Visualising ESS data using ggplot2 and tmap

I stumbled across rOpenSci’s package essurvey that contains data from the European Social Survey (ESS) for European countries, including Lithuania! Also, tmap has a lot of excellent resources to get a head start, so this post shows how to access ESS data and visualise it with tmap.

Downloading ESS data

Downloading ESS data is fairly straightforward - register your email in the ESS and start pulling data! I recommend installing the development version because the stable version was downloading .zip files that Windows couldn’t extract. To replicate this post, you should use:

import_country("Lithuania", 9)

I didn’t want to disclose my email here, so I have downloaded the data separately.

Visualising with tmap

All plotting packages depend on shapefiles. For Europe, it’s best to download the map from GISCO. However, the map does not have municipality data for Lithuania. Fortunately, the shapefile is available over the OpenMap project.

This is how to build a basic tmap plot:

lt_map <- read_sf("data/admin_ribos.shp")

lt_map_sen <- lt_map %>% filter(ADMIN_LEVE == 4)

news <- ess %>%
  group_by(region = labelled::to_factor(region)) %>%
  summarise(`Internet usage` = mean(netustm, na.rm = TRUE)) %>%
  mutate(region = as.character(region)) %>%
  mutate(region = case_when(region == "Klaipedos apskritis" ~ "Klaipėdos apskritis",
                            region == "Šiauliu apskritis" ~ "Šiaulių apskritis",
                            region == "Taurages apskritis" ~ "Tauragės apskritis",
                            region == "Telšiu apskritis" ~ "Telšių apskritis",
                            TRUE ~ region))

lt_plot <- lt_map_sen %>% left_join(news, by = c("NAME" = "region"))

tm_shape(lt_plot) +
  tm_fill("Internet usage") +
  tm_polygons()