For this section we will be using data from the ‘British Election Study’ survey to investigate the effect of perceptions about the economy on party choice. This is part of a broader research examining the effect of political issues on party choice and especialy on the effect of the economy on the party in government.
load("bes.RData")We will analyse data from the 2017 general election, and we will test whether citizens’ who feel that the economy is doing better will support the party in government.
We can download the full dataset from the website below:
We should download the post-election study entitled: 2017 Face-to-face Post-election Survey Version 1.3
We will use the following variables from the dataset:
| Name | Description (name in the dataset in parenthesis) |
|---|---|
| incumbent | (1) The respondent will vote for the party in government (0) will vote for another party (b02) |
| left-right | position on the left- right spectrum (e01) |
| egocentric_retro | retrospective evaluation of personal finances (l01) |
| egocentric_pros | prospective evaluation of personal finances (l02) |
| sociotropic_retro | retrospective evaluation of UK’s economy (l04) |
| sociotropic_pros | prospective evaluation of UK’s economy (l03) |
Our next step is to create a sub-set including only the variables we will include in our analysis
library(dplyr)
eco_voting<-bes %>%select(b02,Age,edlevel,e01,l01,l02,l03,l04,p01)
View(eco_voting)Note: If you use RStudio, you can type the pipe (\(%>%\)) with Ctrl + Shift + M if you have a PC or Cmd + Shift + M if you have a Mac.
Our next step is to replace the negative values (-1, -2) to NAs
eco_voting[eco_voting <=-1] <- NA
eco_voting[eco_voting <=-2] <- NAWe will use the rename() function which is part dplyr:
eco_voting<-eco_voting %>% rename(left_right=e01, egocentric_retro=l01,egocentric_pros=l02,sociotropic_pros=l03,sociotropic_retro=l04,brexit=p01)
View(eco_voting)