7CCSMSDV Data Visualisation Project 2024

King's College London

Divyani Panda

k23033766@kcl.ac.uk

Get Started

Introduction

Track: House Price Analysis

The focus of this track is to analyse the house price data of the 32 London boroughs and the City of London, between January 2010 to December 2023.
These visualisations aim to answer the following research questions:

  • Question 1: Analyse the development of house prices over time. Are there any detectable trends?
    • Visualisation 1:A choropleth map of the London boroughs with their average house prices over the years
    • Visualisation 2:A dual axes line chart displaying the trends of the overall London average house prices and its affecting factors (Population Density, Crime, Average Household Income) over the years
  • Question 2: Analyse the effect of demographic data on house prices. Are there any specific segments of the population (in terms of their average income) driving housing demand?
    • Visualisation 1:An overlay on the choropleth map with a color bar (ranging from the minimum to the maximum population density in any given year) to demonstrate relation between average housing price and population density (one of the demographic factors).
      This interactive choropleth allows users to toggle between the two overlays for population density and crime statistics as described above, in order to have a generic idea of the relation between average house prices and the two factors.
    • Visualisation 2:A dual axis line chart displaying the trends of the overall London average house prices and its affecting factors (Population Density, Crime, Average Household Income) over the years.
    • Visualisation 3:An annulus chart displaying the individual correlation values of average house prices with population density, crime statistics and average household income.
      This shows that out of the two demographic factors, average house prices has a weak positive correlation with population density, and a strong positive correlation with average household income.
  • Question 3: Analyse and compare the impact of crime rates on house prices in a given area.
    • Visualisation 1: An overlay on the choropleth map using the channels of shapes and colors have been used to show crime statistics in the boroughs which would provide the users an idea of how the housing price trends are based upon crime rates. A threshold of 50 crimes have been utilized, where boroughs having 50 or more crimes committed are marked with red danger signs, and the safe boroughs having less than 50 crimes are marked with a green shield symbol.
      This interactive choropleth allows users to toggle between the two overlays for population density and crime statistics as described above, in order to have a generic idea of the relation between average house prices and the two factors.
    • Visualisation 2:A dual axis line chart displaying the trends of the overall London average house prices and its affecting factors (Population Density, Crime, Average Household Income) over the years.
    • Visualisation 3:An annulus chart displaying the individual correlation values of average house prices with population density, crime statistics and average household income.
      This shows that average house prices has a weak positive correlation with the crime statistics factor.

References
  • Website
    • Website template: https://bootstrapmade.com/free-one-page-bootstrap-template-amoeba/
  • Visualisation 1
    • Choropleth map-reference 1: https://gist.github.com/eetuko/4535086c3fabe76a173b432c44b254c6
    • Choropleth map-reference 2: Jonathan Soma (https://www.youtube.com/@jsoma): https://drive.google.com/drive/folders/1c9ZcQUV1w0rBR-NbbyplIJPL9pVAl2MG
    • Symbols on maps: https://observablehq.com/@bradydowling/u-s-state-capitals
  • Visualisation 2
    • Line chart: https://d3-graph-gallery.com/graph/line_filter.html
  • Visualisation 3
    • Annulus/ Donut chart: https://d3-graph-gallery.com/graph/donut_basic.html
  • Datasets
    • Average house prices: https://www.gov.uk/government/publications/about-the-uk-house-price-index/about-the-uk-house-price-index#data-tables?utm_medium=GOV.UK&utm_source=datadownload&utm_campaign=data_tables&utm_term=9.30_21_03_17
    • Population density: https://data.london.gov.uk/dataset/land-area-and-population-density-ward-and-borough?resource=77e9257d-ad9d-47aa-aeed-59a00741f301
    • Crime statistics: https://data.london.gov.uk/dataset/recorded_crime_summary
    • Average household income: https://www.gov.uk/government/statistics/income-and-tax-by-borough-and-district-or-unitary-authority-2010-to-2011

Average House Prices Choropleth Map

Below is a choropleth map of the 32 London boroughs and the City of London, depicting the average house prices over the course of 10 years between 2010 and 2023. Darker shades respresent higher price ranges.
Inference: We can see that Westminster has been consistently offering housing for higher prices even though it records the highest crime numbers. Although along with housing prices, both population density and crime rates are seen to increase over the years, there is no significant affect of either on the average house prices of London boroughs.

Factors affecting Average House Prices

Below are dual axes line charts showing the trends of average house prices in London over the years, with each of the three factors (Population density, Crime statistics and Average household income).
Inference: We can see positive trends of average house prices over the years. Positive trends are seen in the other three factors as well. Average household income seems to be having more impact on the increase in house prices across London.


Correlation between Average House Prices and its affecting Factors
(Population Density, Crime Stats, Average Household Income)

The annulus chart visualisation displays the correlation coefficients of average house prices with various soci-economic factors, which essentially illustrates the strength and direction of the linear relationship between pairs of variables.
Inference: All three factors have positive correlation with average house prices, showing that with increase in the factors, there is increase in the average house prices. For average household income, the correlation coefficient of 0.9 indicates a very strong positive correlation, suggesting that areas with higher average income tend to have higher average house prices. With a correlation coefficient of 0.46, crime statistics has a moderate positive correlation with house prices, indicating that while there is correlation, it does not necessarily imply causation as we have seen in the other visualisations as well. Lastly, population density having the highest correlation coefficient of 0.96 suggests that changes in population density are closely associated with changes in house prices.