All posts from Simple Complexity

An example of a 3D, animated, globe visualization with a heatmap overlay to show temperature anomalies.
August 2009: One of the warmest on record
The world’s ocean surface temperature was the warmest for any August on record, and the warmest on record averaged for any June-August summer season, according to NOAA’s National Climatic Data Center in Asheville, N.C. World-wide records began in 1880. Shown here is a visualization of the August global temperature anomalies–or in other words, how the average temperature in August differs from the average climate of 1961-1990. Notice that in some areas, such as the central United States, temperatures were much cooler than average. But overall, land and ocean temperatures were several degrees above normal.
At the NOAA Environmental Visualization Laboratory, you can see an animated visualization of the temperature data.
A more detailed description of this visualization is available at NOAA News.

Image Credit: clickykbd
Very nice blog post from David Linthicum over at ebizq.
In the world of business intelligence, most of those charged with building BI systems set up data warehouses as the place where decision support data is deposited after being cleansed, aggregated, and restructured. This seemed like the best approach for years, considering that the data we needed to leverage for business intelligence was very different than operational data. . . The magic of combined data is that you have complete control over the data before it’s mined through a BI tool. In essence you’re creating instances of data that are leveraged by BI, and thus you can move from instance to instance as you see fit, and only place an instance of combined data on-line when it’s of the proper content and quality. Thus, the advantages of combined data would be control and quality, where the downsides would be latency. . . Rising up as a popular way to do BI is the notion of abstraction, or abstract data. This means that instead of creating another physical database, or data warehouse, that we leverage the data where it exists, binding it to a virtual database schema that exists only in middleware.
Read the full article here.
The comment by Dyke Hensen is interesting.
Historically, there has always been a challenge when traditional business intelligence meets transactional processing models. “Real time” BI has been both a dream and an oxymoron for years in the data warehousing marketplace. One key reason for data warehousing is the fact transactional systems were inappropriate or inadequate for historical data analysis. Siloed application data didn’t play well with other siloed data and creating BI-like analysis in a transactional application would grind operational performance to a crawl or create glacial query response.
Read the rest here.
What do you think?

Source: Energy Information Administration | The Washington Post

A really nice use of a stacked bar chart in the Monday, October 5, 2009, Washington Post. The infographic accompanied an article by Juliet Eilperin, U.S. Policitical Realities Constrain International Climate Talks.
Stacked bar charts usually do not have a common baseline for its components. It is usually easy to see the first segment that starts at the x-axis, but other segments require you to subtract the previous segments from the later segments highest point.
The message of this infographic is that Energy-related CO2 emissions from Developed Countries is projected to remain almost the same (if you look closely, there is a small increase). At the same time, emissions from developing countries are expected to grow.
The use of the stacked bar chart allows us to see the total amount of projected emissions. The bar chart creates three baselines. One baseline is the normal one at 0. The second is around 13, the level of developed countries. The third is right below 30, which represents the total emissions. The increased emissions is then broken out to show where it is coming from. The map showing developed versus developing countries provides context.
A callout states that “by 2030, developing countries are projected to contribute 63% of the world’s energy-related CO2.” An issue that is not examined is whether this is a fair share or not. Looking at the map in the infographic, the developing countries appear to me to take up more than 63% of the surface area (perhaps population would be a better measure). Perhaps the story should be that developing countries are beginning to create “their share” of emissions. Perhaps developed countries are using more than their share, and should be targeting emission declines. The answer to this is influenced by your perspective.
In any event, good job Washington Post!! I think this is an extremely effective infographic.
What are your thoughts?

Yesterday, I wrote a post about an effective infographic in the Monday, October 5, 2009, Washington Post. Today, I want to point out a missed opportunity to use a data visualization to illustrate a point.

On the left is a scan of the front page of the Sports Section of the Washington Post. The point of the data is that the Redskins (who are doing quite poorly this year) do much better in the second half of the game than the first.
On the right is an alternate visualization of the data. I used a bar graph showing a percentage comparison of first half versus the second half of the game in various categories. I also grouped together the categories where high values are good performance and the data where high values are poor performance (I basically moved the opponent’s categories below the fumbles).
I think with the remade visualization the fact that second half performance in every category is better in the second half is much easier to see. I think particularly because each of the categories are on different scales, standardizing them as parts of a whole is useful.
I’m interested in your opinions.

Let’s try a simple experiment. Look at the two graphics below. Try to determine how much bigger the second item in each graphic is than the first.

How much longer is the second line than the first line?

How much larger is the surface area of the second circle than the first?

Scroll down to see the answers. . .

Editor’s Note: I made an error in the original posting of this message. I’m going to leave the original text of my mistake in tact, but please see the comments below for a more correct discussion of the difference. The second circle has 4 times the width and 16 times the surface area. My original intention was to show a smaller difference in the circle’s surface area.
The second line is two times the first line. The second circle has four times the surface area of the first circle.
Most people find comparing the lines a lot easier than comparing the circles. It has to do with automatic cognitive processes that occur in the brain, generally without conscious thought. People are much better at estimating the length of lines than they are at estimating/comparing surface areas in shapes.
It is something you should think about when constructing data visualizations.
How did you do?

A little while ago, I wrote a blog post about some excellent customer service stories. It was a popular post, so here are a few more stories:

Reedy Carpets in Orlando stands behind their carpet.
We contacted Reedy Carpet who immediately contacted Shaw Carpets. They arranged for an inspection and upon receiving the report they volunteered to replace the carpet entirely even though “pooling” is not covered under their warranty. . .This experience tells me that I was correct in my choice of a dealer and that Shaw Carpets still believes in and practices good customer service.

Strong Elder Care Businesses Focus on Customer Service During Challenging Economic Times – This is a press release, but still has some good information.
While businesses need to focus carefully on cutting expenses where they can, for a company to survive, customer service needs to be bolstered, not cut. aQuire Training Solutions has developed a series of new courses on basic – and advanced – customer service skills. These courses are designed for caregivers and other staff working in the senior care environment: home care agencies, assisted living communities, nursing facilities and more.

A custom canvas company that proves they know their customers. Carrie Lewis, an artist, raves about Signature Canvas. The company provides excellent customer service and included a customized certificate for the artist.
Talk about a Class Act! If you’re looking for a company that gives individual attention to every order, that custom builds supports using first rate materials or if you just like to try new things, give Signature Canvas a try. You will not be disappointed in the product or in the way you are treated.

Customer Service in Switzerland.
I had such a good experience and I want to share it with you . . . Monday, I went to H&M to buy some clothes for my little daughter. I began to look around and found stuff for me as well . . . Anyway, long story short, I bought a nice cardigan and put it in the washing machine. The tag was saying to clean it by machine on the wool cycle, as I did. It came out so small . . . I looked for the bill, and my dear husband throw it away . . . I decided to send an email to H&M and told them my story . . . The very next morning, I received an email back. The lady was going to take care of it and contact the quality department . . . She told me to bring it back where I bought it, she will contact the Manager and they will give me my money back!

Where Has This UPS Envelope Spent The Last 14 Years? The UPS lost a package, found it 14 years later, the contents are missing, and they want to honor their claim.
This week, Paul received a package back from UPS that had somehow never reached its destination. That’s not so unusual. What was unusual was how long it had wandered off for. He had mailed the next day air envelope at least fourteen years ago. . . While it’s curious that the package went missing for so long, lost its contents, and then still found its way back to Paul, we are rather impressed that UPS is willing to pay his claim if he does ever manage to figure out what was in that envelope… a decade and a half ago.

Excellent customer service at Graham & Brown.
My wife took a few hours to browse and add items to her shopping cart to review with me later. The next day we had time to look over her selections. Unfortunately all of the items in her cart had disappeared. She even created an account to make sure that the items in her cart would be saved.
She called Graham & Brown customer service to tell them about the problem. They quickly apologized and explained that the site had just launched. Their web team was working out some bugs, and they were glad to hear our feedback. Customer service also offered to send us the wallpaper samples that we chose free of charge. As we make our final wallpaper selection I’ll be happy to give Graham & Brown my business because of this great experience.

What a great simple visualization.

Via Cool Infographics

Here are ten infographics that recently caught my eye.
The Most Expensive Cities (Via: Digital Monkey)

Are We Over the Worst? (Via: Digital Monkey)

You’re Always At Most 107 Miles From A McDonald’s (Via: Consumerist)

Día de Duarte (Via: Jesus Sabino)

Independencia Dominicana (Via:

I found the following visualization on the web. For some reason, it caught my eye.

Infographic by Jeff Beckham

The caption says:
Colt McCoy has thrown for 9,732 yards in his career at Texas. That’s 5.5 miles of spirals. If he throws for another 3,696 yards this season, he’ll have covered the distance from his hometown of Buffalo Gap to his high school in Tuscola.
I don’t follow College Football, so I had no idea who Colt McCoy was. According to Wikipedia:
Colt McCoy is the starting quarterback for the University of Texas. McCoy won the 2008 Walter Camp Award and was the 2008 Heisman Trophy runner-up. McCoy is considered a first round NFL draft prospect.
He has finished 3 years at the University of Texas, so he is averaging 3,244 yards per year. Hitting the target of an additional 3,696 yards seems possible.
When I saw the two arcs leaving the same point but traveling different paths, I thought I was going to write about how when a line curves, there can be visual difficulties determining proportions between the two arcs. In this image, for the same amount of vertical drop, the outermost arc will travel a longer distance.
This made me think that the best way to show the data is via a map, as we are talking about geographic distances.
I went to Google driving directions and entered Buffalo Gap, Texas and Tuscola, Texas, and I found out that via roads, the distance between the two places is 7.6 miles (13,428 yards). Aha, this is the distance that the original infographic shows as the distance between Buffalo Gap and Tuscola.

I think that seeing this path on a map breaks the football passing metaphor. Passes do not generally take right hand turns.
Using Google Earth, I found out that the city centers of Buffalo Gap and Tuscola are only 5.16 miles apart via a straight line, as the football flies, so to speak. So Colt McCoy has likely already thrown the distance from his hometown to Tuscola.
I think the lesson here is that when the appropriate display mechanism is chosen, information that you might otherwise miss becomes clear.

Click to expand

Indeed.com, a search engine for jobs, has some interesting unemployment visualizations. The one at right is quite effective.
An important aspect as a consumer of data is to ask yourself what does the data mean. When I first saw this graphic, my first impression was that this showed me the difficulty of finding employment in certain geographic areas. Is this the case?
What data is presented?
The data in the table represents the ratio of job posts to unemployed people. The graph clearly states that unemployment data comes from the Bureau of Labor Statistics. I wish the graph equally well defined where the job post numbers come from. I believe job posts are the number of search results from Indeed.com filtered by the city name.
When first looking at the graph, my thought process was that I was looking at # of available jobs versus unemployment rate. How well do the search results at Indeed act as a surrogate for the number of open jobs? How long does a job posting stay in the search results beyond the time that the position is open? How much of the real job market does Indeed capture? This is not known based on information I could find on the site. The lesson here, I think, is that it is important to stop and examine what exact data is being shown on a visualization.
The next thing I thought about was whether it is better to show the ratios of job posts versus unemployed people or present a dot plot showing the number of job posts versus the unemployment number. It might be difficult to construct a dot plot with that many data points and still make it readable. The answer depends on one’s objective.
An important thing to realize is that a job market where there are 10,000 jobs and 10,000 unemployed in a particular city is quite different from a situation where there are 1 million jobs and 1 million unemployed in a city. Although the ratio in both is 1 to 1, the difficulty of finding a job in the two situations may be quite different. The unemployment rate would have an impact.
These are not necessarily criticisms of the data presented, just an indication that the consumer needs to be careful what conclusions can be drawn.
One thing I like is that this graphic does provide some context through showing how much of a change this months results were versus the previous months. This provides you some directionality of change data that could be interesting.

Simple Complexity is on a short vacation break. In the meantime, here is some infographic eye candy.

Source: Infojocks – full size poster available.

Simple Complexity is on a short vacation break. In the meantime, here is some infographic eye candy.

Source: Infojocks