R Notebook

```r
rm(list = ls())

Sys.setenv(LANG = \en\)
library(xlsx)
library(readr)
library(readxl)
library(tidyverse)
library(stringr)
library(stringi)
library(gghighlight)
library(countrycode)
library(countries)
library(janitor)
library(sf)

library(janitor)
library(sf)
library(ggthemes)
library(here)
library(viridis)
library(ggplot2)
library(formatR)
library(gtools)
library(nnet) 
library(patchwork)
library(ggh4x)

# Packages for the regression. Move to different script.
#install.packages(\plm\) # a package of \panel data econometrics in R\ (by Croissant and Millo)
#install.packages(\lme4\) # this is the first multilevel (random effects) package we'll use
library(plm)
library(broom)
library(ggalluvial)

library(pals)

<!-- rnb-source-end -->

<!-- rnb-chunk-end -->


<!-- rnb-text-begin -->


# Load data ---------------------------------------------------------------

## Corpus.

<!-- rnb-text-end -->


<!-- rnb-chunk-begin -->


<!-- rnb-source-begin eyJkYXRhIjoiYGBgclxuYGBgclxuc2Nfc2NvcmVzIDwtIHJlYWRfY3N2KFxcLi4vYWdncmVnYXRlZF9kYXRhL3NjX3Njb3Jlcy5jc3ZcXCkgXG5cbnNjX3Njb3JlcyAlPiVcbiAgICBnZ3Bsb3QoKSArIFxuICAgIGdlb21fYm94cGxvdChhZXMoeCA9IG1pZ19qb3VybmFsLCB5ID0gc2NvcmUsIGNvbG9yID0gYXMuZmFjdG9yKG1pZ19qb3VybmFsKSkpICtcbiAgICB0aGVtZShheGlzLnRleHQueD1lbGVtZW50X2JsYW5rKCkpICtcbiAgICBmYWNldF93cmFwKHZhcnMoY29ycHVzX2Nfc2QxKSxcbiAgICAgICAgICAgICAgIGxhYmVsbGVyID0gbGFiZWxsZXIoY29ycHVzX2Nfc2QxID0gYyhcXFRSVUVcXCA9IFxcSW5jbHVkZWQgaW4gY29ycHVzOiBcXG4gPD0gMVNEIGF2ZXJhZ2Ugc2NvcmUgb2YgbWlncmF0aW9uIGpvdXJuYWxzXFwsIFxcRkFMU0VcXCA9IFxcTm90IGluY2x1ZGVkIGluIGNvcnB1czogXFxuID4gMVNEIGF2ZXJhZ2Ugc2NvcmUgb2YgbWlncmF0aW9uIGpvdXJuYWxzXFwpKSkgK1xuICBsYWJzKGNvbG9yID0gXFxKb3VybmFsIGNsYXNzaWZpY2F0aW9uXFwsXG4gICAgICAgICB4ID0gXFxcXCxcbiAgICAgICAgIHkgPSBcXE1pZ3JhdGlvbiByZWxhdGlvbiBzY29yZVxcKSArXG4gIHNjYWxlX2NvbG91cl9kaXNjcmV0ZShsYWJlbHMgPSBjKFxcT3RoZXIgam91cm5hbHNcXCwgXFxNaWdyYXRpb24gam91cm5hbHNcXCkpICtcbiAgdGhlbWVfbWluaW1hbCgpICtcbiAgdGhlbWUoYXhpcy50aXRsZS54PWVsZW1lbnRfYmxhbmsoKSxcbiAgICAgICAgYXhpcy50ZXh0Lng9ZWxlbWVudF9ibGFuaygpLFxuICAgICAgICBheGlzLnRpY2tzLng9ZWxlbWVudF9ibGFuaygpLFxuICAgICAgICBsZWdlbmQucG9zaXRpb24gPSBcXGJvdHRvbVxcKVxuXG5nZ3NhdmUoXFwuLi9vdXRwdXRfZmlndXJlcy9ib3hfc2NvcmVfbWlnX2pvdXJuYWxfc2QxLnBuZ1xcLCB3aWR0aCA9IDcsIGhlaWdodCA9IDcsIGRwaSA9IDUwMCwgbGltaXRzaXplID0gRiwgcGxvdCA9IGxhc3RfcGxvdCgpKVxuYGBgXG5gYGAifQ== -->

```r
```r
sc_scores <- read_csv(\../aggregated_data/sc_scores.csv\) 

sc_scores %>%
    ggplot() + 
    geom_boxplot(aes(x = mig_journal, y = score, color = as.factor(mig_journal))) +
    theme(axis.text.x=element_blank()) +
    facet_wrap(vars(corpus_c_sd1),
               labeller = labeller(corpus_c_sd1 = c(\TRUE\ = \Included in corpus: \n <= 1SD average score of migration journals\, \FALSE\ = \Not included in corpus: \n > 1SD average score of migration journals\))) +
  labs(color = \Journal classification\,
         x = \\,
         y = \Migration relation score\) +
  scale_colour_discrete(labels = c(\Other journals\, \Migration journals\)) +
  theme_minimal() +
  theme(axis.title.x=element_blank(),
        axis.text.x=element_blank(),
        axis.ticks.x=element_blank(),
        legend.position = \bottom\)

ggsave(\../output_figures/box_score_mig_journal_sd1.png\, width = 7, height = 7, dpi = 500, limitsize = F, plot = last_plot())

```