Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

Make your likes visible on Facebook?

Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.

No, thanks

Las Vegas from a Distance: Visualizing CCCC 2013

Using various modes of visualization--word clouds, maps, charts, graphs, and tables--to represent the program of the 2013 Conference on College Composition and Communication.
by

Eric Detweiler

on 4 June 2013

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Las Vegas from a Distance: Visualizing CCCC 2013

Las Vegas from a Distance:
Visualizing CCCC 2013 Eric Detweiler
Rhetoric PhD Student @ The University of Texas at Austin
eric.detweiler@utexas.edu
@EricSDet
RhetEric.org That’s not to say these scholarly concerns subsumed or superseded my more idiosyncratic interests. If anything, the two ended up informing each other. As a rhetorician, my interest in the emotional appeal (or lack thereof) of various means of data visualization resonated with my own fondnesses and curiosities. Somewhere between literal quantitative representation and the aleatory potential Victor Vitanza ascribes to the Internet Anagram Server, the odd juxtapositions of terms in word clouds—themselves here juxtaposed with graphs, charts, and geographical maps—opened onto a variety of scholarly, rhetorical, aleatory, and heuristic affordances. This interest acquired a more scholarly bent while I was working on a project exploring quantitative differences between the citational practices of scholarship on the “rhetoric” and “composition” ends of the “rhetoric and composition” spectrum. Initially, I represented my accumulated data using relatively traditional forms—tables, primarily. As the project grew, however, I grew curious about alternatives. I ended up encountering, among others, Derek Mueller’s work on word clouds and the “long tail” of rhet/comp scholarship, as well as Louise Wetherbee Phelps and John M. Ackerman’s disciplinary maps and models. Such methods of visualization can be especially revelatory with large data sets that, when taken linearly, seem to sprawl interminably. And as most teachers and scholars who have attended the Conference on College Composition and Communication (CCCC) know, the CCCC program can comprise just such a data set. Choosing which panel to attend during any given conference session, much less sussing out trends and movements in the field(s) the conference represents, can be an aporetic experience. Which brings me to the twofold aim of this project: First, to perform a sort of exploratory postmortem on the CCCC 2013 program by representing its hundreds of pages and panels in the form of clouds, graphs, charts, and maps. Second, to think about the rhetorical effects and affects that might be opened up or foreclosed by these various forms of representation and visualization. I chose Prezi for this project because of its emphasis on spatial relation rather than linear progression. I’ve provided a string of signposts to move users through the project, but I’d encourage users to use Prezi’s zooming and click-and-drag features to (re)visit the project in a less linear fashion, jumping between various areas in search of surprising linkages. In such a spirit, I have refrained from providing much analysis of the visuals. Instead, I stick primarily to brief methodological and/or explanatory notes, leaving the analysis to the user’s curiosity. Just one initial question as a heuristic/example: What might it mean that, though both the conference’s name and its call for papers prominently featured the word “composition,” “writing” is a much more significant term in the titles of the conference’s panels and presentations? I began with a spreadsheet containing the titles of both every CCCC 2013 panel and the individual presentations that made up those panels. I excluded workshops, special interest group meetings, etc., in order to focus on the rhetorical patterns at work in the panels and presentations that comprise the bulk of the conference. These titles represent at least two procedural levels: first, the titling decisions of individual presenters and those presenters who formed and titled their own panels; second, the decisions of the reviewers who accepted those panels/presentations and the conference organizers who titled ad hoc panels. Both submitters and selectors, of course, were constrained by larger scholarly and disciplinary norms, as well as the guidelines of the 2013 conference’s call for papers. Once the panel and presentation columns of the spreadsheet were complete, I generated two word clouds: one based on all the conference’s panel titles and one based on the presentation titles. I did so out of curiosity about the differences and similarities between the two, assuming that the influence of conference organizers would be more pronounced at the panel level.

[Note: All subsequent clouds exclude common English words like prepositions, articles, and common conjunctions. I adjusted the settings of Wordle’s cloud generator to ignore case sensitivity and to include all words in the data set being represented—though as scrupulous users might note, the less frequent words remain essentially illegible due to file-size constraints.] Panel Cloud Presentation Cloud The titular theme of the conference, which was highlighted in its call for papers, was “The Public Work of Composition.” The overlap between that theme’s key terms and the titles of panels and presentations is clear in the clouds above. This isn’t pure causation, of course: one would expect the word “composition” to show up in panel titles at a “composition and communication” conference even if the word wasn’t featured in the CFP’s title. Nevertheless, I was curious to see what the clouds would look like without the dominance of the CFP’s terms and so generated clouds without three key words: “public,” “work,” and “composition.” Panel Cloud Presentation Cloud No Keywords No Keywords The conference and the clouds could barely contain "writing," it would seem. Each panel at the conference was classified under 1 of 14 different “clusters”: [1] “Academic Writing”; [2] “Basic Writing” (a new addition to the 2013 conference); [3] “Community, Civic, & Public”; [4] “Creative Writing”; [5] “History”; [6] “Information Technologies”; [7] “Institutional and Professional”; [8] “Interdisciplinary, Multidisciplinary, and Cross-Contextual Perspectives”; [9] “Language”; [10] “Professional and Technical Writing"; [11] “Research”; [12] “Teaching Writing & Rhetoric”; [13] “Theory”; and [14] “Writing Programs.” Though a list of all panels indexed by cluster is provided in the early pages of the program, I was interested how the relative prevalence of various clusters might come through in a different form. I added a "cluster" column to my spreadsheet that included the 14 primary clusters as well as a separate designation for "Featured" sessions/speakers and the few panels that had no cluster designation. Featured sessions also labeled with a cluster type were included in the relevant cluster's count (e.g. a "Basic Writing Featured Session" was counted under "Basic Writing"). Based on that column, I generated a word cloud representing each clusters' relative prominence.

[Note: As becomes apparent in this cloud, punctuation emerged as an interesting challenge for this project. Since Wordle isolates each word, I had to find workarounds when I wanted clusters of words to stick together. This resulted in a heavy reliance on hyphens and slashes, substituting "and" for "&" (Wordle counts the latter as a space), and—perhaps most prominently—taking a cue from early Latin writers and eschewing spaces altogether.] Cluster Cloud In an attempt to pinpoint the specific points of overlap and difference between the various clusters—keeping in mind that both conference organizers and those who submitted to the conference were forced to contend with the relative generality and necessary ambiguity of such denominations—I created separate word clouds for each cluster, one at the level of panel titles and one at the level of presentation titles. These 30 clouds are what follow. In order to draw some attention to the rhetorical effects of stylistic details, I varied the fonts and the orientation settings for each clusters' pair of clouds. In order to maintain some ease of comparison, however, all clouds include the CFP keywords and were generated using Wordle's alphabetizing feature (hence the tendency of "writing" to be a sort of tail protruding from the clouds' right side). The default Prezi windows for the cloud pairs tend to be large, so do feel free to zoom in and around. Academic Writing Panel Title Clouds Presentation Title Clouds Basic Writing Community, Civic, & Public In contrast, a tabular representation
of the same data might look like this: Creative Writing History Information Technologies Interdisciplinary, Multidisciplinary, and Cross-Contextual Perspectives Institutional and Professional Language Professional and Technical Writing Research Teaching Writing & Rhetoric Theory Writing Programs Featured After examining the presentations across clusters, I became curious about the origins of the presenters themselves. Where, institutionally and geographically speaking, were these presenters from? Before starting to mix in other modes of visualization, then, I added two more spreadsheet columns and generated two additional clouds. The first represents the institutions—by and large colleges and universities, though occasionally companies, publications, etc.—with which presenters were affiliated in the program. When no affiliation was included in the program, I categorized the presentation under "NotListed." As with the cluster clouds, the next two clouds were generated with Wordle's alphabetization preference turned on.

[For the sake of visual compactness, "University" is abbreviated as "U." For university systems with multiple campuses represented in the conference program, I identified specific institutions with a hyphenated location or other identifier. E.g. the cloud contains "UofTexas-PanAmerican" and "UofTexas-Austin." Institutions in the State University of New York or City of New York system are identified with the respective suffixes "-SUNY" or "-CUNY," and I used the colloquial abbreviation "Penn" for the state university system of Pennsylvania.] I based the second cloud in this pair on geographical markers. For institutions located in the United States, I used their home state's postal abbreviations. I also subdivided Canadian institutions by province, spelling those provinces out for unfamiliar users [e.g. "Alberta-Canada"]. Because of a lack of intranational diversity in the geographical locations of other represented countries, I identified them with the home country of the presenters' institution. It seemed to me that an important part of exploring any visual form's potential was pushing its limits while keeping an eye out for its limitations. In the case of word clouds, the geography cloud illustrates a few. First, less represented states and countries are virtually (or literally, in the case of those not represented at all) invisible, making it relatively difficult to comment on or critique any under-representation/absence. And though the scrambled juxtapositions of the cloud might draw attention to certain points of comparison and contrast (the relatively similar prevalence of New York and Texas, for instance), it might cause the user to miss certain bits of data that are readily recognizable in more standard visual forms.

With this in mind, I used two other forms to represent presenters' geographical distribution. The first is a geographical map of the United States. Each state is labeled with the number of times it was represented by presenters and their affiliated institutions as listed in the CCCC 2013 program. Additionally, the states are shaded in accordance with the key at the lower-right corner of the map. In order to clearly delineate the countries represented at the conference—especially since less prominent countries were all but absent from the cloud and, of course, from the single-country map above—I used a traditional table. Interested in additional ways of exploring what sorts of institutions predominated at CCCC, I turned to Carnegie Classifications, the self-described "leading framework for recognizing and describing institutional diversity in U.S. higher education" ("Carnegie Classifications"). Among many other variables, these classifications chart [1] whether institutions offer 4- and/or 2-year degrees; [2] whether institutions are public, private not-for-profit, or private for-profit; [3] the highest degree an institution offers (Associate's, Bachelor's, Master's, or doctorate); and [4] a "Basic Classification" based on a variety of factors explained here: http://classifications.carnegiefoundation.org/methodology/basic.php

Especially given CCCC 2013's particular focus on writing instruction at community colleges and 2-year institutions, the representation of institutions across these various classificatory measures seemed a relevant approach to my data set. I added spreadsheet columns for each of the four measures mentioned in the previous paragraph and quickly discovered I was near the edge of word clouds' utility. Consider the one I got for measure [1] above: whether universities offer 4- or 2-year degrees. As a form of visualization, this seems best classified as "mostly harmless." Though who knows? So though I continued to experiment with word clouds for my final work with Carnegie Classifications, I also generated tables, pie charts, and/or bar graphs as alternative means of representation.

[Because Carnegie Classifications only consider U.S. institutions, all subsequent visuals were based solely on U.S. institutions represented by the conference's presenters. As the classifications were last updated in 2010, a few U.S. institutions that have been founded, merged, or re-branded since that update are also excluded from the data set used to generate the following.] 4- and 2-Year Institutions Public and Private Institutions Degree Type Cloud Pie Chart Basic Classification Cloud Table Continuing in an exploratory and inventional mode, I'll end by beginning to sketch out a few questions.

[1] At the level of form–the level of how these various visuals represent data rather than what data they're representing–what connections, links, and inventional possibilities might these visuals open up? What do the newer forms of representation made possible by technological aggregators and algorithms reveal about both the limitations and the persistent, preferable utility of more traditional modes of data representation? How might these forms change the ways one engages with conferences, disciplinary structures, and linear texts like conference programs? [2] At the level of content–the data itself–what might we glean about CCCC, as well as the disciplines, methodologies, institutions, and structures with which it's connected and bound up, based on what's presented here? [3] Finally, at the level of pedagogy–a key focal point of CCCC–what might it mean to engage students of writing as rhetorical analyzers, gatherers, and artisans of these sorts of graphical, textual, experimental, and traditional (I am, after all, ending with linear text) forms? Though I have no readymade answers for these questions, I am of course happy to discuss them–feel free to contact me using the information nested under the title below with any questions, concerns, comments, etc.

Thanks to The University of Texas at Austin's Digital Writing and Research Lab for providing access to the software and hardware used to create this project. All visuals were created using an amalgamation of screen grabs, Adobe Photoshop CS5, Wordle, Microsoft PowerPoint, Microsoft Excel, and Prezi.

Additional thanks to Pearson for supporting this project via a 2013 Emerging Pedagogies Research & Travel Grant. You can find out more about the grant, as well a gallery of past recipients' projects, here:

http://www.pearsonhighered.com/cccc_emergingpedagogies/

And you can check out my 2012 grant project here:

http://rheteric.org/Four_Movements.html "Carnegie Classifications." The Carnegie Foundation, n.d. Web. 3. Jun. 2013. http://classifications.carnegiefoundation.org/

Detweiler, Eric. “‘-,’ ‘And,’ ‘/’? An Empirical Examination of the Disciplinary Relationship Between ‘Rhetoric’ and ‘Composition.’” Conference on College Composition and Communication. Las Vegas, NV. 15 Mar. 2013.

Mueller, Derek. “Clouds, Graphs, and Maps: Distant Reading and Disciplinary Imagination.” Diss. Syracuse U, 2009. ProQuest Dissertations & Theses Full Text. Web. 3 Jun. 2013.

Mueller, Derek. “Grasping Rhetoric and Composition by Its Long Tail: What Graphs Can Tell Us about the Field’s Changing Shape.” College Composition and Communication 64.1 (2012): 195-223. Web.

“2013 CCCC Conference Program.” Conference on College Composition and Communication. National Council of the Teachers of English, 2013. Web. 3 Jun. 2013. http://www.ncte.org/cccc/review/2013program

Phelps, Louise Wetherbee, and John M. Ackerman. “Making the Case for Disciplinarity in Rhetoric, Composition, and Writing Studies: The Visibility Project.” College Composition and Communication 62.1 (2010): 180-215. Web.

Vitanza, Victor. “From Heuristic to Aleatory Procedures; or, Toward ‘Writing the Accident.’” Inventing a Discipline. Ed. Maureen Daly Goggin. Urbana, IL: NCTE, 2000. 185-206. Print. Munroe, Randall. “Click and Drag.” Webcomic. Xkcd.com, 2012. Web. 3 Jun. 2013. http://xkcd.com/1110/

Ridolfo, Jim. Rhet Map: Mapping Rhetoric and Composition. Rhet Map, 2013. Web. 3 Jun. 2013. http://www.rhetmap.org/ Works Cited Other Works Consulted My interest in methods of data visualization—specifically word clouds—began as a personal tic. When such word-cloud generators as Wordle (http://www.wordle.net) started popping up a few years ago, I found something immensely compelling about feeding them my various writings. The generated clouds were made up of each word that appeared at least once in the piece of writing fed into the generator—the more times a word appeared in the piece, the larger it appeared in the cloud. Perusing these clouds provided me a way of simultaneously stepping back from and engaging more closely with my writings. Something about the clouds spoke to me in a manner linear sentences and tabulations of word frequency did not.
Full transcript