BACK TO TOP

NYC OPENDATA

Dashboard Design for OpenGov. Database - Web AppClient: City of New York (Academic Project)

PROJECT BRIEF

NYC Open Data is a web app for accessing and using data published by New York City agencies and other partners. The Mayor’s Office of Data Analytics (MODA) and the Department of Information Technology and Telecommunications (DoITT) partner to form the Open Data team. As a hub of analytics in the City, MODA advocates for the use of Open Data in citywide data analytics and in the community.
To make the data open and transparent, NYC Open Data also provides online tools for exploring the data sets. For our project, we will be testing, improving and adding analytical features to the existing NYC Open Data website. Users of this app may be not analysts or data scientists. A beginner-friendly data analysis app is important.
We did thorough research of usability tests to create three user journey maps for three profiles of user: Beginner, Policy Maker and Data Veteran. The core of the web app interfaces are the same based on the three views of data, but the tools are provided and organized in diffrent way to better suit different users' needs and ease the learning curve.

USABILITY TESTS

The participants would look at a data set for 311 calls and be given 4 tasks for test in the original website of NYC Open Data. The following five videos of usability tests were conducted by the project instructor(s).

Task 1: It looks like there is data in here from all over the city - that is not what we want. How can you look at *just* the TLC data（Taxi & Limousine Commission）?
Task 2: You should now only be seeing 311 calls regarding the TLC.This dataset goes all the way back to 2010 – can you limit it to only show data created from *last week* (Created Date)?
Task 3: You should now only be seeing 311 calls regarding the TLC, from the *last week*. Now that you have the right data, can you put it on a *map* so we can see where all these calls are coming from?
Task 4: Lastly, every row of data here shows us an *individual* 311 call. How could you look at a *summary* of *total* number of calls shown by zip code in a table?

PAIN POINTS & USER JOURNEY MAPS

A. The user groups are not defined, users without data-analysis background need clearer tutorials and examples, while users with data knowledge need more professional tools.
B. The user flow of analyzing and visualizing data are not fluid. The operation panel needs help users toggle between raw data, filtered data, aggregated data and data visualization.
C. User interface is not friendly with poor layout; it's confusing to click around to find what it needs to finish a task.
D. UX writing and terminology are not consistant and comprehensive, users without data-analysis background could easily get lost by the buzzwords.

BRAINSTORM & SOLUTIONS

A. Introduce three profiles for users to choose: beginner, policy maker and data veteran. Design a wizard to guide new users to ease the learning curve.
B. For beginner, simplify the workflow to "select""filter""visualize"and"share". For policy maker, create pre-set templates, focus on summaries and reports. For data veteran, provide customizable tools to clean raw data, analyze aggregated data and present visualized data.
C. Divide the layout into tab system, operation panel and viewport to clarify the functionality.
D. Create consistant visual symbols to unify the three interfaces. but varify the writing and vocalbulary accordingly for different user groups.

THREE VIEWS OF DATA

In data analysis, a "raw data view" shows the unprocessed, original data points in their entirety, while an "aggregate view" presents a summarized version of that data by grouping and calculating statistics like averages or sums, and a "visualization view" takes the aggregated data and displays it graphically using charts, graphs, or maps to reveal patterns and trends easily.