scatterplots_interactive.Rmd

---
title: "OA percentage of German reserach institutions by sector"
author: "Anne Hobert"
output:
  html_document:
    fig_caption: yes
    keep_md : true
  urlcolor: blue
---

```{r, echo = FALSE, message = FALSE, warning = FALSE}
knitr::opts_chunk$set(
  comment = "#>",
  collapse = TRUE,
  warning = FALSE,
  message = FALSE,
  echo = FALSE,
  fig.width = 6,
  fig.asp = 0.618,
  out.width = "90%",
  fig.align = "center",
  dpi = 300
)
```

```{r}
library(tidyverse)
#library(urltools)
library(cowplot)
library(colorblindr)
library(scales)
library(viridis)
```


```{r}
pubs_cat <- readr::read_csv("data/pubs_cat.csv", col_types = "dccdcclc")
pubs_cat <- pubs_cat %>%
  mutate(sec_abbr = case_when(
    sector == "Hochschulen" ~ "UNI",
    sector == "Helmholtz-Gemeinschaft" ~ "HGF",
    sector == "Max-Planck-Gesellschaft" ~ "MPG",
    sector == "Leibniz-Gemeinschaft" ~ "WGL",
    sector == "Fraunhofer-Gesellschaft" ~ "FhG",
    sector == "Ressortforschung" ~ "ReF"
  ),
  sector = case_when(
    sector == "Hochschulen" ~ "Universities",
    sector == "Helmholtz-Gemeinschaft" ~ "Helmholtz Association",
    sector == "Max-Planck-Gesellschaft" ~ "Max Planck Society",
    sector == "Leibniz-Gemeinschaft" ~ "Leibniz Association",
    sector == "Fraunhofer-Gesellschaft" ~ "Fraunhofer Society",
    sector == "Ressortforschung" ~ "Ressortforschung"
  ),
  oa_category = factor(
    oa_category,
    levels = c(
      "full_oa_journal",
      "other_oa_journal",
      "opendoar_inst",
      "opendoar_subject",
      "opendoar_other",
      "other_repo",
      "not_oa"
    )
  )
  )
```


```{r}
## Exlcude institutions from university sector which are not listed in official statistics
excl_from_analysis <- readr::read_csv("data/exclude_from_analysis.csv")
pubs_cat <- pubs_cat %>%
  anti_join(excl_from_analysis, by = c("INST_NAME" = "Universitäten"))
rm(excl_from_analysis)
```


```{r}
## exclude pubs with less than 100 pubs and administrative institutes, aggregating, residual categories
exclude_from_inst_analysis <- readr::read_csv("data/exclude_from_inst_level_analysis.csv")
oa_shares_inst <- pubs_cat %>%
  anti_join(exclude_from_inst_analysis, by = c("INST_NAME" = "NAME")) %>%
  mutate(oa_category = fct_collapse(
    oa_category,
    is_oa = c(
      "full_oa_journal",
      "other_oa_journal",
      "opendoar_inst",
      "opendoar_subject",
      "opendoar_other",
      "other_repo"
    ),
    not_oa = "not_oa"
  )) %>%
  group_by(INST_NAME) %>%
  mutate(n_total = n_distinct(PK_ITEMS)) %>%
  ungroup() %>%
  group_by(INST_NAME, oa_category, n_total) %>%
  summarise(n_cat = n_distinct(PK_ITEMS)) %>%
  pivot_wider(
    names_from = oa_category,
    values_from = n_cat,
    values_fill = list(n_cat = 0)
  ) %>%
  mutate(oa_share = is_oa / n_total) %>%
  rename(n_oa = is_oa, n_not_oa = not_oa) %>%
  ungroup()

oa_shares_inst_sector <- pubs_cat %>%
  anti_join(exclude_from_inst_analysis, by = c("INST_NAME" = "NAME")) %>%
  mutate(oa_category = fct_collapse(
    oa_category,
    journal_oa = c("full_oa_journal", "other_oa_journal"),
    repo_oa = c(
      "opendoar_inst",
      "opendoar_subject",
      "opendoar_other",
      "other_repo"
    ),
    not_oa = "not_oa"
  )) %>%
  group_by(sector) %>%
  mutate(n_total_sec = n_distinct(PK_ITEMS)) %>%
  ungroup() %>%
  left_join(oa_shares_inst, by = 'INST_NAME') %>%
  group_by(sector, INST_NAME, n_total, oa_share, n_total_sec) %>%
  summarise()

oa_shares_inst_sector_stats <- oa_shares_inst_sector %>%
  ungroup() %>%
  filter(n_total >= 100) %>%
  group_by(sector) %>%
  summarise(mean_oa_share = mean(oa_share),
            median_oa_share = median(oa_share),
            mean_pub_volume = mean(n_total),
            median_pub_volume = median(n_total),
            sd_oa_share = sd(oa_share),
            sd_pub_volume = sd(n_total)
  )
```


```{r, fig.asp=1, fig.cap="Open Access shares of research institutions in Germany with respect to their total publication output grouped by the sector they belong to. Only institutions with at least 100 publications are shown. Blue points correspond to single insitutions, gray lines are obained by linear regression within the sector, gray areas are pointwise symmetric 95% t-distribution confidence bands. Scales of the x-axes vary across subplots in order to adapt to the different publication volumes. Dashed lines show the median value per sector for the OA share (red) and the total number of publications (orange). Tooltip shows the explicit numbers for each institution as well as their names."}
oa_shares_inst_sector %>%
  filter(n_total >= 100) %>%
  left_join(oa_shares_inst_sector_stats, by = "sector") %>%
  ggplot(aes(x = n_total, y = oa_share, label = INST_NAME)) +
  geom_point(color = "#56b4e9", alpha = .7)  +
  scale_x_log10() +
  scale_y_continuous(labels = scales::percent_format(accuracy = 5L),
                     expand = expansion(mult = c(0, 0.05)),
                     limits = c(0,1)) +
  geom_smooth(color = "#999999a0", method = "lm") +
  facet_wrap(~ fct_rev(fct_reorder(sector, n_total_sec)),
             ncol = 2,
             scales = "free_x") +
  geom_hline(aes(yintercept = median_oa_share),
             colour = "#d55e00", linetype ="dashed", size = 1) +
  geom_vline(aes(xintercept = median_pub_volume),
             colour = "#E69F00", linetype ="dashed", size = 1) +
  labs(x = "Total Articles (logarithmic scale)", y = "OA percentage") +
  theme_minimal_grid() +
  theme(legend.position = "none") +
  # bold facet labels
  theme(strip.text = element_text(face = "bold")) -> plot_oa_inst_sec
plotly::ggplotly(plot_oa_inst_sec, tooltip = c("label", "y", "x"))
```