search

CampaignLab / data-pipeline

Scripts and schemas that aim to make data from the inventory easier to analyse
8 stars 8 forks source link

Building an initial model #35

Open hannah-o-rourke opened 5 years ago

hannah-o-rourke commented 5 years ago

The challenge is to build a model to correlate the changes in party vote share at ward level (between the 2014 and 2018 local elections) with ward level demographic data from the census. If you have time you can also build in historic election results.

Use this model to identify the wards that are significant outliers.

Data sources

  1. 2014 & 2018 Election results use this to work out the change in party vote share - explore here for data sets: https://github.com/CampaignLab/data-pipeline or ask someone who has been before.
  2. Ward level demographic census data - use this tool to find different demographic datasets: http://infuse2011.ukdataservice.ac.uk/InFuseWiz.aspx?cookie=openaccess Public health data at ward level can be found here: https://github.com/CampaignLab/healthcare/blob/master/localhealth/health_data_merged_with_election_results.csv
  3. Historic election results: https://www.andrewteale.me.uk/leap/downloads