Looking back, 2014 was a great year for data visualization in Chicago and our group. The group met 10 times this past year in a mix of workshop and seminars. Group membership grew from 700 to 1,400 members over the year. I wanted to review what we did last year, where we excelled and where we fell short. Even though it was a wonderful year for the group, 2015 is an opportunity to grow and offer even more to those interested in data visualization in the Chicagoland area.
City of Big Data
Given my role in Chicago’s government, the city (the entity of residents, companies, and culture; not just the government) tended to be showcased. The City of Big Data exhibit at the Chicago Architecture Foundation was a grand exhibit of the intersection of urban spaces, data visualization, and data. Computer-based 2-D and 3-D models were featured, and a real-life 3-D projection of data onto a scale replica of downtown Chicago.
The organizers of the exhibit were exploring how data is not a spreadsheet or abstract notion, instead, it surrounds us. Some of this data is public, highlighted by its display on the 3D model. But it’s also private, captured by Facebook activity and streaming music. In all, it surrounds us in the same way it does in the exhibit.
Data Science for Social Good is a relatively new program that puts highly-qualified and trained undergraduate, graduate, and post-graduate students in Chicago working on research problems to improve society. Many of these fellows were part of the group over the year. This fall, we were able to see some of those projects which are mostly available as open source projects.
It was obvious that Shiny was a popular tool to quickly prototype and build interactive programs and maps for end users. Grand visualizations were not a primary outcome of those projects, but because of the operational nature of these systems, clear, concise visualizations and maps were rolled out. While visualization is often associated with impressive, large displays, some of the most crucial visualizations follow good practices on charts and maps, enabling exploration through advance filtering. While grand visualizations are the most impressive, the most important visualizations are well-executed visualization of maps and simple graphs.
Importance in design
One of my favorite meetup concepts in 2014 was exploring how design choices leads to a different user experience. This past April, Heather Billings of Chicago Tribune and Derek Eder and Erik Van Zanten of DataMade shared how they visualized the crime data located on the City’s open data portal. It’s a popular dataset from the number of rows (over 5 million) and the pertinent nature of the topic. There seems to be hundreds of analysis and graphs of the data, but the dataset is only 22 variables (columns)–how can the same data look so different?
Each one of the designers took a different approach to visualizing crime data. Chicago Tribune focused on crime in the neighborhood. Derek was candide in admitting that the project was an exploration and learning project as an intentional visualization. Erik demonstrated a tool that allowed people to define their own area of interest. Each project used the exact same data source, but the design and visualization choice led to different user experiences and outcomes.
The city is often the subject of research, an easy topic since it’s a topic that literally surrounds us every day. The newly launched bike-share program led to the DIVVY Data Challenge was quite successful. Similar to crime data, there were a number of drastically different designs based on the same fundamental data, so how different of a design and experience yields from the same data? In this case, there were 99 submissions, all with their own design.
We looked at Divvy Spokes, a chord diagram showing neighborhood-to-neighborhood connections by Shaun Jacobsen; thorough statistical analysis by Drew DePriest on ride patterns and his struggle with seemingly messy data; custom dashboard (with crossfiltering!) by Michael Freeman; and learned Voronoi tiles from Gabe Gaster on his map. Additional “show and tell” presentations by KK shows the usefulness and ease of Tableau and–in one of the most amusing apps–find your match on DIVVY with Serendivvity. The same dataset was literally ranging from a dashboard to a dating site. Ultimately, the design and visualization choices leads to an entirely different experience.
We were treated to nationally renowned data visualization experts. Alberto Cairo previewed his upcoming book and his perspective on the role and epistemology of data visualization. Kaiser Fung, the embodiment of an art critic turned data visualization aficionado, reviewed the best and worst charts–and what makes a good chart, good. Both Alberto and Kaiser, leading speakers in this area, were able to cap off a productive year with outstanding talks.
Review of 2014 goals
Early in 2014, I noted three ways to facilitate a robust meetup group: (1) seminars, (2) teach & learn (workshops), and (3) show & tell. Seminars were a strong part of the year. Visits by Alberto Cairo and Kaiser Fung were enlightening. Generally, we were able to hear from a speaker almost every month. Seminars are a clear staple of this group and will continue to be in the future.
The show & tell sessions were also repeated throughout the year. The DIVVY meetup was particularly wonderful as some great data visualizations were presented during very short presentations. It was successful enough this year where I aim to deliver more of these opportunities in the future.
Sadly, this puttered out throughout the year and I do not think we achieved our mission of providing workshops and conducting adequate hands-on training. We started strong with workshops on D3 (many thanks to Paul Katsen for a wonderful job!), but scheduling started to interfere. We are going to renew our efforts to offer workshops in 2015.
Looking forward to 2015
We will continue to offer the three areas of focus–seminars, workshops, and show & tell–from 2014. I hope to improve the number of workshops that can be provided to the Chicago community. But, in addition, we will take the opportunity to improve upon the past year. Thus, the next year will include new focuses that weren’t covered in 2014.
Data visualization is frequently embodied by business intelligence tools, used by large and small organizations. We will be spending some time this year focusing on business intelligence platforms. There is an unfortunate vacuum in the Chicago community around BI platform awareness.
Our first meetup (TBA) will discuss the Tableau platform. As we learned from the Data Science for Social Good group, some of the best data visualization is good execution of basic graphs. The rise of “data exploration” platforms have focused more on visualization and discovery than automating or “pixel-perfect” reports. We will introduce these platforms and discuss the pros and cons of these tools.
Data visualization is also establishing itself as a profession. Dedicated data visualization-ists(?) are being sought on job boards. Thus, we will be more considerate on making a connection between employers and potential employees. Late in the year, we introduced an opportunity for employers to note they were hiring. We will continue that practice. Likewise, we will increase the offerings for serious professional development in the group.
These lists are never exhaustive, but is a start of a plan for 2015. I am interested to hear your feedback through this blog, @ChicagoDataVis, chicagodataviz@gmail,com, and at the events.