MLA CE: The Diversity of Data Management: Practical Approaches Recorded Webcast, Nov. 11, 2016, Marci Brandenburg

What is Visualization?

“The action or fact of visualizing; the power or process of forming a mental picture or vision of something not actually present to the sight; a

picture thus formed.”

-Oxford English Dictionary

"visualization, n." OED Online. Oxford University Press, June 2016. Web. 27 June 2016.

Outline Overview of Visualization Library Services Instruction/Consultation Types of Visualization Design Principles

Referrals Finding and Manipulating Data

Challenges Acknowledgments

OVERVIEW OF VISUALIZATION

Anscombe’s Quartet

https://en.wikipedia.org/wiki/Anscombe%27s_quartet

https://en.wikipedia.org/wiki/Anscombe%27s_quartet

Anscombe’s Quartet

Danger, Will Robinson

Vigen T. Spurious Correlations. Cambridge, Massachusetts. 2015. Web. 10 March 2014. <http://www.tylervigen.com/view_correlation?id=1703>

http://www.albany.edu/museum/wwwmuseum/work/ lombardi/images/lombardi1.jpg

“Mark Lombardi (1951-2000) draws on the major political and financial scandals of the day to create large-scale linear diagrams that at first glance look like celestial maps….” -University Art Museum. University of Albany, State University of New York.

http://www.albany.edu/museum/wwwmuseum/work/lombardi/images/lombardi1.jpg



Viégas F., Wattenberg, M. Hint.fm. Cambridge, Massachusetts. 2014. Web. 10 March 2014. <http://hint.fm/wind>

Data Visualization Data visualization deals with communicating information about an existing data set through a visual medium. Goals

• Take advantage of the brain’s ability to efficiently process visuals. A high volume of information can often be easily understood through an image.

• Want the viewer to learn something about the data that could not be easily understood by reading the raw data.

• Visual is presented in a simple manner. Don’t overload the viewer with many unnecessary details.

LIBRARY SERVICES

Instruction • Specific visualization

software tools (ex. Cytoscape, R)

• Basic principles (when appropriate, good vs bad, etc)

• Curriculum • Data management

practices

Consultation • Specific visualization

software tools with researcher data (ex. Cytoscape)

• Range of visualization tools to meet a specific purpose or type of data

• Charts/graphs for manuscript (pie chart vs bar chart vs network graph)

• Opportunity to stress good data management

TYPES OF VISUALIZATION Instruction/Consultation

Graphs Goal is to communicate the relationships among the data efficiently.

Common Types of Graphs:

• Simple Graphs • Bar graphs, scatter

plot, pie chart • Network Graphs

• Shows relationships between points

Cytoscape Consortium. 2016. Web. 30 June 2016. <http://cytoscape.org/what_is_cytoscape.html>

http://www.albany.edu/museum/wwwmuseum/work/ lombardi/images/lombardi1.jpg

“Mark Lombardi (1951-2000) draws on the major political and financial scandals of the day to create large-scale linear diagrams that at first glance look like celestial maps….” -University Art Museum. University of Albany, State University of New York.




The American Southwest. 2016. Web. 31 October 2016. <http://www.americansouthwest.net/california/yosemite/glen-aulin-trail-map.html>

Map-Based

Snow, J. Snow on Cholera being a Reprint of Two Papers. London: Oxford University Press, 1936. Print.

Map based example - Cholera Outbreak Created by John Snow

3D can untangle a graph: entire new dimension to place nodes Need to think about if our visualization would really benefit from 3D.

• Does it overcomplicate the visualization? • Does the 3D graph provide more information to the viewer, or does it just look

different?

2D vs. 3D Visualization

Just because we can…

Andy Kirk. Data visualization: a successful design process. Birmingham: United Kingdom, 2012. eBook.

DESIGN PRINCIPLES Instruction/Consultation

Visual Weight Guiding the Eye

Eric Maslowski. University of Michigan 3D Lab. http://um3d.dc.umich.edu. 2008.

Size and Shape


Value and Hue Color


Contrast


Visual Assumptions


Visual Assumptions


Visual Assumptions


Referrals • Visualization landscape at

institution/organization • Locations/units that offer visualization services • Know people and technology

• U-M Library Data Visualization webpage (http://www.lib.umich.edu/data-visualization) • Group email address

• Educate all librarians to be able to make better referrals

http://www.lib.umich.edu/data-visualization

Finding and Manipulating Data

• Subscriptions to various data sets • Help researchers find publicly available

data • Manipulate and clean data to work in

desired visualization software • Deep Blue Data • ICPSR

CHALLENGES

Challenges

• Lot of visualization tools and resources • Constantly changing • Some are very complex

• Wide range of visualization types • Focus on principles; rely on others

• Making the library the go-to place for visualization services

• Resources

Acknowledgments

• Eric Maslowski - Director, Creative Applications. Co-Director, Digital Media Commons. University of Michigan

• Justin Joque - Visualization Librarian, University of Michigan

• Jean Song – Assistant Director Academic and Clinical Engagement, Taubman Health Sciences Library, University of Michigan

Marci Brandenburg [email protected]