By Sankalpa Dashrath; Research Associate, The Media Map project

The image populated my Facebook newsfeed on Diwali morning. It was being ‘liked,’ commented on, and sent around in mass emails. It soon grew into the feel-good story du jour; the consensus was that good had triumphed over evil and light had chased away the darkness. Diwali- the Indian festival of light – and its glittering brightness could be spotted from space this year. At least that’s what everyone was saying. A satellite image, purported to be released by NASA, showed a pretty, colorful map of India as seen from space on Diwali night.

As I watched the image go viral over the internet, I wondered about its source. A quick check on the internet confirmed my suspicions. This was a solid case of data misrepresentation. The image was not from NASA, but from the National Geophysical Data Center (part of the USDOC) website. And no, it wasn’t a single snapshot of a specific night but rather a composite of more than a decade of data imagery overlaid into a single data visual.

Someone somewhere had misinterpreted the image and lots of people everywhere were repeating the fallacy, making it grow stronger by the minute. The truth was lost amongst the hundreds of people who wanted to believe this ‘lie.’There are many unfortunate lessons to be learned from this story.

As the world grows increasingly impatient and attention spans decrease, data visualizations are replacing detailed reports. Data visualizations are just incomplete snapshots. They do not require viewers to really think about the information they contain. They are subject to interpretation and can be misleading, as evidenced from the Diwali debacle.

Social media allows half-truths to become solidified in people’s minds. Repetition by large numbers can validate an untruth. Google’s algorithms don’t check facts; they simply provide results with the most hits. Again, repetition by large numbers triumphs accuracy. The reason why stories go viral on social media is that everyone wants to post ‘breaking news’, so the traditional checks on data sources are neglected. Speed becomes the barometer of newsworthiness.

Lost in the intellectual debates, is the sad fact that these seemingly innocuous mistakes mask unpleasant facts. In this case, the areas on the Diwali image that are colored in blue and green used to be visible from space from 1992 to 1998 but cannot be seen any longer. So the picture is now darker, literally and metaphorically. While the wired world celebrates this fallacy, the situation for many un-connected Indians is getting worse, not better; The New York Times reports that, “Almost half of the Indian population has no access to the electricity grid, and many more people suffer hours without power”. Meanwhile, India’s ranking in the UNDP’s Human Development Index has steadily dropped from 126th in 2006 to 128th in 2007/08 and to 134th in 2010. But, instead of trying to analyze factual data from reputable institutions, the connected world is busy celebrating a fake picture with no source.

A version of this blog also appears on The Morningside Post

By Ericha Hager; Internews Center for Innovation and Learning Intern

I recently attended a US State Department  “Tech@State” conference. These daylong events occur quarterly and focus on different technology topics. This one addressed growing trends in data visualization. Data visualization, the visual representation of data, has been cited as an incredibly impactful and efficient way to transfer knowledge and meaning when done correctly. The amount of time it takes for people to absorb information through the visual representations of data is much faster than by reading data alone. In our increasingly fast paced and visually stimulated world, the ability to catch people’s attention in a quick, direct, and meaningful way is crucial to conveying messages and gaining support for causes.

The data visualization industry is becoming robust as data visualization is becoming a cornerstone element to any modern website. Companies and organizations are looking for ways to create stimulating and interactive ways to display data and apparently many of them are willing to pay for it. I have to admit that for the most part the Tech@State Data Visualization day seemed more like an opportunity for competing data visualization firms to self-promote and pitch their product. Most of the morning panelists were showing examples of data visualization that would completely deplete any medium sized company or organizational budget.  The rest of the day was primarily a show and tell of websites that have been restructured around a strong data visualization element without any tips and tricks on how to do your own data visualization in-house. From the standpoint of a newcomer to the field, I walked away with no new skills to test out, which was rather disappointing for a full day’s attendance.

One valuable thing about Tech@State is it gave me a better perspective as to what Internews’ Center for Innovation and Learning’s Media Map Project could achieve through well-developed data visualization. The Media Map website contrasts with the other websites I witnessed at Tech@State in that one of its primary purposes is data presentation. Media Map is designed to harness the power of data visualization to create a clear, easily navigable, well organized site that has the ability to display large amounts of information that can be cross referenced to draw conclusions about how media develops across countries and over time, and how media relates to development and to governance. There are a few key aspects of data visualization present on Media Map that were highlighted at the conference: the user has control over how the data is presented and what variables they want to test for; the data sets are available for download; and there is supporting text to augment the visualizations. While Media Map has integrated many successful aspects of effective data visualization, I think there is room to expand and change as people utilize it as a resource and identify ways in which it could be a more impactful research tool.

Data visualization can be very powerful and it is important that people develop the ability to critically process it. Data itself is very influential and, as I mentioned previously, data visualization allows people to ingest more information at a higher rate than average. This means  people need to become data literate to some degree, because the presentation of data can be misleading, data can be dirty, and even the data itself can be inaccurate and or simply false. As with all information, it is important to be aware of all influences, circumstances and omissions.  Overall, I think that data visualization is very impactful and look forward to the development of more universally accessible tools so everyone can leverage its utility.

Empowering local media worldwide

1640 Rhode Island Ave. NW, 7th Floor
Washington, DC 20036 (w)
+1 202 833-5740 (p)
+1 877 347-1522      
+1 202 833-5745 (f)