Category Archives: Data Visualization

DataViz: Chart-Topping Songs as Graphs and Diagrams (From FlowingData)

Billboard ranked the top 100 songs since the creation of their Hot 100 list in 1958. The list is based on airplay and sales.

Chart-topping-songs

Tableau Customer Conference 2014 (TCC14): Keynote with Christian Chabot and Chris Stolte on the Art of Analytics

Tableau Keynote 2014

DataViz: Squaring the Pie Chart (Waffle Charts)

Readers:

Robert-Kosara-Tableau-Software-200x200In the past, I would have highly condemned pie charts without giving you much explanation why. However, Dr. Robert Kosara (photo, left), posted a great thought study of pie charts on his wonderful blog, EagerEyes.org, that I want to share with you.

Dr. Kosara is a Visual Analytics Researcher at Tableau Software, with a special interest in the communication of, and storytelling with, data. He has a Ph.D. in Computer Science from Vienna University of Technology.

Also, as part of his blog post, Robert offers an alternative way to create pie charts: using waffle charts or square pie charts.

Dr. Kosara is also one of the great minds behind Tableau’s new storytelling feature. I hope you enjoy his creative thoughts as much as I do.

Best Regards,

Michael

The Pie Chart

Dr. Kosara contends that pie charts are perhaps the most ubiquitous chart type; they can be found in newspapers, business reports, and many other places. But few people actually understand the function of the pie chart and how to use it properly. In addition to issues stemming from using too many categories, the biggest problem is getting the basic premise: that the pie slices sum up to a meaningful whole.

Touchstone Energy Corporation Pie Chart
Robert points out that the circle (the “pie”) represents some kind of whole, which is made up of the slicesWhat this means is that the pie chart first and foremost represents the size relationship between the parts and the entire thing. If a company has five divisions, and the pie chart shows profits per division, the sum of all the slices/divisions is the total profits of the company.

Five Slices

 

If the parts do not sum up to a meaningful whole, they cannot be represented in a pie chart, period. It makes no sense to show five different occupations in a pie chart, because there are obviously many missing. The total of such a subsample is not meaningful, and neither is the comparison of each individual value to the artificial whole.

Slices have to be mutually exclusive; by definition, they cannot overlap. The data therefore must not only sum up to a meaningful whole, but the values need to be categorized in such a way that they are not counted several times. A good indicator of something being wrong is when the percentages do not sum up to 100%, like in the infamous Fox News pie chart.

The Infamous Fox News Pie Chart

Fox News Pie Chart

In the pie chart above, people were asked which potential candidates they viewed favorably, but they could name more than one. The categories are thus not mutually exclusive, and the chart makes no sense. At the very least, they would need to show the amount of overlap between any two (and also all three) candidates. Though given the size of the numbers and the margin of error in this data, the chart is entirely meaningless.

When to Use Pie Charts

Dr. Kosara points out that there are some simple criteria that you can use to determine whether a pie chart is the right choice for your data.

  • Do the parts make up a meaningful whole? If not, use a different chart. Only use a pie chart if you can define the entire set in a way that makes sense to the viewer.
  • Are the parts mutually exclusive? If there is overlap between the parts, use a different chart.
  • Do you want to compare the parts to each other or the parts to the whole? If the main purpose is to compare between the parts, use a different chart. The main purpose of the pie chart is to show part-whole relationships.
  • How many parts do you have? If there are more than five to seven, use a different chart. Pie charts with lots of slices (or slices of very different size) are hard to read.

In all other cases, do not use a pie chart. The pie chart is the wrong chart type to use as a default; the bar chart is a much better choice for that. Using a pie chart requires a lot more thought, care, and awareness of its limitations than most other charts.

Alternative: Squaring the Pie

A little-known alternative to the round pie chart is the square pie or waffle chart. It consists of a square that is divided into 10×10 cells, making it possible to read values precisely down to a single percent. Depending on how the areas are laid out (as square as possible seems to be the best idea), it is very easy to compare parts to the whole. The example below is from a redesign Dr. Kosara did a while ago about women and girls in IT and computing-related fields.

Kosara Square Pie

Links to Examples of Waffle Charts

I did a little Googling and found a few great examples of Waffle Charts. I have provided links to examples in Tableau, jQuery R and Excel.

Squaring The Pie

Sources:

Dataviz Design & Infographics: Abby & Chris’ Wedding

Readers:

This clever wedding package was designed by Abby Ryan Design. The concept was to reflect the playfulness of Abby and Chris’s food truck wedding. First, a 18 x 24 screen printed wedding infographic was created that works as an invitation, program and menu. The poster was designed with the rehearsal dinner invitation as the bottom section so it could be removed for guests only coming to the wedding. Next, the save the date card, which focuses on important events in Abby and Chris’s relationship with the final date being their wedding. A wedding website was also created to keep their guests informed.

I loved the creativity behind this and just had to share.

Best Regards,

Michael

About Abby Ryan Design

Abby Ryan Design is a Fishtown/Philadelphia-based design studio, known for its clean design sensibility with an emphasis on client relationships and open lines of communication. Whether you are looking for a full branding campaign, or help with a small spot illustration, Abby Ryan Design will always give it 110%. With more than a decade of experience and a diverse roster of clients, they will work closely with you to meet all your design needs, from branding to business collateral, digital media to illustration/infographics and everything in between.

Their Process

Abby Ryan Design are visual communicators. They know all pieces of your message must gel for maximum efficiency. They will take a big-picture approach to ensure your design objectives fit into your overarching goal. Design trends and technology are constantly changing, so they will ensure your message is staying current and relevant with your audience. Clients receive personal attention throughout the entire creative process as well as access to our strong network of experts in development, writing, production, and printing. There are many paths they follow to achieve success, all loosely follow these steps:

Research

In order to successfully communicate with your clients it is essential to become an expert in your industry. Abby Ryan research the latest trends and highlights what’s working and what’s not to find the most effective ways to reach your audience.

Conceptualize

Research sparks the creative process, exploration helps to get it where it needs to go. Ideas come from all different places, whether it be surfing the web, creating word maps, or taking a shower. Many ideas are sketched out and thought through before they hit the computer screen.

Realize

Once all concepts are thoroughly explored, the final direction emerges. During this phase the team assembled for your project work together to complete all of the grunt work. Abby Ryan will share with you the progress and look for any feedback you may have to ensure you are happy with the final piece.

01_invite_full1 02_invite_tubes1 03_invite_detail_3 05_invite_detail_2 06_std 07_website1

 

 

 

Jock Mackinlay and Tableau’s Research Team is Building Tomorrow’s UX for Data

Readers:

I thought I would present some interesting information visualization research being conducted at Tableau Software by Jock Mackinlay (photo, right) and his research team.Jock Mackinlay. Source: Tableau Software

Mr. Mackinlay is an information visualization expert and Vice President of Visual Analysis at Tableau Software. With Stuart K. Card, George G. Robertson and others he invented a number of Information Visualization techniques. [1] Mr. Mackinlay, joined Tableau in 2004 after 18 years specializing in data visualization at Xerox PARC.

Tableau Software was born of academic research, and as the company continues to grow, it is building an R&D division to help build a pipeline of innovation. Jock, who heads up the research team, explains how it works and what his team is working on.

I cite references (most of this blog post is based on Derrick Harris’ interview with Mr. Mackinlay in Gigaom) after this blog post for those of you who want to delve deeper into what Jock’s team is doing.

Best regards,

Michael

Tableau Software and Their Research Culture

Tableau LogoTableau Software is many things: a fast-growing thorn in the side of legacy analytics vendorsstock-market gold and the poster child for the next generation of user-friendly data analysis, among them. It’s also a company with a deeply rooted and growing research culture that’s responsible for nearly everything users see when they open its popular visualization application. [2]

Tableau itself is the product of a Stanford Ph.D. dissertation by co-founder and Chief Development Officer Chris Stolte, in conjunction with his then-professor and eventual co-founder Pat Hanrahan. Their project, called Polaris, combined a structured query language with a declarative language for describing data visualization. When they commercialized the research by founding Tableau, that combination – which came together into a technology called VizQL – became the defining feature of the drag-and-drop Tableau experience.

However, the true value of what Stolte and Hanrahan created wasn’t just that let it let mainstream users query data visually and generate graphs, said Mackinlay. There had been a lot of research around ideal ways to visualize data — including his own — but they often focused on customized views of a single problem or type of analysis.“The real power [of Tableau] was to go through a bunch of different views to answer one question,” Mackinlay said. “All you have to be an expert at is your data and the questions you want to ask of it.”

The new research division within Tableau (technically, it was really created about a year and a half ago) is trying to imagine and create the next set of technologies that change the way data analysis is done. The five-person team, which Mackinlay heads, consists of four visualization experts (including Mackinlay), a couple of whom are also specialize in statistics and one of whom specializes in high-performance computing. The fifth member specializes in natural-language processing and computer graphics.

Like most research divisions, the team writes academic papers and works on some projects that might not be applicable for years, but Mackinlay made it pretty clear that the researchers expect everything they’re doing could be commercialized. If there was one thing that separated the famous Bell Labs from Xerox PARC or even Microsoft Research, it’s that Bell was really good at doing really good research that made its way into products, he said. Good research labs need to find the middle ground between nearsighted product upgrades and pie-in-the-sky ideas and, he explained, “You have to have absolutely no gap between the research scientists … and the people who are actually doing the work.”

A still image of an interactive Story Points slideshow. Source: Tableau Public user Matt Francis

Research Leads to Tableau Story Points Feature

Robert-Kosara-Tableau-Software-200x200It’s at a much, much smaller scale than Bell Labs, but Mackinlay thinks Tableau is following down that right path. For example, he said, the Story Points feature in the latest release of the company’s software, allows users to create data slideshows, was the result of tight work between the product team and researcher Robert Kosara (photo, right), who had been doing research into this area for years. As data volumes, dataset complexity and user sophistication all increase, Mackinlay said systems-level research into data processing (including how to optimize for increased client-side computing power) has and will continue to help deliver a smooth user experience.He’s understandably less forthcoming about what, specifically, we can expect to see from Tableau in the near term, but Mackinlay did discuss a few areas of interest. One is making it easier to use aesthetically pleasing icons rather than text labels in charts, an area where he and colleague Vidya Setlur (the aforementioned NLP and graphics specialist) recently published a paper. He’s also interested in text analysis and NLP, and generally adding new types of visualizations — some of which those types of analysis will help enable. For example, “node-link diagrams” (aka graphswill happen, he said, although he can’t put an exact data on when.icons

Mackinlay also suggested that Tableau might expand beyond its current product lineup, which is essentially the same software delivered via the desktop (free and paid), server or cloud. “We can make our existing products easy to use,” Mackinlay said. “We can also make new products that are easy to use — perhaps radically easier than our existing products.”

Although the word “easy” is kind of a misnomer, it’s one that’s used to describe Tableau and other user-friendly software quite often. “Easy” connotes shallowness, Mackinlay said, making an analogy to the evolution of the telephone. Phones have evolved a great deal from those where users just rang the operator, to rotary phones, and now to modern smartphones. With every iteration, manufacturers had to strike the right balance maintaining a recognizable experience but also adding more capabilities.

“We use the two words ‘simple’ and ‘useful,’” he said. “… If you don’t make sure you’re useful, people just aren’t going to stick with you.”

—————————————————————————–

References

[1]  Jock D. Mackinlay, Wikipedia.com, http://en.wikipedia.org/wiki/Jock_D._Mackinlay.

[2] Derrick Harris, A tiny research team at tableau is building tomorrow’s UX for data, Gigaom, July 7, 2014, http://gigaom.com/2014/07/07/a-tiny-research-team-at-tableau-is-building-tomorrows-ux-for-data/.

DataViz: Graphing the ALS Ice Bucket Challenge

Bucket of IceReaders:

As anyone currently on social media knows, the ALS Ice Bucket Challenge has turned into a fun and successful way to help fight Amyotrophic Lateral Sclerosis or ALS. From neighborhood driveways and city streets to Facebook, Twitter and Instagram, people everywhere can be seen dumping buckets of ice water on their heads to raise awareness and funds to fight ALS.

Children, adults and celebrities alike are joining the social media phenomenon to fight back against a disease that currently has no treatments or cures. “We have been moved beyond words by the power of one family’s ability to make such a meaningful difference in the fight against a disease that has taken too many lives,” said MDA President and CEO Steven M. Derks. “All of us at MDA are incredibly grateful to everyone who has taken the ALS Ice Bucket Challenge to raise awareness and donations for ALS. It will take all of us working together to find treatments and cures, and MDA will not rest until we end ALS.”

The viral ALS Ice Bucket Challenge started when 29-year-old Pete Frates, diagnosed with ALS in 2012, posted an ice bucket video on social media and challenged a few friends to follow his lead. The #ALSIceBucketChallenge has since become a social media sensation, sweeping the country with compassion and support. “Increased awareness about ALS is critical to help us learn more about the disease,” Derks said. “But what we need more than ever is action. Together, our collective actions can translate into significant progress against ALS. We hope everyone will join us to fight back by making a donation at mda.org and participating with us at a local MDA event in your community.”

 

Vikesh KhannaVikesh Khanna (photo, right), took a rather unique approach to the ALS Bucket Challenge using data visualization. Mr. Khanna, is a Computer Science Masters student at Stanford University. He was born and brought up in Haridwar, a small religious town in North India. He likes computer programming, reading, badminton, music and wine. If you’re looking for his official symbol, that’d be a crashing Zeppelin.

Vikesh came up with an idea of visualizing all of the people who have taken the ALS Ice Bucket Challenge, who called that person out to do so, and any photos or videos associated with it. Using his application, you can interactively select a person to see how they did the challenge, who they called out to do it, and even the associated video. Below is a screenshot of Vikesh’s data visualization.

Image-1 So, to test Vikesh’s application, I decided to see what Facebook’s Chief Operating Officer, Sheryl Sandberg, did for her ALS Ice Bucket Challenge. So, using the search page of Vikesh’s application, I searched for “Sheryl Sandberg.” The following information appeared (see two screenshots below). You will see information like: who challenged her (Mark Zuckerberg), how long it has been since she was challenged (278.9 hours), has she completed the challenge (she hasn’t yet), and her popularity score (493.914825). 

Sheryl Sandberg - Graph Sheryl Sandberg - Find

For you folks that want to try Vikesh’s application, I recommend you try it on an iPad using the Safari browser. I had problems with Internet Explorer, Google Chrome, and Mozilla Firefox.

Here is a link to his application.

I thought I would finish this blog post by provided you some more information about ALS. Even if you don’t want to have a cold bucket of ice water dumped on your head, please donate.

Best regards,

Michael

 

What is amyotrophic lateral sclerosis?

ALS is a disease of the parts of the nervous system that control voluntary muscle movement. In ALS, motor neurons (nerve cells that control muscle cells) are gradually lost. As these motor neurons are lost, the muscles they control become weak and then nonfunctional.

The word “amyotrophic” comes from Greek roots that mean “without nourishment to muscles” and refers to the loss of signals nerve cells normally send to muscle cells. “Lateral” means “to the side” and refers to the location of the damage in the spinal cord. “Sclerosis” means “hardened” and refers to the hardened nature of the spinal cord in advanced ALS.

In the United States, ALS also is called Lou Gehrig’s disease, named after the Yankees baseball player who died of it in 1941. In the United Kingdom and some other parts of the world, ALS is often called motor neurone diseasein reference to the cells that are lost in this disorder.

Who gets ALS?

ALS usually strikes in late middle age (the late 50s is average) or later, although it also occurs in young adults and even in children, as well as in very elderly people. Some forms of ALS have their onset in youth. Men are slightly more likely to develop ALS than are women. Studies suggest an overall ratio of about 1.2 men to every woman who develops the disorder.

What causes ALS?

Years ago, it was widely believed that there might be one cause to explain all cases of ALS. Today, doctors and scientists know that can’t be the case, and they’re working to identify the multiple causes of the disorder. One thing they do know is that ALS cannot be “caught,” or transmitted from one person to another.

The causes of the vast majority of ALS cases are still unknown. Investigators theorize that some individuals may be genetically predisposed to developing the disease, but only do so after coming in contact with an environmental trigger. The interaction of genetics and environment may hold clues as to why some individuals develop ALS.

Although the majority of ALS cases are sporadic, meaning there is no family history of the disease, about 5 to 10 percent of cases are familial, meaning the disease runs in the family. A common misconception is that only familial ALS is “genetic.” Actually, both familial and sporadic ALS can stem from genetic causes. And some people who have a diagnosis of sporadic ALS may carry ALS-causing genetic mutations that can be passed on to offspring. A genetic counselor can help people with ALS understand inheritance and any associated risks for family members.

What are the symptoms of ALS?

ALS results in muscles that are weak and soft, or stiff, tight and spastic. Muscle twitches and cramps are common; they occur because degenerating axons (long fibers extending from nerve-cell bodies) become “irritable.” Symptoms may be limited to a single body region, or mild symptoms may affect more than one region. When ALS begins in the bulbar motor neurons, the muscles used for swallowing and speaking are affected first. Rarely, symptoms begin in the respiratory muscles.

As ALS progresses, symptoms become more widespread, and some muscles become paralyzed while others are weakened or unaffected. In late-stage ALS, most voluntary muscles are paralyzed.

The involuntary muscles, such as those that control the heartbeat, gastrointestinal tract and bowel, bladder and sexual functions are not directly affected in ALS. Sensations, such as vision, hearing and touch, are also unaffected.

About 50 percent of people with ALS develop some degree of cognitive (thinking) or behavioral abnormality. Usually, cognitive and behavioral symptoms in ALS range from mild (such that only close family members may notice a difference) to moderate.

What is the life expectancy in ALS?

Each person’s disease course is unique. There are a number of examples of people who are leading productive and active lives more than two decades after an ALS diagnosis.

Standard longevity statistics citing an average survival time of three to five years after diagnosis may be somewhat out of date because changes in supportive care and technology — especially for breathing and nutrition — may help prolong life.

What can be done about ALS?

Medical interventions and technology have vastly improved the quality of life for people with ALS, by assisting with breathing, nutrition, mobility and communication. Proper management of symptoms, and proactive use of medical interventions and equipment, can make a positive difference in day-to-day living, and potentially may lengthen survival. The FDA-approved drug riluzole (brand name Rilutek) has been shown to slightly increase longevity.

What is the status of ALS research?

A number of strategies and approaches are being tested around the world, both in the laboratory and in human clinical trials. MDA’s basic science program is constantly pursuing new avenues of research to understand the underlying causes of ALS, with a sharp focus on developing treatments.

As of 2012, intense research is being conducted on genetic factors in ALS, the role of the immune system in ALS, and the role of cells other than nerve cells in this disease. In addition, many medications and other treatments are being tested for potential benefits in ALS. For details about current ALS research, go to Research and Clinical Trials.

Source:

MDA Website, http://mda.org/disease/amyotrophic-lateral-sclerosis/overview.

 

WIRED: A Redesigned Parking Sign So Simple That You’ll Never Get Towed

web-snow-day-1

Your car gets towed, and who do you blame? Yourself? God no, you blame that impossibly confusing parking sign. It’s a fair accusation, really. Of all the questionable communication tools our cities use, parking signs are easily among the worst offenders. There are arrows pointing every which way, ambiguous meter instructions and permit requirements. A sign will tell you that you can park until 8 am, then right below it another reading you’ll be towed. It’s easy to imagine that beyond basic tests for legibility, most of these signs have never been vetted by actual drivers.

Like most urban drivers, Nikki Sylianteng was sick of getting tickets. During her time in Los Angeles, the now Brooklyn-based designer paid the city far more than she would’ve liked to. So she began thinking about how she might be able to solve this problem through design. She realized that with just a little more focus on usability, parking signs could actually be useful. “I’m not setting out to change the entire system,” she says. “It’s just something that I thought would help frustrated drivers.” [1]

Sylianteng notes: [2]

I’ve gotten one-too-many parking tickets because I’ve misinterpreted street parking signs. The current design also poses a driving hazard as it requires drivers to slow down while trying to follow the logic of what the sign is really saying. It shouldn’t have to be this complicated.

The only questions on everyone’s minds are:
1. “Can I park here now?”
2. “Until what time?”

My strategy was to visualize the blocks of time when parking is allowed and not allowed. I kept everything else the same – the colors and the form factor – as my intention with this redesign is to show how big a difference a thoughtful, though conservative and low budget, approach can make in terms of time and stress saved for the driver. I tried to stay mindful of the constraints that a large organization like the Department of Transportation must face for a seemingly small change such as this.

01 two-step

The sign has undergone multiple iterations, but the most recent features a parking schedule that shows a whole 24 hours for every day of the week. The times you can park are marked by blocks of green, the times you can’t are blocked in a candy-striped red and white. It’s totally stripped down, almost to the point of being confusing itself. But Sylianteng says there’s really no need for the extraneous detailed information we’ve become accustomed to. “Parking signs are trying to communicate very accurately what the rules actually are,” she says. “I’ve never looked at a sign and felt like there was any value in knowing why I couldn’t park. These designs don’t say why, but the ‘what’ is very clear.”

Sylianteng’s design still has a way to go. First, there’s the issue of color blindness, a factor she’s keenly aware of. The red and green are part of the legacy design from current signs, but she says it’s likely she’d ultimately change the colors to something more universal like blue. Then there’s the fact that urban parking is a far more complex affair than most of us care to know. There’s an entire manual on parking regulations; and Sylianteng’s design does gloss over rules concerning different types of vehicles and space parameters indicating where people can park. She’s working on ways to incorporate all of that without reverting back to the information overload she was trying to avoid in the first place. [1]

redesigned-parking-inline2

Sylianteng also posted on her blog an illustration of the problem in terms of biocost, as part of her Cybernetics class with Paul Pangaro. [2]

Biocost_ParkingSign

Sylianteng has been going around Manhattan and Brooklyn hanging up rogue revamped parking signs. “A friend of mine called it functional graffiti,” she says. She’ll stick a laminated version right below the city-approved version and ask drivers to leave comments. In that way, Sylianteng’s design is still a ways away from being a reality, but so far, she’s gotten pretty good feedback. “One person wrote: ‘The is awesome. The mayor should hire you.’” [1]

————————————————————————

Sources:

[1] Liz Stinson, A Redesigned Parking Sign So Simple That You’ll Never Get Towed, Wired, July 15, 2014, http://www.wired.com/2014/07/a-redesigned-parking-sign-so-simple-youll-never-get-towed-again.

[2] Nikki Sylianteng, blog, http://nikkisylianteng.com/project/parking-sign-redesign/.

Small Multiples, Tableau and Ben Jones

Readers:

My BI world is changing a bit as I move more towards using Cognos and Tableau at work. In particular, I have a lot of status reports and dashboards to create for my leadership and I have been doing these mostly in Tableau.

I had a situation recently where I wanted to create a small multiples chart versus using a 3D Bar Chart that already existed. I have created small multiples charts fairly easily in Ben JonesMicroStrategy in my previous work, but have never created one before in Tableau. I reached out to Ben Jones (photo, right) at Tableau. I have been a big fan of Ben’s DataRemixed blog for quite some time and have blogged about Ben many times in the past. Ben was gracious enough to create a simple example small multiples chart for me to use to accomplish what I wanted to visualize. I was really impressed that Ben and Tableau did not put me through any red tape for him to help me. He saw I had a need and he helped me.

Much thanks to Ben for his help and I hope this example is useful to you.

Best Regards,

Michael

Small Multiples

Edward TufteA small multiple (sometimes called trellis chart, lattice chart, grid chart, or panel chart) is a series or grid of small similar graphics or charts, allowing them to be easily compared. The term was popularized by data visualization pioneer, Edward Tufte.

According to Tufte (Envisioning Information, p. 67):

At the heart of quantitative reasoning is a single question: Compared to what? Small multiple designs, multivariate and data bountiful, answer directly by visually enforcing comparisons of changes, of the differences among objects, of the scope of alternatives. For a wide range of problems in data presentation, small multiples are the best design solution.

A Small Multiples Example by Andrew Gelman

One of the most well-known examples of the use of small multiples is Andrew Gelman’s analysis of public support for vouchers, broken down by religion/ethnicity, income, and state (see image below).

GelmanMr. Gelman is a professor of statistics and political science and director of the Applied Statistics Center at Columbia University. His books include Bayesian Data Analysis (with John Carlin, Hal Stern, David Dunson, Aki Vehtari, and Don Rubin), Teaching Statistics: A Bag of Tricks (with Deb Nolan), Data Analysis Using Regression and Multilevel/Hierarchical Models (with Jennifer Hill), Red State, Blue State, Rich State, Poor State: Why Americans Vote the Way They Do (with David Park, Boris Shor, and Jeronimo Cortina), and A Quantitative Tour of the Social Sciences (co-edited with Jeronimo Cortina).

Andrew has done research on a wide range of topics, including: why it is rational to vote; why campaign polls are so variable when elections are so predictable; why redistricting is good for democracy; reversals of death sentences; police stops in New York City, the statistical challenges of estimating small effects; the probability that your vote will be decisive; seats and votes in Congress; social network structure; arsenic in Bangladesh; radon in your basement; toxicology; medical imaging; and methods in surveys, experimental design, statistical inference, computation, and graphics.

[Click on Image to Enlarge]

Gelman Voucher Map Using Small Multiples

My Small Multiples Chart

Since I cannot show you what I used the small multiples chart for related to my job, I made an illustrative, simple example related to home sales in different regions for the past six months. Below is an example of my chart, which I created using Tableau.

[Click on Image to Enlarge]

Home Sales Small Multiples

Adding Trend Lines

One of the key features I wanted to use in my chart was to be able to show trend lines for each small multiple.

However, when I clicked on Trend Lines -> Show Trend Lines, I kept getting the following error message:

Trend Lines Error Message Panel

Ben pointed out that in my original chart, the Columns shelf, Month needed to be a Continuous data type (green pill) rather than a Discrete data type (blue pill).  If you click in the Month pill, you should be able to select “Change to Continuous” and then you should be able to add a trend line. This occurs because you can only calculate a trend line when two axes are involved. The way I had it set up, the Columns were just different categories or attributes, rather than continuous measures.

I thought this would be a nice tip to pass along.

I hope to be able to share more Tableau tips as I become more proficient with the tool.

12 JavaScript Libraries for Data Visualization

Readers:

This is from a blog post by Thomas Greco.

Thomas GrecoThomas is a web developer / graphic designer living in New York City. When Thomas isn’t striving towards front­end perfection, he enjoys hanging with friends, going to concerts, and exploring through the wilderness!

Thomas has provided twelve JavaScript frameworks that are extremely useful for data visualization. Thomas feels that a more heavy focus is being placed on JavaScript as a data visualization tool.

I tried the demos for these JavaScript frameworks and they are very impressive. I hope you enjoyed this information as much as I did.

Best regards,

Michael

Dygraphs.js

The Dygraphs.js library allows developers to create interactive charts using the X and Y axis to display powerful diagrams. The more data being parsed, the higher the functionality of the graph. That being said, Dygraphs was built for these visualizations to contain a multitude of views. For example, Dygraphs.js makes it capable to analyze separate portions of a data-set, such as specific months, in addition to the timeframe in its entirety. Also, the Dygraphs.js library is compatible across all major web browsers, and can responds to touch sensitivity, making it a thoroughougly solid choice as a data visualization framework.

D3.js

Eventually becoming the successor to Protovis.js, D3 is capable of creating stunning graphics via dynamically updating the DOM. An acronym for Data-Driven Document, D3.js makes use of chained methods when scripting visualizations, subsequently creating dynamic code that is also reusable. Due to its reliance on the DOM, D3 has been created in accordance with W3C web standards so that the library may render correctly across web browsers. Lastly, D3′s path generator function, defined as d3.svg.line(), gives developers the capability to produce a handful of SVGs by defining different paths, and their properties.

InfoVis

Commonly referred to as InfoVis, the JavaScript InfoVis Toolkit (JIT) also earned its stripes as a JavaScript library for data visualization. Equipped with WebGL support, InfoVis has been trusted by names like Mozilla and AlJazeera, showing its solidarity as a visualization tool. Along with the D3 framework, InfoVis also makes use of chained methods to manipulate the DOM, making it a reliable library for developers of any skill set.

The Google Visualization API

Hailing from the Google Developers Console (GDC), Google’s Visualization API can be called with barely any code. In addition to easy DOM modification, this Google API makes it easy for its user to easily define custom modifier functions that can then be placed into custom groups. Furthermore, this interface’s usability, matched with its support from the GDC’s open source network, place it among the top of the list of data visualization tools.

Springy.js

Springy.js is a JavaScript library that relies on an algorithm to create force-directed graphs, resulting in nodes reacting in a spring-like manner on the web page. Although Springy.js comes configured with a predefined algorithm, options such as spring stiffness and damping can easily be passed as parameters. Springy.js was developed by Dennis Hotson as a library for developers to build off of – a fact that he makes clear.

Polymaps.js

Polymaps.js makes use of SVGs to generate interactive web maps with cross browser compatibility in mind. At the heart of Polymaps lies vector tiles, which help ensure both optimal load speeds and optimal zoom functionality. Although it may come configured with components, Polymaps.js is easily customized, and is able to read data in the form of vector geometry, GeoJSON Files, and more. Check out the graph below of the U.S. created by the U.S. Census borough.

Dimple

This past January, the Dimple API was developed so that analysts at Align-Alytics could develop strong data visualizations without having to possess much development knowledge. That being said, Dimple makes it easy for anyone, analyst or not, to develop stunning, three dimensional graphics without any real JavaScript training. Moreover, dimplejs.org displays several demonstrations, which can be easily manipulated by one’s personal data to render a graph with the same configuration, but different values. So, if you, or anyone you know is trying to segway into the depths of JavaScript, then these examples are perfect for beginners to vist and poke around.

Sigma.js

For people looking to build highly advanced line graphs, Sigma.js provides an unbelievable amount of interactive settings inside its library, and also within its plug-ins. Hailing a motto that states “Dedicated to Graph Drawing”, those developing using Sigma.js cannot help but feel like they have chosen a reliable library to work with. Moreover, Sigma’s developers encourage people to re-configure this library and create plug-ins, which has resulted in a large open-source network. Having said all that, I was extremely pleased with various aspects of Sigma, and it is among my favorite libraries for creating graphical representations in JavaScript.

Raphael.js

The Raphael.js library was created with an emphasis on browser compatibility. The framework follows the SVG W3C Recommendation, which is a set of standards that ensure images are completely scalable and without pixelation. In addition to the use of SVGs, Raphael.js even reverts to the Vector Model Language (VML) if rendered in Internet Explorer browsers prior to IE9. Although VML is very rarely used today, the support for it does a great job of showing the attention to detail that the Raphael.js team placed on this project when developing the library.

gRaphaël

Although Raphael.js is a library used to for the creation of SVGs, it was not built with a total focus on the representation of large datasets. In turn, the gRaphaël JavaScript library was created. Weighing in at a mere 10KB, gRaphaël.js has proven to be a worthy extension to Raphael.js. Although it may have not been developed behind things like a force-driven algorithm, nor does it come pre-configured with any physics properties, gRaphaël is still a well respected library for reasons ranging from its cross-compatible SVG structure, to its ease of use. As long as it coincides with the task at hand, I believe that gRaphaël.js should always be looked at as a viable resource to complete a project.

Leaflet

Whether developing for a smartphone, tablet, or desktop, the Leaflet JavaScript library has ranked atop the list of interactive mapping libraries for several reasons. Lead by the founder of MapBox, Vladimir Agafonkin, the Leaflets team of developers worked to create a library “designed with simplicity, performance, and usability in mind.” Along with Polymaps, Leaflet shares the ability to render SVG pattens via vector tiles, however only Leaflet has been developed to support Retina display. Furthermore, Leaflet can interpret various forms of data such as GeoJSON, making it perfect for a number of tasks.

Ember Charts

For those who already use the juggernaut that is Ember.js, the developers at Addepar Open Source have created a few add-on libraries to extend the Ember experience: Ember Table, Ember Widgets, and Ember Charts. A child of Ember.js and D3.js, Ember Charts utilizes the properties of flat-design. Although limited, the library does have a handful of options that deal with properties such as color and size, making it fairly simple to create impressive visualizations. Nonetheless, Ember’s presence in the front end could really help Ember Chart’s popularity in the future.

Stephen Few: Why Do We Visualize Quantitative Data?

Readers:

Stephen_FewIt has been a while since I have discussed some of the latest creative thoughts on data visualization from Stephen Few. I have read all of Steve’s books, attended several classes from him,  and religiously follow his blog and newsletter on his website, Perceptual Edge.

For those of you who don’t know, Stephen Few is the Founder & Principal of Perceptual Edge. Perceptual Edge, founded in 2003, is a consultancy that was established to help organizations learn to design simple information displays for effective analysis and communication.

Steve has stated that his company will probably always be a company of one or two people, which is the perfect size for him. With 25 years of experience as an innovator, consultant, and educator in the fields of business intelligence and information design, he is now considered the leading expert in data visualization for data sense-making and communication.

Steve writes a quarterly Visual Business Intelligence Newsletter, speaks and teaches internationally, and provides design consulting. In 2004, he wrote the first comprehensive and practical guide to business graphics entitled Show Me the Numbers, now in its second edition. In 2006, he wrote the first and only guide to the visual design of dashboards, entitled Information Dashboard Design, also now in its second edition. In 2009, he wrote the first introduction for non-statisticians to visual data analysis, entitled Now You See It.

Here is his latest thoughts from his newsletter.

Best regards,

Michael

 

Why Do We Visualize Quantitative Data?

Per Stephen Few, we visualize quantitative data to perform three fundamental tasks in an effort to achieve three essential goals:

Web

These three tasks are so fundamental to data visualization, Steve used them to define the term, as follows:

Data visualization is the use of visual representations to explore, make sense of, and communicate data.

Steve poses the question of why is it that we must sometimes use graphical displays to perform these tasks rather than other forms of representation? Why not always express values as numbers in tables? Why express them visually rather than audibly?

Essentially, there is only one good reason to express quantitative data visually: some features of quantitative data can be best perceived and understood, and some quantitative tasks can be best performed, when values are displayed graphically. This is so because of the ways our brains work. Vision is by far our dominant sense. We have evolved to perform many data sensing and processing tasks visually. This has been so since the days of our earliest ancestors who survived and learned to thrive on the African savannah. What visual perception evolved to do especially well, it can do faster and better than the conscious thinking parts of our brains. Data exploration, sensemaking, and communication should always involve an intimate collaboration between seeing and thinking (i.e., visual thinking).

Despite this essential reason for visualizing data, people often do it for reasons that are misguided. Steve dispels a few common myths about data visualization.

Myth #1: We visualize data because some people are visual learners.

While it is true that some people have greater visual thinking abilities than others and that some people have a greater interest in images than others, all people with normal perceptual abilities are predominantly visual. Everyone benefits from data visualization, whether they consider themselves visual learners or not, including those who prefer numbers.

Myth #2: We visualize data for people who have difficulty understanding numbers.

While it is true that some people are more comfortable with quantitative concepts and mathematics than others, even the brightest mathematicians benefit from seeing quantitative information displayed visually. Data visualization is not a dumbed-down expression of quantitative concepts.

Myth #3: We visualize data to grab people’s attention with eye-catching but inevitably less informative displays.

Visualizations don’t need to be dumbed down to be engaging. It isn’t necessary to sacrifice content in lieu of appearance. Data can always be displayed in ways that are optimally informative, pleasing to the eye, and engaging. To engage with a data display without being well-informed of something useful is a waste.

Myth #4: The best data visualizers are those who have been trained in graphic arts.

While training in graphic arts can be useful, it is much more important to understand the data and be trained in visual thinking and communication. Graphic arts training that focuses on marketing (i.e., persuading people to buy or do something through manipulation) and artistry rather than communication can actually get in the way of effective data visualization.

Myth #5: Graphics provide the best means of telling stories contained in data.

While it is true that graphics are often useful and sometimes even essential for data-based storytelling, it isn’t storytelling itself that demands graphics. Much of storytelling is best expressed in words and numbers rather than images. Graphics are useful for storytelling because some features of data are best understood by our brains when they’re presented visually.

We visualize data because the human brain can perceive particular quantitative features and perform particular quantitative tasks most effectively when the data is expressed graphically. Visual data processing provides optimal support for the following:

1. Seeing the big picture

Graphs reveal the big picture: an overview of a data set. An overview summarizes the data’s essential characteristics, from which we can discern what’s routine vs. exceptional.

The series of three bar graphs below provides an overview of the opinions that 15 countries had about America in 2004, not long after the events of 9/11 and the military campaigns that followed.

graph-of-country-opinions

Steve first discovered this information in the following form on the website of PBS:

table-of-country-opinions

Based on this table of numbers, he had to read each value one at a time and, because working memory is limited to three or four simultaneous chunks of information at a time, he couldn’t use this display to construct and hold an overview of these countries’ opinions in his head. To solve this problem, he redisplayed this information as the three bar graphs shown above, which provided the overview that he wanted. Steve was able to use it to quickly get a sense of these countries’ opinions overall and in comparison to one another.

Bonus: Here is a link to where Steve discusses the example above on his website.

2. Easily and rapidly comparing values

Try to quickly compare the magnitudes of values using a table of numbers, such as the one shown above. You can’t, because numbers must be read one at a time and only two numbers can be compared at a time. Graphs, however, such as the bar graphs above, make it possible to see all of the values at once and to easily and rapidly compare them.

3. Seeing patterns among values

Many quantitative messages are revealed in patterns formed by sets of values. These patterns describe the nature of change through time, how values are distributed, and correlations, to name a few.

Try to construct the pattern of monthly change in either domestic or international sales for the entire year using the table below.

table-of-sales-data

Difficult, isn’t it? The line graph below, however, presents the patterns of change in a way that can be perceived immediately, without conscious effort.

graph-of-sales-data

You can thank processes that take place in your visual cortex for this. The visual cortex perceives patterns and then the conscious thinking parts of our brains make sense of them.

4. Comparing patterns

Visual representations of patterns are easy to compare. Not only can the independent patterns of domestic and international sales be easily perceived by viewing the graph above, but they can also be compared to one another to determine how they are similar and different.

In Summary

These four quantitative features and activities require visual displays. This is why we visualize quantitative data.

Follow

Get every new post delivered to your Inbox.

Join 263 other followers