New York Election Data 2010


Visualization

Introduction

For this lab, I had an interest in election data. Through research, I found the Harvard Election Data Archive. The archive contains data on election results, voting behaviour, and electoral politics, with a particular focus on the United States. The core data for the archive are state, county and district level election returns for all recent state and federal elections in the United States. The data I chose to focus on for this lab is New York election data from 2010.

Visualisation Examples

These examples are standard United States election data visualisations. The New York examples show an excellent shapefile and depiction of the uninhabited/uncounted areas as grey.

Materials

Software:

  • Carto

Research resources used in this lab were:

  • Google
  • Carto Tutorials
  • Harvard Election Data Archive

Methods

  1. Find a data source
    1. Harvard Election Data Archive
  2. Download data
    1. NY_Shapefile.zip
      1. This file contained the shapefile and the data
    2. NewYork_notes.rtf
      1. Metadata description of the columns and information
    3. Open Carto.com
      1. Connect you Dataset
    4. Create Map
    5. Edit Layer
      1. Style
        1. Change the visualisation colour
          1. By Value
          2. Choose a colour set or custom create one
        2. Pop-Up
          1. Added values accessed by users by clicking on a section
        3. Legend
      2. Publish

Results and Discussion

I created four visualisations for this lab based on the data downloaded from Harvard Election Data Archive for New York 2010.

New York Election Data 2010 – Population depicts the population distribution of New York based on Census data. Colour is based on value (population) with a gradient colour set from light pink to dark pink (increasing population). A Pop-Up is available to users by clicking on an area to view the exact population. The legend informs the user the data depicted, scale, and the average value.

New York Election Data 2010 – Average Democratic Share depicts average Democratic vote share. The electoral data is showing results greater than 0.5 as primarily Democrat votes. Data containing less than 0.5, indicates mostly Republican votes. Colour is based on value (average Democratic vote share) with a custom divergent colour set (red/blue) to represent the more familiar U.S. electoral party colours. A Pop-Up is available to users by clicking on an area to view the population and average Democratic vote share. The legend informs the user the data depicted, scale, and the average value.

New York Election Data 2010 – Normal Democratic Vote depicts the normal distribution of Democratic vote. Colour is based on value (normal Democratic vote) with a gradient colour set from light blue to dark blue (increasing Democratic vote). A Pop-Up is available to users by clicking on an area to view the population, normal Democratic vote, and normal Republican vote. The legend informs the user the data depicted, scale, and the average value.

New York Election Data 2010 – Normal Republican Vote depicts the normal distribution of Democratic vote. Colour is based on value (normal Republican vote) with a gradient colour set from light red to dark red (increasing Republican vote). A Pop-Up is available to users by clicking on an area to view the population, normal Democratic vote, and normal Republican vote. The legend informs the user the data depicted, scale, and the average value.

The shapefile is creating a couple problems for the visualisation. Unlike my inspiration example discussed earlier, this shapefile is currently including a portion of water off the coastline of New York. This is affecting the colour style by value and the averages in the legend for all visualisations. For example, on New York Election Data 2010 – Average Democratic Share areas of water with a population of zero depict a Democratic Share of zero and are misrepresented as a Republican area. This also lowers the average Democratic Share on the legend and misrepresents the distribution. The visualisation incorrectly depicts parks (e.g. Central Park) and rivers (e.g. East River). Assigning these public areas and bodies of water with a population of zero misrepresents the area as a Republican zone. It is also assigning a population value to these areas and misrepresenting them as a Democratic zone.

Future Directions

There are four areas I would like to continue working on with this data and visualisations. First I would like to edit the shapefile. As stated in the Results and Discussion section, the shapefile is currently including large areas of water and misrepresenting uninhabited areas. Therefore creating a misleading visualisation. Second, I would like to combine the Normal Democratic Vote and the Normal Republican Vote into one visualisation. Third, I would add the remaining 49 states to the visualisations. Last, I would like to do further research and add locations of voting stations to the visualisation. By adding these locations I would like to analyze the data and see if there are any connections between the vote distribution between political parties and access to a voting station.

No related posts.