First, some ground rules for these discussions:
- Don’t blame women and POC for their lack of representation. If there aren’t enough submissions from women and POC, assume the problem is with the competition design and in aspects of the community, not their rate of submission. Look at the competition and community pipeline to see how it’s exclusionary, rather than (or in tandem with) work on submission rates
- Don’t assume data is neutral. Information is socially-constructed, subjective, and fallible – and more information does not necessarily lead to more knowledge.
- Don’t assume the judges/rules/best practices are infallible/objective/correct/neutral. Tufte did his research on mostly male, white U.S. college students; the field of visualization and its best practices grew out of that research. Many people are problematizing conventional wisdom (about pie charts, bar charts, and data cleaning, among others). Not that we shouldn’t value clarity, precision, and standardization, but other values can be just as important, even if they are not explicitly encoded into Tableau.
In a recent Twitter thread, someone shared a link to 10 Ways to Increase Girls’ Participation in Computing Competitions. This was largely drafted before reading that list, but I think many of these ideas are in line with that advice. Also, this is not intended to be comprehensive – nor are all of these necessarily good ideas – but I think it is helpful to have something concrete to react to and adapt.
Small scale (working within the system)
- Provide a single, clear place for people to find out about announcements – done!
- Announce timing of contests in advance so that people with less flexible jobs or family/personal commitments are able to make arrangements in advance if they like
- Consider expanding the Iron Viz submission window for similar reasons
- Provide a description/template for optional blog posts about the contest (or even a round-up of exemplars). Also make it clearer how or if these inform the judging.
- Provide a longer explication of judging criteria. This could be done as a longer write up of the winning feeder entries and finalists (why and how the winning submissions excelled in each area; what about the viz was exciting, innovative, and novel. Don’t let the viz speak for itself!) or a more robust description of each category and how the judges evaluate the entries.
- Announce runner ups!
Medium scale (questioning the system)
- Provide prizes for runner ups and ways to win beyond achieving the #1 spot. This could be a “best in class” along the various criteria (top scorers in “design,” “storytelling,” and “analysis”) or ad hoc awards, like “most technically impressive,” “most creative,” or “something we’ve never seen before.” With “best newbie” replacing community votes, people who have entered before are in an all-or-nothing competition that takes a lot of time and effort. Additional recognition will also have the benefit of improving community understanding of what that category means and what exemplary work looks like.
- Host smaller-stakes competitions throughout the year. Makeover Monday gives opportunities for public review and improvement every week, but there is no “winner” (…aside from perfect attendance, maybe!)
- Offer additional opportunities for community members to reach the Tableau conference without winning the highest-stakes competition or investing over $1,000 in registration fees. This was done in 2016 with the Tableau 10 Olympic Torch competition but I don’t believe anything similar happened in 2017.
- Provide feedback to participants on their IronViz entries. What were some of the judges comments? What were some areas of improvement? This could help strengthen the talent pipeline and ensure people feel seen, even if their work doesn’t make a big splash publicly.
- Have a community-based discussion of judging criteria. Reach out particularly to participants in underrepresented groups. Would they change anything? Would the community want to add additional criteria, like promoting social equity?
- Maybe try blind entries for public comment until the results are announced. This might backfire spectacularly, as I think a lot of people do enter for public recognition, but do any entries get more public attention without recognizable names and logos?
- Build in other best practices (open access, copyright, reproducibility) to the contest rules:
- No graphics, icons, or fonts for which you don’t have permission
- Use publicly-accessible datasets or make the full dataset available for download
- Make workbook available to download
- Make people who are already great stretch themselves and tap into knowledge and forms of creativity other community members may already have but don’t get to show off. Have a way for community members suggest a theme (email address, box submission, etc.)
- Improvement/revision based contest (e.g., re-do an old viz)
- Partnership-based contest (sense of community) not one person:
- Build a viz in collaboration with another designer
- Build a viz in collaboration with the person doing the data collection / someone from the community being studied / an expert in that field
- Build a viz in collaboration with a partner organization (e.g., Data Science for Social Good, Viz for Social Good, volunteer organizations, Tableau Foundation partners) Give those partners a say in the judging/evaluation criteria!
- Rather than a specific topic as the theme, consider offering challenges (similar to the Mobile contest):
- Build a viz using only text data
- Viz about a single person – not yourself!
- Use a dataset that challenges you
- Build a viz with no bar graphs
- Build a viz only with bar graphs
- Build a viz with no graphics or images
- Make a viz that is accessible to someone who is visually impaired. Go beyond a color blind palette! Make it comprehensible to someone using keyboard navigation or someone totally blind. Invite someone with colorblindness or vision impairment into the judging process
Large scale (changing the system)
- Invite a member of the community to the judging panels. Better still, invite community members of underrepresented groups to all judging panels and pay them for their time 🙂
- Have a feeder competition helmed and judged by an all-community member panel
- Consider an IronViz Invitational, or a preliminary/qualifying competition so people know there has been some amount of selection before they are asked to invest a substantial amount of time
- After the deadline, maybe create a mentorship window where participants are given a week to make changes based on feedback from a matched Tableau employee/Iron Viz champion/Iron Viz contestant/Zen master, etc. Or maybe create a different mentorship pipeline people can opt in or out of
- Consider funding a position for a diversity fellow: give them money to mentor underrepresented designers, listen to and learn about their challenges and contexts, make recommendations for changes based on those experiences
- Allow participants to provide their own criteria for evaluation. What do they want to be judged on? Aesthetics? Creativity? Emotional response? This might allow people to set and point out personal goals in line with the contest
Would love to see other ideas – or to hear any thoughts on the ones above!