Know Your Neighbourhood before you commit


Final Projects, Visualization

Introduction Story:

This project was done with the motivation of analysing data of 311 complaints in New York. It is meant to be an exploration of what do New Yorkers complaint the most about when it comes to being residents. It additionally analyses key-factors that can influence renting/buying in a particular area. The genesis of this visualisation came from personal experience of being a New NYC resident. Facing a heating issue exposed me to the aspect of 311 complaints and while I explored this data I realised the gamut of complaints are beyond just heating. The data made me think that this could be a useful tool for a new NYC resident to ‘know a neighbourhood’, and examine and be prepared for recurring problems that may occur in a place that they plan to move to.This visualisation explores

1. Which department or agency gets the most 311 requests

2. The most common complaints amongst NYC residents

3. Analysing and finding the Loudest Neighbourhoods of NYC

4. Analysing and Finding the most unclean neighbourhoods of NYC

Before you explore this data here are a few terms to familiarise yourself with

311 complaints

311 is a helpline service available to citizens that allows them to report community issues or make inquiries without overwhelming emergency lines or facing bureaucratic obstacles. By contacting 311 or the corresponding local number, individuals can speak with a qualified representative who can escalate complaints to the appropriate authorities, such as local government and public utilities, to facilitate solutions to problems or provide answers to queries.

Borough

A borough in NYC is a geographical and administrative division of the city, each with its own distinct character, history, and local government. There are five boroughs in total: Brooklyn, Queens, Manhattan, the Bronx, and Staten Island. Each borough is further divided into neighbourhoods, and they all share a common infrastructure and services provided by the city government, such as public transportation, sanitation, and emergency services.

Zipcode

It is a 5 digit number used in the US to identify the geographic location of an address and assist in the sorting and delivery of mail.

Process:

The process of building the visualisation involved the following steps.

Gathering Datasets

Making of this visualisation required looking for reliable data to base the visualisation on and further draw inferences from. It also involved looking for multiple sets of datasets whose datapoints could be combined to perform a detailed analysis. I explored the NYC Open data website for the 311 data. The original data set was a 17.6 GB file since the database had data from 2010 – till date. I decided to use the filter on the NYC open data website to download the latest 1-month data. This ensured that the file was within the 2GB limit of Open Refine. I initially tried using the api calls to render data in Open Refine but it was trimming the data to 1000 rows only. I also looked for datasets with age demographic data per Zipcode. I used this demographic data to analyse the relation between noise complaints and age of residents of the area. I found this data on the nyc.gov website

Cleaning Data

The 311 data had quite a few columns with a lot of data points. I decided to use OpenRefine to clean and make my data Tidy. The dataset recorded 41 fields like Agency name, Incident Zip, Geographical co-ordinates, Complaint type, Address etc. I deleted columns that wouldn’t help my visualisation like Road Ramp, Bridge highway etc. I additionally only retained rows by the top 4 agencies/ departments that received complaints and were also departments with nature of complaints that would affect a resident from a renting/ buying perspective. I also cleaned the data by clustering them by similarity and finally exported the sheet in a csv format. The demographic data didn’t need much cleaning it was ready to use for the visualisation.

Visualising the data

Know your neighbourhood

To visualise the data I decided to use Tableau as it offered the flexibility of geographic and statistical visualisations and also gave me the flexibility to experiment with types of visualisation. I imported both the 311 and the demography data set and created a JOIN in Tableau for me to create combined visualisations. I started out creating a Treemap for the department that receives the most number of complaints. I then funnelled the visualisation with a bubble chart of the most common complaints reported , for this I combined similar type of complaints in a group with a single label, this made the visualisation clearer. I finally focussed my visualisation on Noise and Unsanitary conditions and a combination of both. I also visualised a combination of demographic distribution of young population aged between 20 yrs – 34 yrs and noise complaints from each borough. Finally I created a dashboard and story to represent these visualisations.

Design decisions

To design my visualisations my main goals were.

1. Clear representation of data which is simple to understand and infer from

2. Using a combination of bar/ area charts and geographical visualisations to best represent each data story and analysis type.

Demography and Noise data by Borough

I additionally decided to experiment to create my own visualisation with multiple data points for the noise and demography combination chart. I used height, width and colour of the bar chart to represent different data fields.

Colour Palette

With my colour palette I decided to use the colours of the 311 website and used their font as an inspiration for the fonts that I picked (DIN Alternate and DIN Condensed). I kept the contrast high so that it would be visually easy to read.

UX Research and Testing

With my UX Research my goals were as follows

1. Test for Visual Accessibility

I used a visual accessibility checker to check for contrast ratios and colour choices of my visualisation. This helped me ensure readability of the visualisation.

2. Test for flow of the data story and understandability

I used simple tasks questions like

– What can you infer from the visualisation

– Did you find anything difficult to understand

– Were you able to read through the visualisation clearly (Rate from 1 to 3 , 3 is great visual clarity and 1 is low visual clarity)

I used a Think-aloud walkthrough technique to perform my research and made changes accordingly.

In terms of recruiting participants I recruited 2 residents of Brooklyn. My participant persona was someone who

1. Lives in NYC

2. Has experience with 311 complaints

3. Looking to move to a new place for self or for a friend.

Improvements post UX Research

The improvements I made to my visualisation post testing was.

1. Change colours on the 311 Incident reporting map since the gray of the map and the scale were conflicting

2. Added zip code boundaries to my mapped visualisations.

3. Added a map visualisation for the Unsanitary neighbourhoods of NYC as well

4. Changed the axis label on the demography+noise complaint graph to read as “Value ranges” so that people wouldn’t be confused with the stepped bar to have a meaning.

Findings :

Noisy and Unsanitary neighbourhoods of NYC

The visualisation led me to interesting analysis and findings

1. NYPD has the most 311 complaints associated with it even though these are non-emergency complaints.

2. The biggest complaints amongst residents is of Illegal activities (drugs, fireworks, housing an illegal animal etc), Noise and heat/hot water

3. Brooklyn is the Noisiest Borough of NYC followed by Manhattan

4. The most noise complaints source come from residential noise. Brooklyn also has the most residential noise and the the maximum population of 20 to 34 yr olds. From this we can infer that it is probably the loud music and partying by this age group that lead to most complaints.

5. In terms of street and sidewalk noise Manhattan is the loudest which also ties to the presence of a lot of bars and moving traffic at all hours in Manhattan

6. Brooklyn also tops the list of most unsanitary neighbourhood followed by Bronx. Flatbush and adjoining areas like Brownsville in Brooklyn tops the list for both Noisiness and lack of cleanliness.

7. Midtown to adjoining areas of Upper Manhattan seem to be the loudest in Manhattan which can be owed to the bars and restaurants in the area as well.

8. Upper Manhattan and adjoining area of Bronx seems to be facing a lot of sanitation complaints

9. Rodents and in general unsanitary premises seems to be the biggest sanitation concern for residents.

10. Areas around parks or parks are fairly clean and well maintained.

Reflections and Way Forward :

I started out this project wanting to explore 311 complaints in general and extrapolate data with regard to noise. While I started building out the visualisations the data made me realise that sanitation is another major concern that would influence a resident and that pushed me to account for both and explore both aspects. As a build up on this topic in the future I would like to overlay median income data on the sanitation data and see the correlation between the maintenance and attention to an area in terms of sanitation and median income. I would also like to overlay restaurant and bar data with that of noise and sanitation data and see how much of an overlap there is due to close proximity of these commercial establishments to the residential ones. I also thoroughly enjoyed experimenting with a multi-dimensional visualisation and felt that omitting the axis for that particular visualisation might be worth considering.

Bibliography :

“Planning-Population-American Community Survey-DCP.” Accessed May 1, 2023. https://www.nyc.gov/site/planning/planning-level/nyc-population/american-community-survey.page.

“311 Service Requests from 2010 to Present | NYC Open Data.” Accessed May 2, 2023. https://data.cityofnewyork.us/Social-Services/311-Service-Requests-from-2010-to-Present/erm2-nwe9.

SafeWise. “What Is 311 and When Should I Use It? | SafeWise.Com,” May 23, 2022. https://www.safewise.com/blog/what-is-311/.

Coolors.co. “Coolors – The Super Fast Color Palettes Generator!” Accessed May 2, 2023. https://coolors.co/.

“WAVE Web Accessibility Evaluation Tools.” Accessed May 2, 2023. https://wave.webaim.org/.