Data Literacy

It has been almost a year since the great IATUL conference on Data, probably one of the best conferences I have ever attended. I was thinking about DiRT today, and remembered that the brilliant Lisa Johnston had done a write-up of E-Science tools for Research in the column that she writes for SLA’s SciTech News.

One of my favorite science librarians, Michael Fosmire, has given some talks about the importance of information fluency in an e-science context. These are some of the competencies he identifies as required of e-scientists:

• Discovery and application of data in repositories, and the ability to import and convert it to a suitable format for further processing,
• Data management and organization: understanding the life cycle of data and creation of standard operating procedures for processing it,
• Understanding metadata and the structure and purpose of ontologies to facilitate better sharing,
• Data curation and re-use: recognizing that data may have purposes other than the original one for which it was intended and understanding that data curation is a complex and often costly process,
• Cultures of practice:  recognition of the practices, values, and norms of one’s chosen field as well as relevant data standards,
• Data preservation:  recognition of the benefits and costs,
• Data analysis:  becomes familiar with the basic analysis tools of the discipline,
• Data visualization:  understanding the advantages of different types of visualizations, and
• Ethics:  develops an understanding of intellectual property, privacy, and confidentiality issues and appropriately acknowledges external sources.

A recent paper, Carlson, Jacob, Michael Fosmire, C.C. Miller, and Megan Sapp Nelson. “Determining Data Information Literacy Needs: A Study of Students and Research Faculty.” 2011 points out the need to use ethnographic research to inform decisions based around data literacy. Great stuff!

Edward Tufte

I was lucky enough to attend an Edward Tufte workshop last month,and had a meeting today with one of the archivists at my library to discuss it. It reminded me about several things that were great about it.

• Sparklines  are probably the number one best takeaway. They are basically graphic images that get embedded into the text. You have probably seen more of these than you think–for example, they are used in this financial report from Yahoo. So, instead of having one data point, you basically have a whole lot of information coming to you at once, so you can put one data point into context. Apparently, there is now a TrueType font available, but I haven’t tried it. Doing a Google Image search for sparklines yields a lot of interesting examples.
• One of the most beautiful examples of design dates way back to Saturn images in Galileo’s text .
• The books! Edward Tufte’s books are beautiful.
• Never do powerpoint again. Do written reports that get handed out ahead of time. Tufte has an interesting article on Lousy PowerPoint presentations. Also, he has done work on conveying technical information in powerpoints, and how NASA had relied heavily on powerpoints during the Challenger days. This is fascinating, and sad days in scientific communication.
• Tufte has an amazing book collection. At one point he pulled out a personal copy of Euclid’s Geometry (1570) He noted that his copy was previously owned by Shakespeare’s contemporaneous playwright, Ben Jonson. This is the oldest pop-up book I have ever seen: three flaps of triangular paper come together in a 3-D point, to form a tiny pyramid. Tufte reprints an identical pop-up over 400 years later in Envisioning Information, on page 16. 

It has been a beautiful data kind of day!  I just ran across a story in Computerworld that has a great listing of 22 free tools for data visualization. And, Lauren Meyers, in our archive, just sent me this cool link to Victorian Infographics!

So, we are brainstorming around the library about how to make beautiful data, and mining the archives for some lovely examples of 3-D design in Euclid books, and old maps. We’ll let you know what we find!

Additonal note 5/5/2011: I had forgotten to mention anything about the Galileo sunspot animation. Tufte showed a film that one of his students had produced. In a similar effort, Rice’s Galileo Project had done short animations of the daily sunspot observations that Galileo had done in 1612. Really brilliant.

Beautiful data….

I just got a copy of Beautiful Evidence by Edward Tufte. This truly does make information a beautiful thing to behold.

You might have seen Beautiful Data (link only for Rice) already; it came out in the summer of 2009,by Toby Segaran and Jeff Hammerbacker.

I love this new push to wonderful visual design in presenting data, even if it does come with some challenges. The DiRT wiki has a whole slew of data visualization tools to play with; so, try a Wordle or a chart done with Many Eyes, and see how much beauty can add to your information.

Pretty neat, huh!

We are all thinking about ways to present our information in a way that is graphic, and catches the eye of fatigued users. I just stumbled on this BBC video of a very enthusiastic Dr. Hans Rosling explaining the past 200 years of world progress in only four minutes using a fascinating graph. It is  a clip from a one-hour long documentary produced by Wingspan Productions and broadcast by BBC, 2010.  Dr. Rosling says at the end that it involved the plotting of 120,000 numbers. I love how he becomes part of the graph, part of the movement, part of the dance, if you will, of dreaming and thinking about the past, and the possible future. Enjoy.

Data Sites I love

I just returned from the 31st meeting of IATUL. IATUL provides a forum for the exchange of ideas relevant to librarianship in technological universities throughout the world. Data was on everyone’s mind! It was amazing to see the work being done on DataCite, out of Germany, and also, ANDS, an Australian effort. These types of projects make datasets extremely visible. There is obviously still a lot of work to be done by subject bibliographers and collection development specialists to acquire datasets, and also, to work with faculty to make sure that datasets get deposited into institutional repositories. But I have a lot of faith that with deep collaboration, large amounts of data will become publicly accessible over the next few years. The recent NIH and NSF mandates make this work crucial. The Library wants to be at the center of a strategic data management plan! Hard work between Datacite and publishers, such as Elsevier, make sure that the user has easy linking from an article, to the publicly accessible data behind the research. Even if they don’t have access to the article, they will have access to the abstract, and to the data that supports the article. A recent research paper in NaturePrecedings has shown that 48% of open data received 85% of total citations, “48% of trials with publicly available microarray data received 85% of the aggregate citations.”

Libraries can incorporate this work into their local catalogs, and make data easily findable.

We will see lots of efforts around data curation in the next few years.  We are in a whole new dataland!