EDA Chamber decisions (gessdoc300796s in Konflikt stehende Kopie 2019-09-05).Rmd

---
title: "EDA Chamber Decisions GFCC"
author: "Sebastian Sternberg"
date: "2 September 2019"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```



Load data and packages

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

require(magrittr)
require(dplyr)
require(ggplot2)
require(lubridate)

#load data

load("chamber_decision_cleaned.Rda")

df <- chamber_decisions

#fill NAs of kammer_txt (with kammer of citation)

df$kammer_txt[is.na(df$kammer_txt)] <- df$kammer_cit[is.na(df$kammer_txt)]

#some NAs in the AZ:

df$link <- as.character(df$link)

sum(grep("[0-9]bvr", df$link))

grepl("[0-9]bvr[0-9]+", df$link[is.na(df$senat)])



http:https://www.bverfg.de/e/rk19990315_2bvr037599.html

sapply(df, function(y) sum(length(which(is.na(y)))))

```


# Summarize by date

### Summarize by year, month, week day

```{r}

df$year <- year(df$date)
df$month <- month(df$date)
df$weekday <- wday(df$date, label = T)

```


```{r}

df_year <- df %>% group_by(year) %>% tally()
df_year_senat <- df %>% group_by(year, senat) %>% tally()
df_year_senat_chamber <- df %>% group_by(year, senat, kammer_txt) %>% tally()



#in one call
ggplot(tally(group_by(df, year, senat)),
    aes(x = year, y = n, fill = senat)) +
    geom_bar(stat="identity") + labs(fill="Employment")


ggplot(df_year, aes(x = year, y = n)) +
    geom_bar(stat="identity")


#year by Senate
ggplot(df_year_senat, aes(x = year, y = n, fill = senat)) +
    geom_bar(stat="identity") + labs(fill="Senate")



#year by Senate chamber


df_senat1 <- df %>% filter(senat == 1)
df_senat2 <- df %>% filter(senat == 2)

#Senate 1 by chamber

ggplot(tally(group_by(df_senat1, year, kammer_txt), format = "percent"),
    aes(x = year, y = n, fill = kammer_txt)) +
    geom_bar(stat="identity") + labs(fill="Chamber")

ggplot(tally(group_by(df_senat2, year, kammer_xt)),
    aes(x = year, y = n, fill = kammer_txt)) +
    geom_bar(stat="identity") + labs(fill="Chamber")

#for each Senate Chamber separately



df %>% group_by()

count_pct <- function(df) {
  return(
    df %>%
      tally %>% 
      mutate(n_pct = 100*n/sum(n))
  )
}

```