Overview

The package offers functions that calculate coalition probabilities in multi-party elections, based on a survey results. It offers two main functions:

  • get_surveys: Scrapes surveys on German Bundestag elections from wahlrecht.de
  • get_probabilities: Calculates coalition probabilities based on provided survey results.

Installation

Install from GitHub using:

devtools::install_github("adibender/coalitions", build_vignettes=TRUE)

Usage

Detailed workflow is outlined in the vignettes enclosed in the package (see browseVignettes(package="coalitions")).

A short overview is presented below.

Scrape surveys

The wrapper get_surveys() which takes no arguments, downloads all surveys currently available at wahlrecht.de and stores them in a nested tibble:

library(coalitions)
library(dplyr)
library(tidyr)
surveys <- get_surveys()
surveys
## # A tibble: 7 x 2
##   pollster   surveys           
##   <chr>      <list>            
## 1 allensbach <tibble [38 × 5]> 
## 2 emnid      <tibble [212 × 5]>
## 3 forsa      <tibble [218 × 5]>
## 4 fgw        <tibble [79 × 5]> 
## 5 gms        <tibble [93 × 5]> 
## 6 infratest  <tibble [102 × 5]>
## 7 insa       <tibble [290 × 5]>

Each row represents a survey insitute and each row in the surveys column again contains a nested tibble with survey results from different time-points:

surveys %>%
    filter(pollster == "allensbach") %>%
    unnest()
## # A tibble: 38 x 6
##    pollster   date       start      end        respondents survey         
##    <chr>      <date>     <date>     <date>           <dbl> <list>         
##  1 allensbach 2017-10-25 2017-10-07 2017-10-19        1454 <tibble [7 × 3…
##  2 allensbach 2017-09-22 2017-09-13 2017-09-20        1074 <tibble [7 × 3…
##  3 allensbach 2017-09-19 2017-09-06 2017-09-14        1083 <tibble [7 × 3…
##  4 allensbach 2017-09-06 2017-08-22 2017-08-31        1043 <tibble [7 × 3…
##  5 allensbach 2017-08-22 2017-08-04 2017-08-17        1421 <tibble [7 × 3…
##  6 allensbach 2017-07-18 2017-07-01 2017-07-12        1403 <tibble [7 × 3…
##  7 allensbach 2017-06-20 2017-06-01 2017-06-15        1437 <tibble [7 × 3…
##  8 allensbach 2017-05-26 2017-05-05 2017-05-19        1457 <tibble [7 × 3…
##  9 allensbach 2017-04-25 2017-04-01 2017-04-13        1407 <tibble [7 × 3…
## 10 allensbach 2017-03-28 2017-03-06 2017-03-19        1397 <tibble [7 × 3…
## # ... with 28 more rows
survey <- surveys %>% unnest() %>% slice(1)
survey %>% unnest()
## # A tibble: 7 x 8
##   pollster   date       start      end        responde… party perce… votes
##   <chr>      <date>     <date>     <date>         <dbl> <chr>  <dbl> <dbl>
## 1 allensbach 2017-10-25 2017-10-07 2017-10-19      1454 cdu    33.0  480  
## 2 allensbach 2017-10-25 2017-10-07 2017-10-19      1454 spd    20.5  298  
## 3 allensbach 2017-10-25 2017-10-07 2017-10-19      1454 gree…   9.50 138  
## 4 allensbach 2017-10-25 2017-10-07 2017-10-19      1454 fdp    12.0  174  
## 5 allensbach 2017-10-25 2017-10-07 2017-10-19      1454 left    9.00 131  
## 6 allensbach 2017-10-25 2017-10-07 2017-10-19      1454 afd    12.0  174  
## 7 allensbach 2017-10-25 2017-10-07 2017-10-19      1454 othe…   4.00  58.2

Calculate coalition probabilities

For each survey (row) we can calculate the coalition probabilities

survey %>% get_probabilities(nsim=1e4) %>% unnest()
## # A tibble: 6 x 4
##   pollster   date       coalition       probability
##   <chr>      <date>     <chr>                 <dbl>
## 1 allensbach 2017-10-25 cdu                   0    
## 2 allensbach 2017-10-25 cdu_fdp               0.840
## 3 allensbach 2017-10-25 cdu_fdp_greens       99.2  
## 4 allensbach 2017-10-25 spd                   0    
## 5 allensbach 2017-10-25 left_spd              0    
## 6 allensbach 2017-10-25 greens_left_spd       0