova_table % rowwise() %>% mutate(contcol = str_count(term, ":")) 

summary_table % rowwise() %>% mutate(contcol = str_count(term, ":"))

merge(summary_table, anova_table, by = c("ssyk", "interaction"), all = TRUE) %>%
  filter (term.x == "year_n") %>%
  filter (term.y == "sector") %>%    
  filter (interaction == "none") %>%
  
  mutate (estimate = (exp(estimate) - 1) * 100) %>%  
  ggplot () +
    geom_point (mapping = aes(x = estimate, y = statistic.y, colour = interaction)) +
    labs(
      x = "Increase in salaries (% / year)",
      y = "F-value for sector"
    )   

Leia Também  3 razões para investir parece complicado

Figura 1: A significância do setor sobre o salário na Suécia, uma comparação entre diferentes grupos ocupacionais, ano 2014 - 2018

merge(summary_table, anova_table, by = c("ssyk", "interaction"), all = TRUE) %>%
  filter (term.x == "year_n") %>%
  filter (contcol.y > 0) %>%    
  # only look at the interactions between all four variables in the case with interaction sector, year, edulevel and sex
  filter (!(contcol.y % 
  
  mutate (estimate = (exp(estimate) - 1) * 100) %>%  
  ggplot () +
    geom_point (mapping = aes(x = estimate, y = statistic.y, colour = interaction)) +
    labs(
      x = "Increase in salaries (% / year)",
      y = "F-value for interaction"
    ) 

Figura 2: A significância da interação entre setor, nível de ensino, ano e sexo no salário na Suécia, uma comparação entre diferentes grupos ocupacionais, ano 2014 - 2018

As tabelas com todos os grupos ocupacionais classificados por valor F em ordem decrescente.

merge(summary_table, anova_table, c("ssyk", "interaction"), all = TRUE) %>%
  filter (term.x == "year_n") %>%
  filter (term.y == "sector") %>%   
  filter (interaction == "none") %>%
  mutate (estimate = (exp(estimate) - 1) * 100) %>%  
  select (ssyk, estimate, statistic.y, interaction) %>%
  rename (`F-value` = statistic.y) %>%
  rename (`Increase in salary` = estimate) %>%
  arrange (desc (`F-value`)) %>%
  knitr::kable(
    booktabs = TRUE,
    caption = 'Correlation for F-value (sector) and the yearly increase in salaries')

merge(summary_table, anova_table, c("ssyk", "interaction"), all = TRUE) %>%
  filter (term.x == "year_n") %>%
  filter (contcol.y > 0) %>%   
  filter (interaction == "sector and sex") %>%
  mutate (estimate = (exp(estimate) - 1) * 100) %>%  
  select (ssyk, estimate, statistic.y, interaction) %>%
  rename (`F-value` = statistic.y) %>%
  rename (`Increase in salary` = estimate) %>%
  arrange (desc (`F-value`)) %>%
  knitr::kable(
    booktabs = TRUE,
    caption = 'Correlation for F-value (sector and sex) and the yearly increase in salaries')

merge(summary_table, anova_table, c("ssyk", "interaction"), all = TRUE) %>%
  filter (term.x == "year_n") %>%
  filter (contcol.y > 0) %>%   
  filter (interaction == "sector and edulevel") %>%
  mutate (estimate = (exp(estimate) - 1) * 100) %>%  
  select (ssyk, estimate, statistic.y, interaction) %>%
  rename (`F-value` = statistic.y) %>%
  rename (`Increase in salary` = estimate) %>%
  arrange (desc (`F-value`)) %>%
  knitr::kable(
    booktabs = TRUE,
    caption = 'Correlation for F-value (sector and edulevel) and the yearly increase in salaries')

merge(summary_table, anova_table, c("ssyk", "interaction"), all = TRUE) %>%
  filter (term.x == "year_n") %>%
  filter (contcol.y > 0) %>%   
  filter (interaction == "sector and year") %>%
  mutate (estimate = (exp(estimate) - 1) * 100) %>%  
  select (ssyk, estimate, statistic.y, interaction) %>%
  rename (`F-value` = statistic.y) %>%
  rename (`Increase in salary` = estimate) %>%
  arrange (desc (`F-value`)) %>%
  knitr::kable(
    booktabs = TRUE,
    caption = 'Correlation for F-value (sector and year) and the yearly increase in salaries')

merge(summary_table, anova_table, c("ssyk", "interaction"), all = TRUE) %>%
  filter (term.x == "year_n") %>%
  filter (contcol.y > 1) %>%   
  filter (interaction == "sector, year, edulevel and sex") %>%
  filter (!(contcol.y %  
  mutate (estimate = (exp(estimate) - 1) * 100) %>%  
  select (ssyk, estimate, statistic.y, interaction) %>%
  rename (`F-value` = statistic.y) %>%
  rename (`Increase in salary` = estimate) %>%
  arrange (desc (`F-value`)) %>%
  knitr::kable(
    booktabs = TRUE,
    caption = 'Correlation for F-value (sector, year, edulevel and sex) and the yearly increase in salaries')

Vamos verificar o que encontramos.

temp %
  filter(`occuptional  (SSYK 2012)` == "242 Organisation analysts, policy administrators and human resource specialists")
 
model <- lm (log(salary) ~ year_n + eduyears + sector + sex, data = temp) 
 
plot_model(model, type = "pred", terms = c("eduyears", "sector"))

## Model has log-transformed response. Back-transforming predictions to original response scale. Standard errors are still on the log-scale.

Figura 3: Setor de maior valor F, analistas da organização, administradores de políticas e especialistas em recursos humanos

temp %
  filter(`occuptional  (SSYK 2012)` == "941 Fast-food workers, food preparation assistants")
 
model <-lm (log(salary) ~ year_n + eduyears + sector + sex, data = temp) 
 
plot_model(model, type = "pred", terms = c("eduyears", "sector"))

## Model has log-transformed response. Back-transforming predictions to original response scale. Standard errors are still on the log-scale.

Figura 4: Setor de menor valor F, trabalhadores de fast-food, assistentes de preparação de alimentos

temp %
  filter(`occuptional  (SSYK 2012)` == "911 Cleaners and helpers")
 
model <- lm (log(salary) ~ year_n + eduyears + sector * sex, data = temp) 
 
plot_model(model, type = "pred", terms = c("eduyears", "sex", "sector"))

## Model has log-transformed response. Back-transforming predictions to original response scale. Standard errors are still on the log-scale.

Figura 5: Setor de interação com maior valor F e gênero, Produtos de limpeza e ajudantes

temp %
  filter(`occuptional  (SSYK 2012)` == "411 Office assistants and other secretaries")
 
model <- lm (log(salary) ~ year_n + eduyears + sector * sex, data = temp) 
 
plot_model(model, type = "pred", terms = c("eduyears", "sex", "sector"))

## Model has log-transformed response. Back-transforming predictions to original response scale. Standard errors are still on the log-scale.

Figura 6: Setor de interação com menor valor de F e gênero, assistentes do escritório e outros secretários

Leia Também  Ciência de dados colaborativa: orientação de alto nível para análises éticas de pares científicos

temp %
  filter(`occuptional  (SSYK 2012)` == "335 Tax and related government associate professionals")
 
model <- lm (log(salary) ~ year_n + eduyears * sector + sex, data = temp) 
 
plot_model(model, type = "pred", terms = c("eduyears", "sector"))

## Warning in predict.lm(model, newdata = fitfram, type = "response", se.fit =
## se, : prediction from a rank-deficient fit may be misleading

## Model has log-transformed response. Back-transforming predictions to original response scale. Standard errors are still on the log-scale.

Figura 7: Setor de interação com maior valor F e nível profissional, profissionais associados a impostos e governos relacionados

temp %
  filter(`occuptional  (SSYK 2012)` == "911 Cleaners and helpers")
 
model <- lm (log(salary) ~ year_n + eduyears * sector + sex, data = temp) 
 
plot_model(model, type = "pred", terms = c("eduyears", "sector"))

## Warning in predict.lm(model, newdata = fitfram, type = "response", se.fit =
## se, : prediction from a rank-deficient fit may be misleading

## Model has log-transformed response. Back-transforming predictions to original response scale. Standard errors are still on the log-scale.

Figura 8: Setor de interação de menor valor F e nível de nível, Produtos de limpeza e ajudantes

temp %
  filter(`occuptional  (SSYK 2012)` == "129 Administration and service managers not elsewhere classified")
 
model <- lm (log(salary) ~ year_n * sector + eduyears + sex, data = temp) 
 
plot_model(model, type = "pred", terms = c("eduyears", "year_n", "sector"))

## Model has log-transformed response. Back-transforming predictions to original response scale. Standard errors are still on the log-scale.

Figura 9: Setor de interação com valor F mais alto e ano, gerentes de administração e serviços não classificados em outra parte

temp %
  filter(`occuptional  (SSYK 2012)` == "159 Other social services managers")
 
model <- lm (log(salary) ~ year_n * sector + eduyears + sex, data = temp) 
 
plot_model(model, type = "pred", terms = c("eduyears", "year_n", "sector"))

## Model has log-transformed response. Back-transforming predictions to original response scale. Standard errors are still on the log-scale.

Figura 10: Setor de interação com menor valor F e ano, Outros gerentes de serviços sociais

temp %
  filter(`occuptional  (SSYK 2012)` == "264 Authors, journalists and linguists")
 
model <- lm (log(salary) ~ year_n * eduyears * sector * sex, data = temp) 
 
plot_model(model, type = "pred", terms = c("eduyears", "year_n", "sex", "sector"))

## Warning in predict.lm(model, newdata = fitfram, type = "response", se.fit =
## se, : prediction from a rank-deficient fit may be misleading

## Model has log-transformed response. Back-transforming predictions to original response scale. Standard errors are still on the log-scale.

Figura 11: Setor de interação com valor F mais alto, nível de ensino, ano e gênero, Autores, jornalistas e linguistas

## TableGrob (2 x 1) "arrange": 2 grobs
##   z     cells    name           grob
## 1 1 (1-1,1-1) arrange gtable[layout]
## 2 2 (2-2,1-1) arrange gtable[layout]

temp %
  filter(`occuptional  (SSYK 2012)` == "534 Attendants, personal assistants and related workers")
 
model <- lm (log(salary) ~ year_n * eduyears * sector * sex, data = temp) 
 
plot_model(model, type = "pred", terms = c("eduyears", "year_n", "sex", "sector"))

## Warning in predict.lm(model, newdata = fitfram, type = "response", se.fit =
## se, : prediction from a rank-deficient fit may be misleading

## Model has log-transformed response. Back-transforming predictions to original response scale. Standard errors are still on the log-scale.

Figura 12: Setor de interação de menor valor F, nível de ensino, ano e gênero, Atendentes, assistentes pessoais e trabalhadores relacionados

## TableGrob (2 x 1) "arrange": 2 grobs
##   z     cells    name           grob
## 1 1 (1-1,1-1) arrange gtable[layout]
## 2 2 (2-2,1-1) arrange gtable[layout]

var vglnk = {key: '949efb41171ac6ec1bf7f206d57e90b8'};

(função (d, t) {
var s = d.createElement
s.src = '//cdn.viglink.com/api/vglnk.js';
var r = d.getElementsByTagName
} (documento, 'script'));