winner
Our group won the Data and Digital Storytelling Award at the 2020 Research, Innovation, & Scholarship Expo (RISE).
summary
As researchers partnered with the city of Boston, our group worked to understand how community-based and advocacy organizations in Boston interact with public data. Through interviews with several organizations from a variety of sectors, as well as meetings with the head of the city’s Open Data department, we identified several pain points and inefficiencies in the flow of public data.
Through our research, we found:
Inefficient use of resources in clarifying the data-related inquiries of individual organizations
A lack of knowledge from the city regarding how public data is used by organizations
Little-to-no connection or cooperation among organizations with similar data needs
To aid the first issue, we developed a web interface to streamline the process of data inquiries from organizations to the city. This interface will also aid the second issue by providing a method of gathering and displaying projects from organizations that use public data. Lastly, to improve the third issue, we developed and hosted a series of events designed to connect organizations with similar data needs, as well as provide general data education.
initial research
After first meeting with the head of Boston’s Open Data Department, we learned that she spends a lot of time and resources working with organizations to clarify what data they want from the city and how they could use it. We discussed ideas for designing an system that would help alleviate the work done on her end by allow organizations to form better data-related questions.
Goals:
Input: questions from organizations about data and specific datasets on Boston’s open data portal.
Output: better, well-formed questions.
the first prototype
so what makes a good question?
In order to develop a system that helps users form better questions about data, we first had to understand what makes a good question. In our investigation, we discovered the Right Question Institute, a non-profit, educational organization that promotes a learning strategy based on “asking better questions.” Through their educational materials, as well as other research, we developed five key elements that a good question must have:
A good question, given the scope of our system, should be:
Relevant. Ensure that it relates to the problem being discussed.
Purposeful. Include the reason why the person is asking the question.
Specific. Limit the scope of possible answers.
Productive: Understand how expected or unexpected answers will be useful to the proposed solution.
prototype concepts
After brainstorming a list of possible prototypes to create, we went through all of our concepts and grouped them by idea and goal.
Narrowing down to five ideas through group discussion, we drew up storyboarding sketches of the remaining prototypes.
engaging users
In testing our ideas, we found that the best way to engage users with our interface was to provide a mad-libs style framework, where they could add the information they were seeking. At this stage in our project, we had not yet made connections with actual users, so we performed testing with our colleagues.
testing
We tested our frameworks with a Google Form, where users could fill in answers to each of the prompts. Unfortunately, we did not get enough responses to have significant data from this test, but the results helped inform us about how users approached the wording in each of the prompts.
results
The frameworks that garnered the most productive responses were the following:
My neighborhood needs ___ to help ___
I want to help my neighborhood by ___
user interviews
Up until this point in our project, the ‘good question’ prototype was only being developed with input from the city of Boston, not from actual users. It was important to understand how the city infrastructure could support our proposed solution, but in order to create a successful product, we needed to hear from users directly.
Over the course of a few months, we were able to interview people from the Boston Area Research Initiative, Asian Women for Health, LGBTQ Youth Organization, and the Boston Public Health Commission. Through these interviews we were able to better understand how different types of organizations interact with public data.
We found that, in the flow of public data, from initial collection by city departments, to the end uses by organizations, there were many pain points that organizations faced. Based on our discussions with organizations, we grouped these issues into two, large categories:
Issues with obtaining public data.
Issues with analyzing the data, once it had been obtained.
visualizing the results
We set about creating a gigamap as a sense-making tool for the information we had gathered in our interviews with organizations. Since the organizations we talked to largely focused on the two issues described above (obtaining data, and analyzing it), we decided to map out the flow of public data in Boston, and highlight points at which the flow broke down or failed.
initial sketches
final visualization
the new interface
Having created the sense-making tool above, we then began designing a high fidelity version for the ‘good question’ prototype.
sketches and storyboards
screens
data events
After completing our initial research and developing the high fidelity web prototype, we set about finding a solution for the third issue we found in our research: little-to-no connection or cooperation among organizations with similar data needs.
Having spoken with many organizations, most expressed that they would enjoy face-to-face interactions with those doing similar work. So, we began developing a series of events designed to connect organizations with similar data needs, as well as provide general data education.
Unfortunately, we were only able to run one event, as the rest were cancelled due to COVID-19.
planning
overview
We planned for our first event to be small and intimate and at Northeastern Crossing, a public space that the university uses to interface with the surrounding communities. We posted our event on Eventbrite, and sent invitations to the organizations we had previously interviewed, as well as a handful of new organizations.
agenda
Introductions: Time for everyone to get comfortable and discuss the purpose of the event and the agenda.
Data Training Session: We had a researcher from the Boston Area Research Initiative (BARI) come in to give a brief overview of a data mapping tool that organizations could explore.
Interactive Data Activity: We organized an activity for organizations to learn from other organizations for how they approached challenges with data.
feedback
survey results
After the event, we provided questionnaires to those who attended, so that we could gather feedback to improve the following events. We found that:
Most attendees were comfortable interpreting data visualizations but not analyzing the raw data itself. Many expressed interest in having more training in this area as part of future events.
Most attendees wanted more information on what the city is doing to help them.
Many attendees wanted more time for networking among their peers.
retrospective
For our team retrospective, we found a couple of things to improve upon:
Spend more time explaining our research, and why we are performing certain functions on behalf of the city.
Focus more on networking and person-to-person interaction.
Offer more context for everything we may be presenting, as there were moments where attendees mentioned they were confused about material being presented.
next steps
We are packaging up both the prototype and the event content for the City of Boston to hopefully implement over the coming years. We hope that the integration of our UX research into the City of Boston’s open data portal will help community organizations have better access to the data and resources they might need to reach their goals.