Lab 3: Mayoral Mystery

The goals for this tutorial are:

to experiment working with maps and geospatial data
to build data exploration skillsets with ambiguity
to iterate with visualizations to discover data insights
to hone the data → visualization workflow

The assignment requirements are as follows:

Create a dashboard in the index.md file using the data provided in the /data folder. Please also include written context for the answers that is supported by visualizations.
Create at least two, but no more than four visualizations with Observable Plot (or d3, if you feel ready for that, but not recommended).
One of your visualizations should be a map of the nyc districts with some data visualized. That data can be points on the map, or a choropleth. The index.md file has some initial work for the geoJSON data to get you started.
This lab is different in that you are not obligated to answer any data questions in particular, but some relevant insights do exist. Since this lab has less structure, here are some questions that you may explore to get you started if you feel lost:
- How did the candidate do, overall? How do the results vary by district? By income level?
- How did people feel about the issues? What other information do the survey responses tell us?
- What was the extent of the Get Out The Vote campaign? Was it successful?
- This candidate would like to run again. What would you suggest for a successful next campaign?
Submit your deployed link as a comment on the lab 3 commons post.

Since this lab doesn't have a set slate of research questions, take time to understand the data, consider what you think is important context for the staff, and present it thougtfully and clearly. As mentioned in the syllabus (pdf in the home page of commons), for labs 3 and 4, you will be graded on:

On time submission
Technical requirements met
Thoughtful responses to research questions (if applicable)
Well articulated data presentations (new)
Design considerations (new)

Data Context

A political campaign has collected data from their recent election cycle and needs analysis to understand their performance. The campaign ran from May through November 2024 across New York City's borough-community districts. The candidate had some exiting momentum, and would like to run again. The campaign needs to know areas in which they succeeded, and areas where they could make changes for next time.

Some notes:

Borough-community districts (boro_cd) are coded as 3-digit numbers where the first digit represents the borough (1=Manhattan, 2=Bronx, 3=Brooklyn, 4=Queens, 5=Staten Island)
Income categories are derived from median household income: Low (<$50,000), Middle ($50,000-$100,000), High (>$100,000)
Survey responses include both voters and non-voters, with open-ended feedback on campaign effectiveness

The data has been provided in multiple datasets as CSV files, along with a topojson file of NYC districts.

NYC TopoJSON

There is a topoJSON file called nyc.json in the data folder. This file is a simplified version of the NYC voting districts geoJSON sourced from NYC open data. Debugging and projecting this data can be fairly advanced, so the file is already imported, converted to geoJSON, and rendered in a plot in the index.md file for you. To make a visualization of this map, you can expand on the existing plot by changing the Plot.geo mark or adding more marks in addition.

Campaign Datasets

campaign_events.csv (May - November 2024)

Column Name	Type	Range/Values	Description
event_id	Integer	1-130	Unique identifier for each event
boro_cd	Integer	101-503	Borough-community district code
income_category	Categorical	Low, Middle, High	Income classification of the district
event_type	Categorical	6 types	Community Meeting, Volunteer Training, Rally, Roundtable, Town Hall, Canvassing Kickoff
event_date	Date	2024-05-01 to 2024-11-02	Date the event was held
latitude	Numeric	40.58-40.90	Geographic latitude of event location
longitude	Numeric	-74.15 to -73.74	Geographic longitude of event location
estimated_attendance	Integer	33-296	Number of attendees at the event

survey_responses.csv (Post-Election Survey)

Column Name	Type	Range/Values	Description
respondent_id	Integer	1-100	Unique identifier for each respondent
boro_cd	Integer	101-503	Borough-community district of respondent
age	Integer	25-74	Age of respondent
heard_of_candidate	Categorical	Yes, No	Whether respondent was aware of the candidate
affordable_housing_alignment	Integer	1-5	Agreement with candidate's housing policy (1=strongly disagree, 5=strongly agree)
public_transit_alignment	Integer	1-5	Agreement with candidate's transit policy
childcare_support_alignment	Integer	1-5	Agreement with candidate's childcare policy
small_business_tax_alignment	Integer	1-5	Agreement with candidate's small business policy
police_reform_alignment	Integer	1-5	Agreement with candidate's police reform policy
voted	Categorical	Yes, No	Whether respondent voted in the election
voted_for	Categorical	Candidate, Opponent, blank	Who the respondent voted for (if applicable)
open_response	Text	Free text	Optional feedback about the campaign

election_results.csv (Election Results by District)

Column Name	Type	Range/Values	Description
boro_cd	Integer	101-503	Borough-community district code
median_household_income	Integer	$34,560-$142,890	Median household income in the district
income_category	Categorical	Low, Middle, High	Income classification of the district
total_registered_voters	Integer	30,366-49,906	Number of registered voters in the district
votes_candidate	Integer	7,394-21,328	Votes received by the candidate
votes_opponent	Integer	4,963-28,347	Votes received by the opponent
turnout_rate	Numeric	39.56%-84.92%	Percentage of registered voters who voted
gotv_doors_knocked	Integer	108-7,940	Number of doors knocked for Get Out The Vote efforts
candidate_hours_spent	Integer	1-28	Hours the candidate personally spent in the district