I’m Susan Kistler, the American Evaluation Association’s Executive Director, and I contribute each Saturday’s aea365 post. What a great week of content on Data Visualization and Reporting (DVR)! Thanks to everyone who contributed, commented, and collaborated to make it happen. I’m excited to explore the breadth and depth of DVR resources and believe that this is a fantastic addition to AEA’s cadre of 40+ Topical Interest Groups.
Cool Trick: Sponsored by our colleagues in the DVR TIG, we’re hosting aea365’s first drawing! To enter, all you have to do, by midnight on Thursday, February 17, in any time zone, is to add a note to the comments section of this post that includes:
- A link to a data visualization that you believe is either particularly strong or weak, and
- A sentence or two explaining what you believe makes it so good or so bad.
Your example doesn’t have to be from your own work (but it can) and can come from any source as long as the visualization is publicly accessible on the web. You don’t have to be a member of AEA (but if you aren’t, wouldn’t you like to join?) to enter or win. If you are receiving this via email, just click back to http://aea365.org/blog/?p=2733 and scroll down for comments.
One entry per person please (although you are welcome to comment on each other’s suggestions all you wish)! We will randomly select two winners from all those who add a comment with a visualization link. Each will receive a copy of Rosalie Torres, Hallie Preskill, and Mary Piontek’s Evaluation Strategies for Communicating and Reporting thanks to Torres, Preskill, and Piontek and our colleagues at SAGE publications (note, if you don’t have a copy, this is definitely worth a read – even if you don’t win…).
The American Evaluation Association is celebrating Data Visualization and Reporting Week with our colleagues in the new DVR AEA Topical Interest Group. The contributions all this week to aea365 come from our DVR members and you may wish to consider subscribing to our weekly headlines and resources list where we’ll be highlighting DVR resources. Do you have questions, concerns, kudos, or content to extend this aea365 contribution? Please add them in the comments section for this post on the aea365 webpage so that we may enrich our community of practice.
Pingback: Anonymous
Congratulations to our winners, Sarah Katherine Collins and Dominica McBride! We’ll get your books out to you this week.
Thanks to everyone who contributed. I enjoyed looking at the lot!
Pingback: Susan Kistler on Sharing Your Ideas on Including Gender and Identity to Win a Copy of Finding Out | AEA365
Here’s a nice visualization of population changes in the United States over the past 100 years.
http://www.statesman.com/news/nation/2010-census-apportionment-data-1252736.html
http://www.biostat.wisc.edu/~kbroman/topten_worstgraphs/bell_fig3.jpg
the figure is found on the top ten worst graphs webpage, which I love. Pie charts are bad enough to begin with, but this figure has not one, not two, but NINE (yes, 9) pie charts. All labeled with molecule names (N03 for example) and percentages. It’s a train wreck of a graphic – far too confusing and it is nearly impossible to compare values. this information would have been better presented in a table. See Few’s “Saving the Pies for Dessert” at http://www.perceptualedge.com/articles/visual_business_intelligence/save_the_pies_for_dessert.pdf.
Donut charts are notoriously difficult to interpret. Here’s a set of charts published a few years ago in New York Times Magazine. The use of concentric circles and the absence of numbers makes these charts particularly difficult to comprehend.
http://www.nytimes.com/imagepages/2008/03/16/magazine/16wwln-lede.graphic.ready.html
http://www.improving-visualisation.org/visuals/tag=qualitative [note: page may be slow to load]
It can be difficult to find good or at least interesting examples of visualizations of qualitative data. The site above has 16 examples of varying quality–some could be more self-explanatory and less crowded visually. I like the one on visualizing surveys and think a similar chart could be made to highlight different themes between groups.
http://www2.ed.gov/offices/OUS/PES/higher/upward.pdf
I’m actually going to take the reverse course from many commenters and post an example where data visualization is poorly executed (read: non-existent!). The report is nothing but a white document with no color, little graphs, and a bunch of text. I would be hard pressed to read this.
http://www.ssec.wisc.edu/data/
While the visualizations on this site focus on maps and geography, I believe it is inspiring and can generate visualization ideas and our evlauative imaginations around how to apply this beauty and mapping to other areas.
In my demonstration at last year’s annual conference, I mentioned this report: http://www.greenpeace.org/international/Global/international/publications/climate/2010/fullreport.pdf
What I love about this work is that it is chock full of evaluative information and it is displayed so well. See page 27 for a great example of “small multiples”
Here’s the one I nominate for worst pie chart ever! From NY TIMES: http://economix.blogs.nytimes.com/2009/01/15/stimulus-pie-chart/
Hard to read, hard to compare, and colors are overwhelming. There is also too much shown here. Imagine this information in a bar graph in descending (or ascending) order of amount spent. SO much easy to read and compare! General rule of thumb: Keep away from pie charts!
Here’s an example of data visualization that I think work really well.
http://www.webdesignerdepot.com/2009/06/50-great-examples-of-data-visualization/
This is my current favorite data mapping source: http://www.patchworknation.org
I love this one because of the wealth of information it has, from simple population counts all the way to literacy rates and election results. I also like the comparisons over many years and the overlap feature: it provides an easy way to visualize how social indicators have changed over time and interact with one another. I think the user-friendly and visually appealing interface makes the tool more practical for everyday use than many government primary data sources.
On a more personal level, my organization puts together an annual health report card, aggregating a variety of health indicators for the state of Colorado, and I think the visual presentation of that data is quite nice: http://www.coloradohealthreportcard.org/ReportCard/2009/subdefault.aspx?id=4133
I like that it presents some key indicators to aggregate into a “grade” but then if you want more information, you can use the navigation on the left to dig down into the indicators and see national and longitudinal comparisons as well as a great summary of why that indicator is important.
This one is a classic:
http://www.edwardtufte.com/tufte/posters
Charles Joseph Minard’s 1869 map of Napoleon’s 1812-1813 march on and retreat from Moscow.
I always find it particularly strong in the way it depicts proportion in motion over a defined area. It also at a glance corrects a mis-assumption; it is clear from Minard’s work that, as brutal as the retreat from Moscow was, most of the Grande Armée never made it to Moscow in the first place. I always think of depicting inputs such as program dollars in the same way–can you imagine being able to depict how most of a program budget never made it to the program in the first place?! Cost-effectiveness redefined!
I love any visualization that accounts for geography. This chart shows the amount of oil reserves by country. The sizes of the countries may look abnormal, and that is because the larger the country is portrayed on the map, the greater the oil reserves that country has. In my opinion geographically based data drives home the point far better than any table could do!
http://www.dailygalaxy.com/photos/uncategorized/2008/01/20/map01_1024_2.jpg
While not a visualization technique, I hope this qualifies.
Stephen Few is well now for his work in visualization techniques. He has a short (19 pages) document entitled Visual Communication: Core Design Principles for Displaying Quantitative Information (you can find this here: http://www.perceptualedge.com/articles/Whitepapers/Visual_Communication.pdf). Very good information. This has greatly enhanced all of my visual presentations, not just charts and graphs.
David
A website that I have found particularly interesting displays their top 7 modern approaches to data visualization.
http://www.smashingmagazine.com/2007/08/02/data-visualization-modern-approaches/
While each approach is innovative, many are cumbersome and cluttered; they detract from the essence of information they are trying to portray vs. supplying clear data.
Science Magazine and the National Science Foundation has been organizing the International Science & Engineering Visualization Challenge since 2003. Their Web site offers a range of creative and inspirational visualizations in different categories including illustration, photography, informational graphics, and multimedia.
http://www.sciencemag.org/site/special/vis2009/show/
If you scroll down at
http://healthit.ahrq.gov/portal/server.pt/community/ahrq_national_resource_center_for_health_it/650
and click to open the PDF of “Using Health IT: Eight Quality Improvement Stories,” then go to page 37, you will see a simple visualization that captures results of an evaluation of an intervention where physicians switched from paper prescribing to e-prescribing. With a visual equation leading to $ going into a piggy bank, it is so intuitive, and brings a smile.
Here is a visualization about Dr Who Villains.
https://newsletter.bimeapp.com/players/dashboard/D98E2B6E62A0ED53E5A5E805A48E3092
It’s cool because it offers different ways of displaying different information – bar charts and a treemap. I like the interactivity – you can click on the individual data points and bring up information about each of them. Lastly, it has a fun but informative aspect, it’s a serious data visualization tool but can equally be used to display fun data sets – the best of both worlds.
With a background in geography and a lifelong passion for maps, I am always drawn to cartographic visualizations. And National Geographic’s representations of vast quantities of data are some of the best.
http://earthpulse.nationalgeographic.com/earthpulse/earthpulse-map
If you find yourself in need of color advice, for maps or other graphics, check out http://colorbrewer2.org/
Several years ago I signed up for the Free-Range Thinking newsletter by Andy Goodman where he gives interesting presentation tips for non-profit organizations. I’ve picked up a lot of good ideas about data visualization and evaluation reporting from the newsletter and the website:
http://www.agoodmanonline.com/newsletter/index.html
You can download his publication “Why bad presentations happen to good causes” from this website, too, although tips and advice in it are similar to some other good books on creating Power Point presentations. My advice to evaluation students is to put a good presentation guide book on your resrouce shelf and refer to it often to stimulate creative and critical thinking.
Here’s an example of data visualization that I think work really well. It’s from David McCandless’s blog/book (informationisbeautiful.net) on the intricacy of intimate partnerships: http://www.informationisbeautiful.net/2010/the-varieties-of-intimate-relationship/. It’s gorgeous! *But* be sure to check the data links — he’s been criticized a bit for using sources that are less than perfect.
I’ll offer three examples of strong data visualization.
1) “A gadget’s life: From gee-whiz to junk” from the Washington Post. http://www.washingtonpost.com/wp-srv/special/business/a-gadgets-life/
This visualization is strong because it simultaneously (and effectively) tracks two variables over time – average cost and number of purchases. Together, the visual impact is not an exact calculation but a thumbprint impression of the graphs’ larger purpose, to demonstrate technological life cycles from obscurity to ubiquity and back again.
2) “Adults With College Degrees in the United States, by County” from the Chronicle of Higher Education. http://chronicle.com/article/Adults-With-College-Degrees-in/125995/
The census data here is about as basic as it gets. What’s compelling is the level of detail provided and the interactivity of the visualization. Viewers manipulate the map to see change over time and variation between subgroups, deepening (and perhaps renewing) our understanding of something commonplace to demography.
3) “Budget Puzzle: You Fix the Budget” from the New York Times. http://www.nytimes.com/interactive/2010/11/13/weekinreview/deficits-graphic.html
Again, the strength of this visualization is its interactivity. To simply share the data, federal spending categories could be displayed in a traditional bar graph or pie chart. But the visualization’s higher purpose is to communicate what a huge “puzzle” balancing the budget is and to challenge viewers to explore different ways to solve it.
Pingback: Tweets that mention Susan Kistler on Suggesting Data Visualizations to Win a Copy of Evaluation Strategies for Communicating and Reporting | AEA365 -- Topsy.com
Pingback: Tweets that mention Susan Kistler on Suggesting Data Visualizations to Win a Copy of Evaluation Strategies for Communicating and Reporting | AEA365 -- Topsy.com
I’m not sure if this is what you are looking for, because it is more a discussion and comparison of two data viz.
http://www.stat.columbia.edu/~cook/movabletype/archives/2009/12/healthcare_spen.html
I find the author’s commentary comparing the two ways to portray healthcare spending per country to be insightful. I also found the extensive discussion in the comments to help me to understand the limitations, and strengths, of both the bubble and parallel-coordinates graphs. Both could be fancied up a bit for public consumption, but through the article I gained a better understanding of each option.
I will offer this one, from Power of Data Visualization: An Infographic Inspiration Site. Here’s the link to a timely data visualization on Valentine’s Day:
http://www.pdviz.com/valentines-day-the-price-of-love
It shows several graphics that are easy to read and interpret, and use color and shape well to enhance the visual appeal.
I also like this one they offer on cats:
http://www.pdviz.com/strange-facts-about-cats
I like the way they use words in different sizes and colors to draw the eye to each section. You don’t need to read the graph in order. You just naturally go to the places of interest.
While neither are “perfect” examples of Data Visualization and I’m sure the experts can suggest improvements, I do like this site for the inspiration it offers.