top of page
This site was designed with the
.com
website builder. Create your website today.Start Now

STATSCAN OPEN DATA ACCESS

Through a learning-infused and iterative process of researching, prototyping and testing, we produced design improvement suggestions for the Statistics Canada database website that reflect user goals and lower the barriers to working with open data on this platform.

My roles

  • User research

  • Research & Survey Design

  • User Persona Development

  • Wireframing & Prototyping

  • Project Documentation

Results

  • Streamlined data search & retrieval

  • Improved exploratory previews of datasets

  • Simple & clean user interface

BACKGROUND 
 

This project was conducted as a term project for UBC’s INFO 300: Information and Data Design 13-week course where I collaborated in a team of 5 students with economics and statistics backgrounds. 

 

As global shifts towards information-centred economies and state structures continue to produce vast amounts of data and new sets of power relations around who owns and controls data are surfaced, the project of “open data” and its corollary, citizen science, have emerged. In the spirit of democratizing information access and analysis, open data projects function as ways to hold governments accountable and expand knowledge production to a “wide range of actors across government, academia, and industry to use information for public good” (1). In the economics domain in particular, applied microeconomist Velichka Dimitrova (2) states that “making  economics  research  data  and  code  available  serves  to enable scholarly enquiry and debate and to ensure that the results of economics research can be reproduced and verified.” Spurred by this call to action as well as our lived (and frustrated) experiences accessing open data, our team began our UX research process.

​

UNDERSTANDING USER GOALS
 

To begin the project, we gathered user feedback from interviews and a survey to explore how 18 academics of varying levels (students, faculty, etc.) experienced data accessibility on the Statistics Canada page. People were eligible to participate if they had prior experience conducting research with open data and were recruited by word of mouth, through whatsapp groups made up of UBC students and a subreddit made specifically for student surveys. From the interview and written feedback data, I led the development of a user persona to capture key elements of our user’s needs and concerns.

Our user research provided rich insights into the sequential steps taken by an open-access data user and allowed us to pinpoint crucial junctures where design failures were holding them back from getting the data they needed. What centrally emerged from our research was a shared frustration among users regarding the lack of visualization capabilities in the online web tools and the confusing filter system which made it difficult to find relevant data on the topic and at the granularity users needed.

CONSULTING DOMAIN EXPERTISE
 

Our next step in the project was guided by consultation with scholarly research from the information retrieval domain. In particular, key insights from a paper by Matheus et al. (3) were especially formative for the design direction we took in our development of dataset dashboard previews to address the central user goal of viewing data attributes before downloading.

IDEATION
 

In order to satisfy our users’ key goals within a limited timeframe, we revisited their needs statements in order to establish our priorities before beginning the ideation process. Our identified goals included: ensuring a clean and simple user interface to enhance the accessibility of relevant information, creating a preview capacity for verifying data before downloading it, and implementing a tailored filter feature to assist users in refining their search results. 

PROTOTYPING
 

Following our new to-be journey, we created a low-cost paper prototype and conducted a lean evaluation in which our users interacted with the prototype and provided feedback for further improvement and direction. 

Based on the user feedback, we updated our design and created a mid-fidelity prototype.

Once again, we found three participants to test the mid-fidelity prototype. Each participant was an upper-year undergraduate student in a Statistics, Computer Science, or Neuroscience major. To conduct the evaluation, we asked the user to find and download a specific dataset on GDP from our redesign of the Statistics Canada website. We first showed the users the mid-fi prototype, walked them through the pages, and briefly explained the functions that the design included. Then we took note of the users’ feedback as they navigated the clickable mid-fi prototype in Figma. We recorded the verbal comments and feedback made while interacting with each of the improved functions during the users’ attempt at the task. Finally, we gathered and categorized their feedback according to four data retrieval steps.

Despite the crunch of the academic timeline, we were able to swiftly translate our mid-fidelity feedback into a high-fidelity design.

REFLECTIONS & POSSIBLE NEXT STEPS
 

The next steps of our design could include:

  • Conducting further user testing on our proposed filter categories on both our Data Search Launch Page and Search Results and Search Summary Page to better understand how current drop down options match users needs, and to see if we are missing any filter category headings

  • Researching how our design might be compliant with triple A accessibility standards, given that Statistics Canada is a government website - and how a pleasing visual interface can still be included in an accessible design

  • Exploring how the added data preview features render on mobile devices for users who may wish to access and preview data from their phones

  • Developing diverse preview layout options for different kinds of datasets (ie, time-series vs. cross-sectional data) that require unique visual summaries depending on the variable attributes contained within 

  • Developing visual markers and layout consistency for data objects within the Statistics Canada database that are not downloadable - such as data dashboards, written reports about data, etc.

 

Conducting a UX project from start to finish yielded a number of learning opportunities. The project taught me the importance of:

  • Clearly defining the design problem at hand & retaining an open posture to redefining it when needed and

  • Constantly returning to the key user goals and priorities when details and feature solutionism get in the way of the design process

​

​

​

​

​

  1. Davies, Tim, Stephen B Walker, Mor Rubinstein, and Fernando Perini, eds. 2019. The State of Open Data: Histories and Horizons. Cape Town and Ottawa: African Minds and International Development Research Centre. https://muse.jhu.edu/book/67112 

  2. Dimitrova, V. 2014. Open Research Data in Economics. In: Moore, S. A. (ed.) Issues in Open Research Data. Pp. 141–150. London: Ubiquity Press. DOI:http://dx.doi.org/10.5334/ban.i

  3. Matheus, R., Janssen, M., & Maheshwari, D. (2020). Data science empowering the public: Data-driven dashboards for transparent and accountable decision-making in smart cities. Government Information Quarterly, 37(3), 101284. https://doi.org/10.1016/j.giq.2018.01.006

bottom of page