December 08, 2013

Remembering Mandela

Coming of age as I did in the 1980s, Nelson Mandela loomed large and it was with a profound sadness that I learned of his death.  He was a revolutionary fighting against both the unjust racial system of apartheid as well as larger political and economic structures that enabled (and continue to enable) the some to exploit the many.

With his death, the world has lessened; but the dream and the work for justice remains, both in South Africa and the world.

I take comfort in this visualization of the globalness of his legacy, presently partially and imperfectly, in the animation below. Individuals from around the world sending a message containing his name, the merest flash in time-space, that when combined with others sparks, can set the world aglow.

Nkosi Sikelel' iAfrika….

Tweets from December 5th and 6th 2013 containing "Mandela" 

-------- technical notes -----
- This animation represents 190,000 geocoded tweets mentioning 'mandela' sent from 14:00 December 5th to 16:00 on December 6th. Geocoded tweets are only a fraction of all tweets and so the actually number of tweets sent is likely up to 100 times more.
- The use of Twitter varies across countries and low levels of use means that some countries with large populations do no stand out in this map (e.g., China and India). This is most likely an artifact of differences in technology use as well as non-Roman alphabets rather than an indicator of interest.

November 14, 2013

two new researcher positions at the Oxford Internet Institute

We're happy to announce two new researcher posts that have been made available at the Oxford Internet Institute. Both positions link into Mark's larger ERC and IDRC funded research into knowledge economies and virtual labour in Sub-Saharan Africa.

One post is for a researcher with quantitative and statistical skills. 

The other is targeted towards a researcher that has experience using qualitative methods. 

Mark will be working closely with the two successful candidates, and is looking forward to the exciting research possibilities in both projects. Please feel free to get in touch with any questions.


Grade 7: Salary £29,541 - £36,298 p.a.

The research focuses on how new economic practices and processes are taking root in Sub-Saharan Africa as a result of changing connectivities. We plan to map formal and informal types of participation in ‘knowledge economies’ in order to investigate why certain places have sustained their dominance, why others have become more central, and why some places, practices, and initiatives have declined.

To do this we are seeking a researcher with experience in quantitative social research. The researcher will work on three stages of the project. First, collecting and bringing together all necessary data. While some of the data are readily available in existing and open datasets, others require the creation of custom scripts and data collection tools. Second, using GIS and statistical packages to comprehensively analyse the data. We plan to employ both inferential models and descriptive graphics and maps. Finally, broadly disseminating this work in a variety of open and accessible formats including a data-sharing tool, an interactive website, open reports, and peer-reviewed academic journal articles. The work will also be used as a base for detailed qualitative research performed by two other members of the research team.

The successful applicant will demonstrate an ability to carry out social and spatial statistical analysis, visualise results, write for both public and academic audiences, and work with an interdisciplinary team. We also welcome applications from candidates who are additionally eager to design a future research programme in order to extend the position.

Based at the Oxford Internet Institute, this position is available from 1st March 2014 for 36 months in the first instance, with the possibility of renewal thereafter funding permitting.

Only applications received before 12:00 midday on 9th January, 2014 can be considered. Interviews for those short-listed are currently planned to take place in the week commencing 27thJanuary 2014.

To apply for this role and for further details, including a job description, please click on the link below:


Grade 7: Salary £29,541 - £36,298 p.a.

The Oxford Internet Institute is a leading centre for research into individual, collective and institutional behaviour on the Internet. We are looking for a full-time Researcher to work with Dr Mark Graham and Dr Vili Lehdonvirta on the IDRC-funded project Microwork and Virtual Production Networks in Sub-Saharan Africa and Southeast Asia. Combining archival research, surveys, and interviews, this ambitious project will critically assess the impact of Internet and mobile connectivities on social and economic development, particularly insofar as they open up opportunities for novel forms of online work, such as ‘e-lancing’, ‘microwork’, and ‘game labour’.

In this exciting role, the Researcher will carry out a total of approximately six months of fieldwork among virtual workers and organizations in Sub-Saharan Africa and Southeast Asia, as well as working at OII’s premises in Oxford. The Researcher will also contribute to the dissemination of the findings through peer-reviewed academic papers, project reports, events, blogs and social media.

Candidates should have experience of social science research in Development Studies, Geography, Sociology, Social Anthropology, Communications, Organization Studies, Management or related disciplines, training and practical experience in qualitative research methods.

Based primarily at the Oxford Internet Institute (with periods of fieldwork), this position is available immediately for 2.5 years in the first instance, with the possibility of renewal thereafter, funding permitting. For qualified candidates, there may also be opportunities to teach course modules on our ‘Social Science of the Internet’ MSc course.

Only online applications received before 12:00 midday on 13 December 2013 can be considered. Interviews for those short-listed are planned to take place on 16 January 2014.

To apply for this role and for further details, including a job description, please click on the link below:

November 12, 2013

Crisis Mapping in the Philippines: Efforts and Resources

Typhoon Haiyan/Yolanda in the Philippines has resulted in catastrophic loss of life and deprivation and our hearts go out the people and towns affected. Response to this crisis (as in the case of Hurricane Katrina, the Haitian Earthquake and Hurricane Sandy) includes a range of efforts that leverage user-generated data and the volunteer mapping efforts of individuals and organizations. We thought it useful to highlight some of these unfolding efforts here as resource for our readers and ourselves. At present this is just a listing of some of the things we've seen in our feeds, please add additional ones to the comments.

MANY of these Maps are actively seeking volunteers so please join in. 

A Variety of Crisis Maps for the Philippines:
News Articles about Current Crisis Mapping Responses:
Previous academic work on crowd-sourced crisis mapping responses:

November 06, 2013

The Geography of Top Level Domain Names

Some of the sheep team have just published a new map over at the Information Geographies project. This one draws on some of Matt's long term research into the geography of domain names which dates back more than fifteen years.[1]   Egads, where did the time go?.

The map offers a detailed overview of one-facet of the geography of content production.  While not the normal kind of user generated data we use in our work at FloatingSheep, domain name registrations are an indicator of content production.

Some results are unsurprising (for instance the low scores in many countries that have low numbers of internet users). However, other unexpected patterns also reveal themselves (such as the relatively low numbers of domains in many Asian countries).  For a more detailed description of results take a look at the discussion on the map's Internet Geography home.

Also, to give some perspective on how much things have changed, here is map of .com domains in San Francisco (and zoomed into just the South of Market region) back in 1998 when there were less than 2 million rather than 110 million that exist today.

Distribution of .com domains by Registrant Address, San Francisco, Summer 1998
Source: Matthew Zook, 2005 (see below)

Distribution of .com domains by Registrant Address, 
South of Market, San Francisco, Summer 1998
(apologies for the low quality image, it is the only one available)
Source: Matthew Zook, (see below)
[1] If you are interested in reading more domain name work check out.

Zook, M.A. (2001). Old hierarchies or new networks of centrality? The global geography of the internet content marketAmerican Behavioral Scientist. (June). Vol 44. No. 10. 1679-1696.
Zook, M.A. (2000). The web of production: The economic geography of commercial internet content production in the United StatesEnvironment and Planning A. Vol. 32. 411-426.  

October 31, 2013

One more CFP for AAG 2013

And here is one more call for papers for the Association of American Geographers meeting in Tampa, FL.  

Alternative Computation and Unconventional Spaces

Newly emergent features of the computational turn (e.g., Berry 2011)
posit new challenges for geographers practicing computer-mediated
research.  While geographers maintain a strong relationship with
geographic information systems, new technologies, hardware, and
practices suggest exciting new avenues for computational research.
Geographic “big data” demand new, computationally intensive approaches
to geospatial analysis.  Textual artifacts from social media sources
augment traditional geospatial inquiry, but also serve as data for
non-GIS computational work such as natural language processing, topic
modeling, or social media analysis.  These provocative treatments
suggest ways in which information can be geographic, yet not
necessarily require explicit Cartesian expression. Results of such
analyses have determined uneven distributions of data, and limits to
the representational abilities of GIS.

We’re excited to push beyond traditional GIS techniques to explore
other ways in which our digital beings are expressed through space and
place.  This session welcomes both empirical and theoretical work that
advances computer-mediated research in novel ways.

We welcome papers on the following topics (or any closely related):

1)  Digital humanities-inspired inquiry for Geography

2)  Alternative methodologies for the digitally underrepresented

3)  Novel geospatial and other computer-mediated approaches to “big
data” analyses

4)  Non-Cartesian geographic information and its analyses

5)  Computer-mediated research located in underrepresented spaces
(rural areas, impoverished places, etc.)

6)  Geographic natural language processing, topic modeling, or other
textual analysis

7)  Relational spaces of Social Network Analysis

Please send related abstracts to Joe Eckert ( and
Monica Stephens (

October 01, 2013

CFPs for AAG 2014 Meeting in Tampa

Though the deadline for abstracts for the 2014 Meetings of the Association of American Geographers isn't until December this year, we thought we'd get a jump on things and get our sessions organized early. Below are CFPs for two Floatingsheep-organized sessions, one by Mark and Matt on data shadows and another by Taylor on smart cities.

1) CFP: Data Shadows and Urban Augmented Realities  

Most parts of our urban areas have become both digitally connected and represented by digitalized information. Digital layers of geographic information (commonly referred to as "augmented reality" by computer scientists) can take myriad forms. The most visible of which are probably the digital maps that many people use to navigate through cities. Google, Yahoo!, Wikipedia, Apple, OpenStreetMap, Baidu, and many other companies and organisations all host publicly accessible platforms that partially reflect parts of our world. These services also become the platform for an almost unimaginable amount of additional content that both reflects the materiality of cities and augments it with additional content. This additional volunteered (and emitted) geographic information is comprised of photographs, blogs, tweets, social media checkins, webcams, videos, and encyclopedia articles. These layers of digital representations are then further reproduced and repurposed in the ways that they annotate the urban environment

The ambition of this session is to interrogate the increasing prevalence of both geographically referenced digital information and the code through which it is regulated. By asking what these augmented realities are, where they are and where they are not, and how they are brought into being, we can both unpack the language we use to speak about digital augmentations and explore the ways in which digital extensions of place are becoming increasingly important in everyday, lived geographies.

This session seeks two kinds of papers. First it aims to provide space for papers that explore the ways in which we should imagine, describe, critique, and even name, the digital and informational augmentations of our lives. Second, the session seeks papers that critically examine information geographies and augmented realities in specific contexts. How do informational augmentations impact on how we bring our worlds into being? What and where do they exclude? What narratives and discourses do they allow, and what do they conceal? How are they governed, regulated, and challenged?

Please submit abstracts of less than 250 words to MarkGraham ( and Matthew Zook ( before October 31, 2013.  We will review abstracts in order to form cohesive sessions.

2) CFP: Thinking the ‘smart city’: power, politics and networked urbanism
The fact that cities are increasingly being augmented by digital hardware and software, producing massive amounts of data about urban processes, has been well documented in recent years. Discourses around so-called ‘smart cities’ and tend to position them as either a panacea, an entirely new conceptual and material breakthrough, or as a kind of dystopian imposition of technological rationality onto cities, leaving the precise nature of this social and spatial reorganization unclear. This session will engage these issues through empirically-focused, but conceptually-rich, research on how digital information and communication technologies do not simply connect cities to distanciated networks, but also drive new forms of urban development and new methods of civic exchange and political contention between municipalities and their residents. 

This session seeks papers that document and analyze how these new socio-technical systems are reconfiguring the relationships of urban governance, and how these systems remain embedded in longstanding social structures at both local and global scales. We are also interested in how geographers might offer a unique perspective on the processes and outcomes of smart urbanism, especially given the dominance of computer scientists and management consultants in the making of these projects. Possible topics might include, but are not limited to:

-- Policy mobilities and the ‘smart city’ model
-- Politics of urban data
-- Smart cities and technocratic planning
-- Smart cities as new urban entrepreneurial assemblages
-- Virtual spaces in the networked city
-- Role of transnational corporations in promoting smart city developments
-- Smart cities and urban environmental sustainability
-- Smart cities in the Global South
-- Cybernetics and the intellectual history of smart urbanism

Please submit abstracts of no more than 250 words to Alan Wiig ( and Taylor Shelton ( by October 15th to ensure sufficient time for review.

September 18, 2013

What do Twerking and Syria have in common? Not much, except for the Twerking Tramp Stamp of America

User-generated data, especially generated via social media, provides a useful look into the day-to-day experiences and conversations that occur around the world. This kind of data provides insights, however partial, into our hopes, our triumphs and our fears. Sometimes it reveals things that we would rather not think about, like hate speech or racially charged discourse. But in all cases, it enlightens us to what matters to people, or at least what matters to the people participating in online discussions.

The past month has seen a sharp increase in two topics of discussion, the first representing an African American cultural meme from the bounce music tradition of New Orleans but more recently (and cynically) appropriated by white pop artists, and the other tied to an ongoing conflict that rapidly garnered calls for international intervention. We speak, of course, of twerking [1] and the ongoing civil war in Syria. While the two topics share little in common apart from recent media attention, the divergences between the space-time patterns of these geo-coded tweets show how online and offline actions and characteristics are intricately and imperfectly connected.

Using DOLLY, we extracted all geocoded tweets from July 1, 2012 to September 11, 2013 from North America that referenced either "twerk*" or "syria*". It quickly became apparent that twerking is a much more popular topic on Twitter, with 775,000 geocoded tweets during this time period, while there were only 75,000 references to Syria [3]. These numbers have changed significantly over the past month as the U.S. has called for military strikes in response to reports of chemical weapon attacks in Syria, but nevertheless there have still been three times as many references to twerking as Syria in August and September, with 133,000 tweets against just 43,000.

Indexed Volumes of Twerk and Syria Tweets, July 2012 to August 2013
The evolution in volume of each kind of tweet overtime is also strikingly different.  Compared to the number of mentions in July 2012, twerking has steadily become a more popular topic of Twitter conversations over the course of the past thirteen months. In contrast, discussion about Syria largely declined over the course of the year, and only in August did it became a hot topic. Since July 2013, the relative amount of conversation about Syria increased almost tenfold, from an index value of 61 to an index value of 526. Though they took much different trajectories, both topics have around five times as much discussion as was the case a year ago.

Looking at the geography of these tweets, just for the month of August 2013, illustrates how tweeting behavior varies across space during a time when there was a lot of national attention to both topics. Using a simple ratio of the # of Syria Tweets / # of Twerking Tweets, which we term the Twerkyria Index, the maps below show this distribution at the state and county levels.

The Twerkyria Index by State, August 2013

One of the most compelling results is the clear difference in ratios for Washington D.C., which has three times as many Syria tweets as twerking tweets, bucking the national average which is three to one in the opposite direction. The next closest areas are Vermont and Alaska, with relatively small African American populations, which have roughly the same number of tweets for each topic. The rest of the country is divided into red states that have more than the national average of twerking and pink states that are slightly less twerking obsessed (at least relative to attention to events in Syria). The concentration of states in the southeast -- from Texas to South Carolina -- forms a Twerking Tramp Stamp across America [4], with a few other concentrations tastefully tattooed across the Great Plains and Midwest.

The pattern suggests a number of possible connections between the Twerkyria index of Twitter activity and offline demographics. Despite Miley Cyrus's "act of cultural appropriation being passed off as a rebellious reclamation of her sexuality after a childhood in the Disneyfied spotlight", twerking's roots are within
African American culture, especially as it relates to southern hip-hop. In this context, sending a tweet containing twerk is likely much more about cultural expression of local and identity politics than the more recent appropriation of the work by the dominant culture to work through "a raft of personal, socioeconomic and third-wave-feminist issues". Likewise, tweets containing Syria represent a wide range of political views and stances towards possible U.S. intervention.

In short, it's complicated. Far from a simple unitary meaning, the use of both twerking and Syria on Twitter are complicated expressions of cultural and political expression in the U.S., relating these locations both to particular regional cultures within the country and particular geopolitical configurations that span the globe.

In an effort to test some of these relationships, we ran a quick (and relatively crude) OLS regression at the state level with the Twerkyria index as the dependent variable and a range of demographic variables including:
  • Population under 18 years, percent, 2012  
  • African American Population, percent, 2012 
  • Median value of owner-occupied housing , 2007-2011
  • Population per square mile, 2010 
The model excludes Washington DC, given that it is a city rather than a state [5] and includes dummy variables for Hawaii and Alaska given their unique spatial position vis-a-vis the lower 48 states.

Modelling the Twerkyria Index

While there are any number of ways to critique/improve upon this basic model, it works well for simple illustration [6]. The model explains about 65% of the variation in the Twerkyria Index. States with younger populations and a higher percentage of African Americans are associated with more twerking tweets. Interest in foreign affairs is more difficult to measure (at least with standard Census data), but a combination of housing price and population density provides a measure of a state's urbanity and presumed interest in international affairs. Locations with more expensive real estate and higher population densities, i.e., more urbanized states, have relatively fewer tweets about twerking and more with Syria. This model shows that the Twerkyria Index correlates fairly well with some reasonable theoretical expectations about the nature of the offline demographics of the points of origin of these tweets.

The Twerkyria Index by County, August 2013

The state level, however, masks many of the subtle distinctions that emerge within these relatively large spatial units.  Examining the Twerkyria index at the county level (see below) shows that the higher number of Syria tweets in Washington DC is also evident in a number of counties, roughly equal in size, to the district. While these counties are scattered across the country, a particularly large concentration is found in the San Francisco Bay with Alameda County, containing Oakland and Berkeley, representing the source of fully 8% of all Syria-related tweets. Given this large concentration, it is not surprising, that the Twerkyria Index diverges greatly from the rest of the country. In contrast, southern California largely conforms with the larger national trend of tweeting considerably more about twerking.

In summary, we have no easy answer to the age old, "To twerk, or not to twerk" [7], but looking at the socio-spatial dimension of online activities provides useful insight on the complicated interconnections between our online and offline activities.

[1] For those few unaware of what twerking is, we refer you to the Wikipedia definition, which defines it as "a type of dancing in which the dancer, usually a woman, shakes her hips in an up-and-down bouncing motion, causing the dancer's buttocks to shake, "wobble" and "jiggle".  If you still have trouble understanding it we suggest this thoughtful overview provided by the New York Times. After all, where else would one go to truly understand an artifact of twenty-century African American urban culture than the grand grey lady of journalism?  Or you could view the three videos at YouTube with the most hits 1) Booty Me Down Song By Kstylis; 2) How to Twerk and 3) Jimmy Kimmel Reveals "Worst Twerk Fail EVER - Girl Catches Fire" Prank".

We're still working on obtaining video of someone from the FloatingSheep collective twerking but an array of technical and legal difficulties have prevented this thus far. [2]

[2] Also the fact that no one has volunteered has been a bit of problem. But we have high hopes that we'll eventually wear Mark's resistance down.

[3] For sake of international comparison, the UK has 26,607 tweets on Syria and 28,342 tweets containing Twerk.  Apparently, twerking still has room for expansion in Britian.

[4] Not to be confused with the Beer Belly of America, which is an entirely different socio-spatial phenomenon that we anthropomorphized into a regional definition.

[5] Although models that include DC actually have a higher r-squared. It just seems better to exclude it from a state level model.

[6] Seriously, this is just a blog post comparing twerking and Syria. For this, you expect peer review?

[7] Except in the case of Mark (see [2]) in which case the answer is yes.

September 09, 2013

Hiring a full-time researcher to work with Mark at the Oxford Internet Institute

The Oxford Internet Institute is hiring a full-time researcher to work with Mark on an ESRC-DFID funded project, The Promises of Fibre-Optic Broadband: A Pipeline for Economic Development in East Africa. Employing case-studies, interviews, surveys and textual analysis in Kenya and Rwanda, this project examines the expectations and stated potentials of broadband Internet and compares those expectations to on-the-ground effects that broadband connectivity is having in three economic sectors: tea production, tourism, and business process outsourcing.

This is an exciting role in which the Researcher will conduct in-depth qualitative research on the topic of connectivity, value chains, information flow, and exclusion in Rwanda. The researcher will also contribute to the dissemination of this work through academic papers and project reports.

Candidates should have experience of social science research in Development Studies, Geography, Sociology, Social Anthropology, Communications, or related disciplines and a strong record of training and practical experience in qualitative research methodology.

Based primarily at the Oxford Internet Institute (with periods of fieldwork in Rwanda), this position is available from immediately for 10 months in the first instance, with the possibility of renewal thereafter, funding permitting. We will soon be starting a multi-year project focusing on knowledge economies in Sub-Saharan Africa. We would therefore also welcome applications from candidates who are keen to be part of a larger research programme in order to extend the position.

The deadline is September 27 and interviews for those short-listed are planned to take place on October 15th and 17th. More info and an application package is available here, but feel free to get in touch if you have any question about the job.

September 05, 2013

Schmos, Schmucks and Schlongs, Oy vey!

Oy vey. It has been a very busy, long summer and due to some glitches we've fallen behind in producing posts for the blog like some kind of nebbish. We could kvetch some more but no one likes a nudnik and beside we know all of our readers are real mensches and won't complain and become pains in our tukhus.

Besides, Rosh Hashanah is upon us and we have just enough time to power up the patented FloatingSheep mapping chutzpah and create a special holiday post... Mazel Tov!

And in case you haven't figured it out, today's theme is Yiddish, that wonderfully expressive language of the Jews of central and eastern Europe and more recently (by which we mean the past century) of New York. Drawing from the DOLLY database, aka the golem of the geoweb, we compiled maps of tweets in the USA for the most common yiddish words used in English. Ok, well, Wikipedia complied the list and we made the maps.

Since it is a holiday, we'll keep things short and simple. A key finding is that Yiddish words are alive and well on Twitter within the US, albeit primarily used as single words rather than in whole phrases or sentences. For example, there is a whole lot of "Oy" and "Oy vey" in the Twitterverse. Likewise, the surprisingly long list of Yiddish terms for penis (putz, schlong, schmuck) are running amuk like some kind of meshuggener, which upon reflection makes sense. Nosh is also very popular relative to other Yiddish terms, such as the delightful zaftig which is not as heavily used.

Below, you'll find a series of maps showing how these various terms are distributed across the U.S. Shalom.

Yiddish words are predominantly used in large cities in the US. The map of Yiddish speakers on Wikipedia suffers from the modifiable area unit problem, so not aggregating to the level of the state is more illustrative here.

chutzpah: nerve, guts, daring, audacity, effrontery (Yiddish חוצפּה khutspe, from Hebrew)

kvetch: to complain habitually, gripe; as a noun, a person who always complains (from Yiddish קװעטשן kvetshn 'press, squeeze', cf. German quetschen 'squeeze')

People in the north east kvetch more on Twitter than in other areas of the country.

mensch: an upright man; a decent human being (from Yiddish מענטש mentsh 'person', cf. German Mensch

nosh: snack (noun or verb) (Yiddish נאַשן nashn, cf. German naschen)

oy or oy vey: interjection of grief, pain, or horror (Yiddish אוי וויי oy vey 'oh, pain!' or "oh, woe"; cf. German oh weh

schlep: to drag or haul (an object); to walk, esp. to make a tedious journey (from Yiddish שלעפּן shlepn; cf. German schleppen)

schlong: (vulgar) penis (from Yiddish שלאַנג shlang 'snake'; cf. German Schlange)

There was more intense discussion of schlongs in smaller cities and suburbs throughout the United States.

schmo: a stupid person. (an alteration of schmuck; see below)

schmuck: (vulgar) a contemptible or foolish person; a jerk; literally means 'penis' (from Yiddish שמאָק shmok 'penis', maybe from Polish smok 'dragon')

schmutz: dirt (from Yiddish שמוץ shmuts or German Schmutz 'dirt')

schnoz or schnozz also schnozzle: a nose, especially a large nose (perhaps from Yiddish שנויץ shnoyts 'snout', cf. German Schnauze)

shtup: vulgar slang, to have intercourse (from Yiddish שטופּ "shtoop" 'push,' 'poke,' or 'intercourse'; cf. German stupsen 'poke')

Shtup is used evenly across the country, perhaps as a misprint for "shut up" in conversations, but then again shtuping is a popular activity across time and space.

spiel or shpiel: a sales pitch or speech intended to persuade (from Yiddish שפּיל shpil 'play' or German Spiel 'play')

Spiel is used more in small cities, such as around Marion, Illinois and Sandusky, Ohio.

tush (also tushy): buttocks, bottom, rear end (from tukhus

yutz: a fool 

zaftig: pleasingly plump, buxom, full-figured, as a woman (from Yiddish זאַפֿטיק zaftik 'juicy'; cf. German saftig 'juicy') 

There is no particular pattern of where 'zaftig' is used more, apparently the pleasingly plump are distributed throughout the continental United States.

August 14, 2013

Visualizing the Relational Spaces of Hurricane Sandy

Nearly a year ago, Hurricane Sandy made landfall on the eastern seaboard of the US, wreaking havoc on the lives of millions of people in its path. At the time, we threw together some quick maps of where Sandy was being talked about on Twitter, and how the geographies of Sandy-related tweets were both intensely connected to the material impacts of the storm, but also somewhat incongruent.

Since then, we've been putting the finishing touches on a paper that extends our initial interest in the data shadows of Hurricane Sandy to a more comprehensive look at how we can use Sandy-related tweeting to understand the multidimensionality of the geographies of social media activity. All too often, a one-to-one connection is made between the location of a geotagged tweet or other piece of social media content and the content of that tweet [1]. We have instead been attempting to understand how we can think through, and then visualize, how geotagged tweets reflect and produce much more complex socio-spatial relations, which include both intense connections to the places where such content is produced, as well as much more physically distant locations which are brought closer in relational space through such informational flows. The rest of this post is adapted from our paper-in-progress, and outlines how we can map and measure the relational spaces of Hurricane Sandy.

Using T-100 Domestic Market data from the Research and Innovative Technology Administration (RITA) on flights and the number of passengers between city pairs in 2012, we determined the 50 cities that have the most passenger traffic with New York City, ranging from Chicago (3.5 million passengers back and forth) to Kansas City (175,000 passengers). Since operations and activities at some airports close to New York were directly affected by Sandy’s landfall, we exclude any airport within 500 kilometres of Manhattan in this analysis. For the remaining airports we used a buffer of 5km to collect all Hurricane Sandy related tweets and calculated the lower bound of the odds-ratio (or location quotient).  This metric measures the level of Hurricane Sandy tweets relative to overall Twitter activity . If relational networks did not play a significant role in Sandy-related tweeting, one would expect to see a direct distance decay effect: as the distance from New York City increases the odds-ratio should decrease.

Twitter Activity vs. Physical Distance

Our map shows, however, that physical distance has no significant relationship with the relative level of tweeting activity about Hurricane Sandy as is evidenced by both the scatterplot and the map (Spearman’s rho is -0.05). The map uses an azimuthal equidistant projection with New York City as the center, where the size of each airport is proportional to its odds ratio. Airports that are equally distant in physical terms from New York have widely diverging measures of Sandy-related Twitter activity. In addition, the average odds ratio in each 1000km zone does not decrease the further away one travels from New York.

In contrast, a slightly altered version of our map shows that the number of passengers between each city and New York City exhibits a much stronger positive correlation with the odds-ratio metric of Twitter activity (Spearman’s rho is 0.34). This figure preserves the directional bearing of each city with respect to New York City, but instead uses an inverse of the number of passengers to recalculate the relational distance between the cities. Airports are thus no longer displayed according to their physical distance from New York City, but rather based on the intensity of air traffic between the two cities. Since the bearing has remained the same, airports with a higher intensity will move closer to New York along that line, and vice versa. In addition to the correlation coefficient, we can also visually determine that cities with a lower odds-ratio, such as Pittsburgh and Memphis, have a tendency to move towards the outer circles while cities with a higher odds-ratio, such as San Francisco and Los Angeles, move relatively closer.

Twitter Activity vs. Air Traffic Interactivity

In other words, it is the relational connection to New York, measured by number of air travelers, not physical distance, which better explains the level of concern with Hurricane Sandy as expressed via Twitter. This concern, however, can vary within metropolitan territories depending upon the scale of analysis; some parts of an urban area may have much stronger relational ties to distant cities, while other parts are largely disconnected from such global flows.

To test the extent to which the data shadows of Sandy-related tweeting are a localized phenomenon within certain parts of metropolitan areas (rather than a more generalized territorial phenomenon), we increased the initial buffer around each airport from 5km to 25km. Thus, rather than just capturing neighborhoods that are spatially proximate to the airport, this measure captures a much wider swath of each metropolitan area. With this larger buffer, there is a near-reversal of the correlations illustrated in our first map, as Pearson’s rho for total number of passengers is now 0.06 (rather than 0.34), while the distance effect starts to emerge (rho is -0.15). In other words, even though the sociospatiality of a phenomenon like Sandy is expressed partly through a network of connections between territories, these connections are very much bounded by the locally-specific practices of place. This once again highlights the complex ways in which the digital data shadows of a material event are manifest through the intertwinement of different dimensions of social space.

As evidenced by these examples, Sandy’s data shadows are not evenly distributed through the continental United States. They are instead quite intense in some locations, while hardly reaching others at all, demonstrating the multiple spatial dimensions of social processes such as the response to Hurricane Sandy.
[1] We're as guilty of this as anyone.

July 17, 2013

Tweeting for Trayvon

While the not guilty verdict is in for George Zimmerman, the discussion about and ramifications of Trayvon Martin's killing seventeen months ago are only beginning, from protest marches throughout the country to tweeting with hashtags like #MillionHoodies. Plenty of people smarter than us have weighed in on what this means for the persistent racism and inequity of the justice system in the United States, so we'll leave that side of the analysis to them. But as we specialize in thinking about and analyzing the geographies of social media, we want to offer our own two cents on what we can collectively take away from the case based on an analysis of geotagged tweets reacting to George Zimmerman's acquittal for Trayvon Martin's slaying.

First, some quick notes on our methodology and general trends in the data. Using DOLLY, we collected all the geotagged tweets from July 1 through July 15, referencing either "JusticeForTrayvon" or "Not Guilty", capturing the usage of these phrases with or without an accompanying hashtag. There were a total of 27,863 tweets referencing "Not Guilty" in this time frame, and just 6,614 referencing "JusticeForTrayvon". We calculated location quotients using hexagonal binning in order to normalize the data based on a relative measure of tweeting activity, as well as to account for differential size of counties or other similarly arbitrary areal units [1]. More simply, this allows us to compare the relative level of Twitter activity in any particular location, rather than relying on raw counts which are biased by population density.

Timeline of Tweets Referencing "Trayvon" from July 13th-14th

In addition to our primary interest in the spatial dimension of tweeting, we're also able to visualize a timeline of tweeting activity, which shows a clear spike immediately following the verdict on Saturday evening around 10 pm. While we're sure that many people's timelines were filled with reactions to the verdict throughout the day on Sunday, it seems as though much of the tweeting became more dissipated throughout the day as protests heated up and others went back to their usual routines.

Taking a look at the spatial patterns of these keywords, there are some clear differences. While there are many fewer JusticeForTrayvon tweets overall, they tend to be generally scattered, but with some relative concentrations largely in the south, in cities like Shreveport and Alexandria, Louisiana and Durham, North Carolina. Again, these measured are normalized for overall level of Twitter activity and thus show that these places were more engaged in this topic via Twitter than other parts of the country.

References to Not Guilty, however, in addition to being far more prevalent, demonstrate significantly more clustering in areas of the country outside the south, especially in Texas (depending on whether or not you consider it to be Southern) and some of the Midwestern or Mid-Atlantic states. We should note that there is a greater concentration references to Not Guilty in the vicinity of Sanford, Florida, the location of Trayvon Martin's killing and the subsequent trial, than was visible in references to JusticeForTrayvon. 

It is also important to note, however, that large cities on the west coast, like Los Angeles, San Francisco and Seattle, have relatively little tweeting about the case for either term, as do major cities along the eastern seaboard, like New York, Boston, D.C. and Philadelphia, despite being the sites of the major protests following the verdict.

Comparing references to the two terms -- while keeping in mind that they are not entirely oppositional, i.e., "Not Guilty" is a much more neutral and contextually dependent phrase than JusticeforTrayvon, which explicitly 'takes sides' in this debate -- reveals a much clearer geographic pattern. This comparison brings the different geographies of these phrases into a stark contrast, with many more references to JusticeforTrayvon concentrated throughout the southern states of Arkansas, Louisiana, Mississippi, Alabama, Georgia and Kentucky (highlighted in purple), with a greater number of more generic references to the verdict (highlighted in green) scattered throughout much of the rest of the country. In short, the hashtag that is more closely associated with protesting the outcome of the court case, is more highly concentrated in Southern states.

One thing that is clear is that although the experience of racism isn't unique to the American South, it is uniquely associated with and experienced in that place when viewed through geotagged social media content [2]. This isn't to say that the tweeting about the case throughout the south is, in and of itself racist, as many, if not most, tweets express outrage at Zimmerman's acquittal, as evidenced by the large number of tweets referencing the JusticeForTrayvon hashtag. But given the back-and-forth around the particularity of racism in the south or the universality of racism across the United States, the higher concentration of this Twitter discussion within the region suggests a process distinct from the rest of the country.

The fact that Trayvon Martin's killing took place in Florida, which shares a similar history with regard to race as the rest of the south, has clearly elicited a broader reaction from those in a (relatively) similar geographic context. The complexities of racism (both historical and contemporary) as expressed in part through problematically-enforced laws like stand-your-ground come to the fore in the south at a time like this, as can be seen in the much higher-than-usual tweeting about the case in Alabama, Georgia, Mississippi, Louisiana and Arkansas. If anything, the outpouring of tweets throughout the south in support of the Martin family and in favor of a more sensible and equitable justice system serves to destabilize the common narrative that the south is unitary, coherent region populated by those clinging to nineteenth century racial mores. The south is, like any other place, marked by conflict and contradiction, something evident nowhere more than in the way it continues to deal with (or ignore) persistent racial inequality like that seen in Trayvon Martin's killing and George Zimmerman's acquittal.
[1] We've previously demonstrated the utility of this method for mapping concentrations of tweets about a given phenomena.
[2] See, for example, our work on mapping racist tweets in response to President Obama's re-election last November.