README.Rmd

---
output: github_document
---

<!-- README.md is generated from README.Rmd. Please edit that file -->

```{r, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "",
  echo = FALSE,
  warning = FALSE,
  message = FALSE,
  fig.retina = 3
)
```

```{r, include = FALSE}
source("_targets_packages.R")
```

```{r, echo = FALSE}
targets::tar_load(syst_all)
targets::tar_load(syst_pocty_long)
targets::tar_load(orgdata_graph)
targets::tar_load(orgdata_nodes_processed)
targets::tar_load(orgdata_edges_processed)
targets::tar_load(orgdata_rect)
```

```{r, include=FALSE}
cnf <- config::get()
nms_orig <- names(cnf)
names(cnf) <- paste0("c_", names(cnf))
list2env(cnf, envir = .GlobalEnv)
names(cnf) <- names(nms_orig)
rm(nms_orig)
```


# systemizace

<!-- badges: start -->
<!-- badges: end -->

This repository contains R code to download and process official data on the organisation of public servants in the Czech central public administration - "systemizace služebních míst".

It works with two sources of data:

1. The official tables
2. Organisational structures published as [open data](https://data.gov.cz/datov%C3%A1-sada?iri=https%3A%2F%2Fdata.gov.cz%2Fzdroj%2Fdatov%C3%A9-sady%2F00007064%2F846439662)

### Official tables

The code downloads, loads and processes the published excel files such that

- the data is in tidy format
- civil servants and employees are in one data frame
- pay grades (platové třídy) and management levels (představení) are correctly identified

See https://www.mvcr.cz/sluzba/clanek/systemizace-sluzebnich-a-pracovnich-mist.aspx. For [April 2022 edition](https://www.mvcr.cz/sluzba/clanek/zmena-systemizace-sluzebnich-a-pracovnich-mist-s-ucinnosti-od-1-dubna-2022.aspx), only PDF table is provided.

Data currently available from `r min(syst_all$rok)` to `r max(syst_all$rok)`.

#### Sample

```{r}
syst_pocty_long |>
  dplyr::filter(ustredni_organ, kapitola_vladni) |>
  mutate(trida = as.integer(trida)) |>
  group_by(rok, kapitola_zkr, level_nazev) |>
  drop_na(pocet) |>
  dplyr::filter(date == "2022-04-01") |>
  summarise(trida_mean = weighted.mean(trida, w = pocet, na.rm = TRUE), 
            .groups = "drop") |>
  mutate(kapitola_zkr = as.factor(kapitola_zkr) |> fct_reorder(trida_mean, min)) |>
  ggplot(aes(trida_mean, kapitola_zkr, colour = level_nazev)) +
  geom_point(size = 3) +
  scale_color_manual(name = "Úroveň řízení", values = c("grey50", "darkblue")) +
  scale_x_continuous(limits = c(10, 15)) +
  ptrr::theme_ptrr("both", multiplot = FALSE, legend.position = "top") +
  labs(title = "Průměrná platová třída (2022)",
       subtitle = "Podle systemizace 2022.\nJen ústřední orgány ve vládních kapitolách",
       caption = "Plánovaný stav podle systemizace 2022")
```


Wide format

```{r}
syst_all
```

Long format - staff numbers only

```{r}
syst_pocty_long
```



### Organisational structures

These are now in a [separate repository](https://github.com/petrbouchal/mapa-statu/).

### Technical: how to run this

Code organised as a {targets} pipeline, with packages tracked by {renv}, so can be reproduced like so:

```r
renv::restore()
targets::tar_make()
```

Tidy data ready for analysis are in `data-export`. Codebook is TBA.

Rendered `r Sys.time()`