Women in Data is an international non-profit organization started in 2015 whose mission is to bring women together for career advancement and an opportunity to uplift one another. They have chapters throughout the world and each hold quarterly symposiums that include enlightening talks, expert panels and networking opportunities. 

Co-presenter Rebecca Pop is the founder of Vizlogue, a data visualization and storytelling lab that offers training and consulting services. 

Watch the Recorded Panel 

What’s Inside 

Here are some of the topics addressed during the panel. 

• Can you share your personal journey and how you got started with data visualization? 
• How do you approach data visualization problems? When you are working on a dataset, do you have standard steps/best practices that you follow every time? Are there any key focus areas one should be mindful of? Ann said, “Something so important to know in advance, is whether your audience is technical or non-technical. Technical audiences are people who like data, who love opening a spreadsheet, and are in a data career on purpose. Non-technical audiences are the opposite. They’d rather hire a consultant or let another staff member handle it. It’s probably the last thing on their to do that they want to tackle (and they probably procrastinate!)” 
• Important aspects to keep in mind while working with data are data integrity and data ethics. What is your take on data integrity and data ethics?  
• For someone just getting started in data visualization, it can be overwhelming with the number of tools and courses available these days, what is your advice for beginners? Can you also share some resources? Ann said, “At first, learn the one-hour version of about 10 different tools, but then take a 10-hour training on just one tool and go deeper and specialize. There’s a lot of great courses out there.”  
• What is the future of data visualization? How do you anticipate data visualization to differ in the coming years? 
• Data visualization is a very competitive field, how can one stand out from the crowd and make an impression? Ann said, “Don’t worry too much about having to be the best at everything, I don’t think it’s even possible. Just pick one and play on the strength that you already have and make that public in some way… For example, if you like Tableau post a lot of visualizations on your Tableau public profile. If you like R, post to your code on Github and connect with other people.” 
• What are the key skills required to be successful in data viz? How important is the tool? Ann said, “Chart choosing [is so important]. Are you going to use a pie chart, bar chart or something else altogether? It’s very difficult to take a table, rows and columns of summary statistics and figure out what chart that is going to be. I think a lot of people go to the standards like pie charts or bar charts.” She added, “One activity that you can try for yourself is find a table of data, set a timer for 10-15 minutes and see how many ideas you can come up with in that time period. When I started doing this, I could only come up with a couple of ideas in a 15-minute brainstorming session. Now I come up with 15 ideas in that same time period.” 

Learn More 

Here are some of the resources we mentioned during the panel: 

• The 3-step process for sharing data with users through data placemats: https://onlinelibrary.wiley.com/doi/full/10.1002/ev.20181  
• Ann’s favorite dataviz podcasts, which were mentioned in the Q&A after the recording ended: Explore Explain with Andy Kirk, Data Stories with Enrico Bertini and Moritz Stefaner, Data + Love with Zach Bowders, Data Viz Today with Alli Torban and Storytelling with Data with Cole Knaflic.

The post Women in Data appeared first on Depict Data Studio.

1. Commit to Easier-to-Read Layouts

The end of clustered bar charts is near. I’m on a personal mission to banish these cluttered, impossible-to-grasp-at-a-glance graphs from your deliverables.

This year, resolve to swap your clustered bar chart for a small multiples bar chart, slope graph, or dot plot.

2. Try a New Graph Type

Try a new format, just for fun, then hide it away for a few weeks and forget about its shiny new novelty before sharing it with viewers. Tree maps, social network maps, Sankey diagrams, and more! Shiny! New! Fun amazing toys! How pretty! Give each of these chart types an honest try during your lunch break. But please don’t use them just because they’re cool. Use them when they’re the clearest chart for your viewers.

3. Share Disaggregated Findings in Addition to Aggregated Findings

Small multiples layouts involve multiple small charts, like one chart per company, school, city, or department that you’re reporting on.

Repeat with me: “It doesn’t matter if it takes a little longer to produce a small multiples layout. This is the information my viewers want and need–the details about their specific program, not the means and medians from all the groups lumped together–so my extra effort will pay off tenfold.”

4. Stop Looking for the Perfect Data Visualization Book

Stop looking for the perfect visualization book that will answer all of my questions and transform me into a visualization guru overnight. The field is young. A good number of visualization professionals have written books, and dozens of partnerships for new books are in the works. In the meantime, turn your attention to blogs and tweets.

Here are my highly-recommended blogs for non-programmers who want to get more from common software tools like Excel:

• My blog
• Stephanie Evergreen’s blog
• Cole Nussbaumer’s blog
• Jon Schwabish’s blog

And here are my highly-recommended Twitter accounts:

• @NYTgraphics
• @PostGraphics
• @WSJgraphics

5. Stop Looking for the Perfect Data Visualization Software Program

One day soon, we’ll stop arguing about which software program is best, and we’ll agree that our brains are the best visualization tools of all time.

One day, in a galaxy far, far away, the software salespeople will also stop pretending that their tool is best. Until then, resolve to speak up when you hear their crazy claims.

6. Don’t Trust Your Software Program’s Default Settings

On a related note… As novices, the first few times we use a software program, we’re tempted to trust the program. We might think, “I don’t really understand how this 3D/bevel/shaded chart is effectively communicating my message, but the software company must be full of experts, so I’ll just go with it.” It’s no secret that you’ll need to adjust almost every one of Excel’s default settings to produce something clear and comprehensible. Keep questioning Tableau, R, SPSS, and others, too.

7. Take Your Data Analysis and Data Management Skills to the Next Level

First comes planning… then comes identifying, modifying, or creating data collection tools and instruments… then data collection… then data cleaning…. then analysis… and, finally, data visualization–all while weighing stakeholder information needs, and building a data culture, and anticipating the best communications mode(s) for sharing the completed visualizations.

Do yourself a favor by building your foundational data management and analysis skills this year. Check out my pivot table webinars and spreadsheet tips, or take a course from one of my trusted fellow Excel gurus.

8. Embrace Ranges, Guesses, and Uncertainty in Your Dataset

Only have access to the minimum and maximum values in a dataset? Try a span chart. Want to graph the standard deviation? Add light shading above and below your values. Are you predicting what might happen down the road? Display the estimated range of values with light shading.

9. Don’t Forget to Visualize Your Qualitative Data

Create a chart, diagram, illustration, photograph, or cartoon to display patterns in qualitative data.

Experiment with strategies for displaying data from documents, interview transcripts, and focus group conversations, and you’ll be the next visualization rockstar. This is one of the most underdeveloped areas in the entire field of visualization.

I recently shared ideas for displaying qualitative data.

What do you think of the Washington Post’s simple color-coding for eyewitness statements? I’m a fan.

Visualizing the eyewitness statements: Did Michael Brown charge? http://t.co/dXrtGOcdUr via @postgraphics pic.twitter.com/45R2etOH1L
— Post Graphics (@PostGraphics) November 30, 2014

10. Give Virtual High-Fives to Data Visualization Designers You Admire

Like what you see? Tell the designer. Write them an email, comment on their blog post, send them a tweet, borrow Hedwig the owl. Your kind messages inspire new blog posts, conference presentations, and more, which keeps the field healthy and thriving.

Don’t like what you see? Shhh. Best to keep your thoughts to yourself. It’s bad karma and nobody wants to be friends with Debbie Downer.

11. Use Something Besides the Computer to Display Data

Some of my most valuable breakthroughs with organizations have come from drawing graphs on whiteboards during meetings.

Perhaps you’ve seen Hans Rosling’s 52-second explanation of population trends using stones.

You’re probably familiar with my affinity for sketching graphs on paper.

12. Create an Animated GIF or Video

Of course you’ve watched Hans Rosling’s 200 Countries, 200 Years, 4 Minutes video.

But have you seen the remake GIFs from Dark Horse Analytics? My personal favorite is their Clear Off The Table remake:

How about these changes-over-time GIFs from the L.A. Times Graphics Team?!? Swoon. 

I’m also a big fan of this semi-animated, semi-interactive model of breast cancer causation. Well done, University of California researchers and designers. Here’s to hoping that we’ll see many more GIFs and videos in data visualization’s future.

Join the Conversation

Now it’s time to share your feedback: What are some other ways that you can improve your data visualization skills?

The post A Dozen Ways to Improve Your Data Visualization Skills appeared first on Depict Data Studio.