Mini-project

Aim

To create a one-page report that summarises the relationship between certain socio-demographic variables and EU referendum voting patterns. You can find an example report here.

Workflow:

1. Import: download and read data
2. Tidy: clean data for analysis in R
3. Transform: create new variables and merge different data sets
4. Visualise: create a set of scatterplots
5. Report: put together a summary document of your findings

Import

1. Open a new R script.
2. Load the tidyverse package
3. Download and read EU referendum results by Local Authority from the Electoral Commission https://www.electoralcommission.org.uk/find-information-by-subject/elections-and-referendums/past-elections-and-referendums/eu-referendum/electorate-and-count-information.
4. Read the variables.csv file which contains socio-demographic data and a census code for each Local Authority area. The data derive from the 2011 Census via the UK Data Service and the Annual Survey of Hours and Earnings (2016) by Local Authority area.

There are 7 different variables:

• degree: % residents with higher education
• pay: median gross annual pay (full time)
• percent_aged_18_30: % residents aged 18-30 years
• median_age: Median age of residents
• non_uk: % residents not born in the UK
• no_passport: % of residents without a passport
• ABC1: % residents in professional occupations

Tidy

1. Subset the referendum data frame to only include the following columns:
   • Area_Code
   • Area
   • Electorate
   • Pct_Remain
   • Pct_Leave
2. Filter the referendum data frame to exclude “Northern Ireland” and “Gibraltar”. We have no socio-demographic data for these areas.

Transform

1. Create a new variable in the referendum data frame called Result which records the outcome of the vote in each Local Authority area. Tip: use an ifelse() statement.
2. Transform Area_Code in the variables data frame into a factor.
3. Transform degree and percent_aged_18_30 in the variables data frame into percentages.
4. Create a new data frame by merging the variables data frame with the referendum data frame.

Visualise

1. Create a scatterplot using the ggplot2 package showing the relationship between the share of the Leave vote with the percentage of residents having a degree.
2. Create a scatterplot showing the relationship between the share of the Leave vote and a different socio-demographic indicator. What is the strength of the relationship?
3. Use the plotly R package to make your scatterplot interactive.

Report

1. Create a new R Markdown document.
2. Paste all of your code into separate code chunks.
3. Interpret your results in Markdown.
4. knit your R Markdown document to HTML.
5. Optional: Publish your report on rpubs.com