PreprintPDF Available

Disruptive social media virality: Amplifying indigenous resistances to shale gas and the Dakota Access Pipeline

Authors:
Preprints and early-stage research may not have been peer reviewed yet.

Abstract and Figures

We use shale gas exploration in New Brunswick, Canada and the Dakota Access Pipeline, as case studies of what we term disruptive social media virality. The Dakota Access Pipeline did not gain widespread mainstream media and public attention until it was nearly complete, despite extensive opposition from the Standing Rock Sioux Tribe and its supporters. It was not until images depicting repression circulated on social media that mainstream media and public attention spiked. We consider disruptive social media virality to involve constituents who perceive themselves as outsiders to decision-making processes on climate and energy issues. In both cases protest was ongoing for significant periods of time before they received mainstream media and public attention. Using mixed methods, we show that law enforcement and company crackdown on indigenous communities, and the circulation of dramatic visuals via Twitter, is associated with spikes in social media, as well as mainstream media and public attention. Keywords: Dakota Access Pipeline, Elsipogtog First Nation, Google Trends, Standing Rock Sioux Tribe, Twitter
Content may be subject to copyright.
1
2
3
4
5
Disruptive social media virality: Amplifying indigenous resistances to shale gas and the Dakota
6
Access Pipeline
7
8
Jill E. Hopke
1
9
DePaul University
10
Molly Simis-Wilkinson
11
Independent Researcher
12
Patricia A. Loew
13
Northwestern University
14
15
November 26, 2018
16
17
Working paper based on research presented at the 2017 Conference on Communication and the
18
Environment (COCE) at the University of Leicester in Leicester, UK and at the 9th International
19
Conference on Social Media and Society at the Copenhagen Business School,
20
Copenhagen, Denmark.
21
22
23
1
Author contact information: College of Communication, DePaul University, 1 East Jackson Blvd., Chicago IL
60604; email: jhopke@depaul.edu.
RUNNING HEAD: Disruptive social media virality
2
24
25
26
ABSTRACT
27
We use shale gas exploration in New Brunswick, Canada and the Dakota Access Pipeline, as
28
case studies of what we term disruptive social media virality. The Dakota Access Pipeline did
29
not gain widespread mainstream media and public attention until it was nearly complete, despite
30
extensive opposition from the Standing Rock Sioux Tribe and its supporters. It was not until
31
images depicting repression circulated on social media that mainstream media and public
32
attention spiked. We consider disruptive social media virality to involve constituents who
33
perceive themselves as outsiders to decision-making processes on climate and energy issues. In
34
both cases protest was ongoing for significant periods of time before they received mainstream
35
media and public attention. Using mixed methods, we show that law enforcement and company
36
crackdown on indigenous communities, and the circulation of dramatic visuals via Twitter, is
37
associated with spikes in social media, as well as mainstream media and public attention.
38
Keywords: Dakota Access Pipeline, Elsipogtog First Nation, Google Trends, Standing Rock
39
Sioux Tribe, Twitter
40
41
RUNNING HEAD: Disruptive social media virality
3
Disruptive social media virality:
42
Amplifying indigenous resistances to shale gas and the Dakota Access Pipeline
43
Indigenous communities are on the frontlines of resource extraction, as well as the
44
impacts of climate change. In October 2013 an iconic photograph circulated in social media
45
across North America. In it, a young First Nations woman holding an eagle feather kneels before
46
a wall of heavily armed Canadian police. Three years later, in November 2016, images of law
47
enforcement using water cannons on anti-Dakota Access Pipeline demonstrators captured public
48
attention. Unarmed water protectors, as the demonstrators called themselves, near the Standing
49
Rock Sioux Indian Reservation brace against the icy blast of water cannons fired by sheriff’s
50
deputies. In both cases, imagery visually represented indigenous communities struggling against
51
multinational energy projects. Mainstream media largely ignored the concerns of the
52
demonstrators on-the-ground and their supporters until participation in the protests became
53
disruptive. These disruptions prompted crackdown by authorities and co-occurred with peaks in
54
social media posting.
55
To study the role of social media in amplifying activism we examine how the Elsipogtog
56
and Standing Rock protests spread on Twitter in relation to spikes in attention and salience on
57
the mainstream media and public agendas. In recent years, climate activists have increasing
58
forged alliances with native communities in opposition to pipelines, including those involved
59
with protests against the Keystone XL and Dakota Access projects, connecting climate activism
60
to public opposition to large-scale energy projects (see Johnston, 2017; McKibben, 2016).
61
We argue that Standing Rock water protectors, like the Elsipogtog demonstrators before
62
them, used Twitter to reach beyond their core supporters. We compare these two cases, separated
63
by three years at their respective peaks, in order to make more generalizable statements about the
64
RUNNING HEAD: Disruptive social media virality
4
function of what we term disruptive social media virality within activism originating from
65
marginalized communities. There are parallels in the historical experiences of colonialism
66
between the Elsipogtog Mi’kmaq and Standing Rock Sioux, as well as the violent crackdown
67
demonstrators in both cases faced from law enforcement. Their respective movements both used
68
Twitter to reach bystanders. As our two case studies show this type of virality is possible when
69
political opportunities are perceived to be closed-off to social movements (see McAdam, 1999).
70
We studied Twitter because the platform serves as, according to technology journalist
71
Farhad Manjoo, a “global gathering space for live events” (2015, para. 6). Beyond that, research
72
shows that influential actors—such as journalists, activists, and politicians—gravitate to it for
73
political communication purposes (Himelboim, Hansen, & Bowser, 2013). This study fits into
74
Slater and Gleason’s (2012) framework as content analysis (strategy 8a).
75
Social Media and Issue Virality
76
Individual protest events are not independent. Rather protest events need to be
77
understood within their political, economic, and social context (Koopmans, 2004). Social
78
movement scholar Sidney Tarrow defines protest cycles as periods of “heightened conflict and
79
contention” (1994, p. 153). These are “waves,” or periods of expanding, transforming, and
80
contracting political contention (Koopmans, 2004, p. 22).
81
As a message circulates there is a “sharp acceleration” in its spread (Nahon & Hemsley,
82
2013, p. 25). “Interest networks” form around viral content, with emotional impact for audiences,
83
heightening public attention (Nahon & Hemsley, 2013, p. 72). At times virality cannot be
84
isolated to a single piece of content, what Nahon and Hemsley (2013) term a “viral topic,” made
85
up of multiple streams of visually similar viral and non-viral content (p. 11).
86
RUNNING HEAD: Disruptive social media virality
5
To successfully challenge the status quo, movements need attention and salience to gain
87
traction for their demands. While social media are potentially powerful tools, activists still need
88
mainstream media attention to capture the hearts and minds of bystanders. Activists can use
89
social media, particularly in cases of government crackdowns, to enhance “disruptive social
90
movement capacity” (Tufekci, 2017, p. 199).
91
To explore the relationship between how Standing Rock and Elsipogtog topics were
92
discussed on social media and the respective media and public agendas, we ask:
93
RQ1a: What is the relationship between Twitter post volume and mainstream media
94
attention to each issue?
95
RQ1b: What is the relationship between Twitter post volume and public attention to each
96
issue?
97
In the case of the real-time circulation of images showing police brutality on social media
98
(Tufekci, 2017), these apps and platforms can serve as mechanisms for movements to enhance
99
their disruptive capacity. To theorize the underlying processes, we conceptualize this as
100
disruptive social media virality. We consider disruptive virality to involve stakeholders
101
perceiving themselves to be outsiders in the decision-making processes, or who are from
102
historically marginalized groups, who make use of social media to amplify and document dissent
103
when traditional modes of public participation have been exhausted or are perceived to be
104
ineffective for achieving their goals. Thus, we hypothesize:
105
H1: The Twitter volume for each issue will peak during periods of heightened crackdown
106
on demonstrators on-the-ground.
107
Social Movement Framing, Media and a Political Process Approach
108
RUNNING HEAD: Disruptive social media virality
6
We take a political process perspective on social movements, as outlined by sociologist
109
Doug McAdam (1999) in his groundbreaking analysis of the Civil Rights Movement. This
110
framework is well-suited for our analysis because of its aim of studying movements as holistic
111
processes, rather than sets of discrete events. It is not enough to share a collective identity and be
112
oppressed or marginalized by mainstream political actors. A shift in traditional political power
113
structures and opportunities is required (McAdam, 1999).
114
Historically, mainstream media have been sites where contests over the symbolic
115
meanings of social movements have played out (Gamson, 2004; Gitlin, 1980). The 1973
116
Takeover at Wounded Knee is one such example. Some broadcast news reports depicted
117
members of the American Indian Movement as political terrorists while other portrayed them as
118
champions of civil rights (Briggs, 1973; Miller, 1973). Thus, media have a “dual role” as sites
119
where the symbolic interests of movements, their opponents, and authorities are contested, and
120
also as cultural outputs (Gamson, 2004). Media define the “standing” of actors to speak in the
121
public sphere and whether a group is seen as an active “agent,” rather than an “object” discussed
122
by bystanders (Gamson, 2004, p. 251). The goal activists have in propagating frames is to sustain
123
media standing in order to advance collective aims (Gamson, 2004).
124
Political sociologists have theorized that in order to understand individual protest events,
125
an explanation for changes to existing power structures is required (see McAdam, 1999; Piven &
126
Cloward, 1979). In this political process model, an increase in collective action can be explained
127
by the confluence of three factors: 1) the level of organization within groups 2) their assessment
128
of the possibilities for success, and 3) the ability of collective actors to appeal for the support of
129
other sectors in society (McAdam, 1999). Movements are also impacted by external factors, such
130
RUNNING HEAD: Disruptive social media virality
7
as the response of authorities, within which they are engaged in a cyclical relationship
131
(McAdam, 1999).
132
Social media and discursive opportunities. Social media have opened new discursive
133
opportunities for social movements from the 2011 Arab Spring uprisings in the Middle East to
134
Black Lives Matter from 2013 onward and #MeToo campaigning against sexual violence and
135
harassment that started in 2017. Social media allow fluid organizing structures for movements,
136
which Bennett and Segerberg (2012) term “connective action” (p. 744). This happens when
137
organizers use social media to amplify internet-mediated protests by spreading personalized
138
messages, or action frames, through their social networks (Bennett & Segerberg, 2012). Thus,
139
social technologies enable activists to challenge mainstream media gatekeepers (Meraz &
140
Papacharissi, 2013; Papacharissi & de Fatima Oliveira, 2012).
141
We examine whether Standing Rock-themed Twitter posts include a greater degree of
142
collective action frames related to the concerns of activists and their supporters than do those
143
posts focused solely on the pipeline. To compare these two conversation streams to the anti-
144
fracking protests in Elsipogtog, an issue that spiked for a limited time period and lower volume
145
of posts, we ask:
146
RQ2: What are the social movement collective action frames and mainstream, counter-
147
frames present in each of the three Twitter data streams?
148
Case Background
149
Elsipogtog and Shale Gas Exploration
150
Elsipogtog First Nation, a small Mi’kmaq community in New Brunswick, is thousands of
151
miles away and culturally distinct from the Standing Rock Sioux Tribe, a Lakota reservation that
152
RUNNING HEAD: Disruptive social media virality
8
straddles North and South Dakota.
2
However, the two indigenous communities have much in
153
common. Historically, both encountered settler forces. Both saw their land bases shrink and their
154
economies ravaged by expansionism. Both were targets of forced assimilation through
155
residential/boarding school policies and each continues to experience crippling poverty and
156
joblessness. Through their protests against shale gas exploration and the Dakota Access Pipeline
157
respectively, Elsipogtog and Standing Rock respectively asserted their sovereignty in attempts to
158
stop what each perceived to be an environmental threat to their land, culture, and way of life
159
(Meyer, 2016; Schwartz & Gollom, 2013).
160
In Elsipogtog, English and French legacy media
3
only sporadically covered the concerns
161
voiced by the Mi’kmaq over fracking and usually framed the coverage within the context of Idle
162
No More, a national indigenous environmental movement (see Kelley, 2014). In South Dakota,
163
legacy newspapers like the Bismarck Tribune carried stories about the pipeline, but most of them
164
focused on the permitting process and the economic impact of the project. National television
165
coverage of the Dakota Access Pipeline protests was non-existent until September 2016 when
166
events became violent.
167
Elsipogtog First Nation historical background. The 3,300 citizens of Elsipogtog First
168
Nation identify as Mi’kmaq, a large Algonquian-speaking community residing in Canada’s
169
Maritime Provinces and Newfoundland (Indigenous and Northern Affairs Canada, 2017).
170
As early as the sixteenth century, the Mi’kmaq encountered European fishers and traders
171
off the coast of present-day Nova Scotia (Howe, 2015; Discover Canada, 2015). Beginning with
172
2
There are three divisions of Sioux: the Lakota, Dakota, and Nakota, who live in the United States and Canada.
About 170,000 Lakota live primarily on five reservations, including Pine Ridge, Cheyenne River, Rosebud, Lower
Brule, and Standing Rock.
3
The term legacy media refers to older, established media such as radio, television, and especially newspapers.
RUNNING HEAD: Disruptive social media virality
9
Jacques Cartier’s first voyage in 1534, King Francis I claimed vast portions of Mi’kmaq land for
173
France and provided land grants and rights of occupancy to French aristocrats. As Howe pointed
174
out, “these rights were never the French monarchy’s to give [to settlers], and the indigenous
175
peoples were never consulted” (2015, p. 22). This pattern would repeat itself throughout history.
176
By the eighteenth century, upwards of 6,000 Acadian settlers were living in Mi’kmaq
177
territory as were a new set of occupiers, the British, who signed peace and friendship treaties
178
with the Mi’kmaq. Of the nearly two million acres that originally comprised Elsipogtog, today
179
barely 2,000 acres remain. In 1999, the Canadian Supreme Court ruled that the Peace and
180
Friendship treaties of 1760-61 did not cede land or resources. In other words, the land had been
181
taken from the Mi’kmaq without their consent.
182
Shale gas exploration. In 2013 fracking became a crucial talking point when the
183
provincial government of New Brunswick—without consulting Elsipogtog—signed a contract
184
with Southwestern Energy, authorizing the company to explore for shale gas near the Elsipogtog
185
Reserve. Despite promises that fracking would stimulate the economy and reduce
186
unemployment, a women-centered grassroots opposition movement emerged. They argued that
187
the provincial government had no right to compromise Mi’kmaq’s clean water without
188
consulting the First Nation (Vowel, 2013). The Elsipogtog found their message amplified by
189
“Idle No More,” a national indigenous movement opposed to legislation perceived to dismantle
190
Canada’s environmental protections.
191
Standing Rock and the Dakota Access Pipeline
192
RUNNING HEAD: Disruptive social media virality
10
Standing Rock is one of five Lakota Sioux reservations in the United States.
4
Its 8,200
193
tribal members struggle with poverty and an unemployment rate of 86 percent (Bureau of Indian
194
Affairs, 2005). Pre-contact, the Lakota were primarily a horse culture and survived on bison.
195
They first encountered French fur traders and later Americans making their way west.
196
Standing Rock historical background. In 1851 the U.S. government pressured the
197
Lakota and seven other nations to sign the first Fort Laramie treaty. Under the agreement, the
198
Lakota identified their boundaries, which included the Black Hills in present-day South Dakota.
199
In exchange for annuity payments, the Lakota agreed to allow settlers heading west along the
200
Oregon and Bozeman Trails safe passage across their territory. The treaty was almost
201
immediately broken.
202
In 1868, the United States negotiated a second Fort Laramie Treaty with the Lakota that
203
also attempted to protect white travelers passing through Lakota lands and again acknowledged
204
Lakota claims to the Black Hills. Within a few short years, however, gold was discovered in the
205
Black Hills, and miners poured into the territory. The U.S. sent military forces to protect the
206
miners and for the next nine years, fighting was constant. In June 1876, a combined force of
207
2,500 Lakota, Cheyenne, and Arapaho annihilated George Armstrong Custer and the battalion
208
under his command at the Battle of Greasy Grass (Little Big Horn).
209
Fourteen years later the 7th Cavalry—Custer’s old regiment—killed an estimated 250
210
men, women, and children at Wounded Knee, South Dakota (West Film Project, 2001). In 1973,
211
during the so-called “Red Power” movement, members of the American Indian Movement
212
occupied the site of the massacre, asserted tribal sovereignty, and spoke out against broken
213
4
There is a Lakota Reserve in Canada, the Wood Mountain Reserve near Moose Jaw, Saskatchewan, where
descendants of Sitting Bull’s Hunkpapa Band reside.
RUNNING HEAD: Disruptive social media virality
11
treaties. They drew attention to the two counties straddling Standing Rock, where crushing
214
poverty, substandard housing, and poor health care are a way of life.
215
The Dakota Access Pipeline. In 2014 Energy Transfer Partners announced it intended to
216
build a 1,172-mile pipeline to carry crude oil from the Bakken oil fields in North Dakota to
217
Patoka, Illinois. Over the next two years it held public hearings, received approvals, and
218
negotiated contracts with or filed condemnation suits against landowners., Initially, the route
219
took the pipeline under the Missouri River north of Bismarck. After residents protested that it
220
threatened the city’s water supply, company officials re-routed it so that it would travel under the
221
Missouri River 500 feet north of the Standing Rock Sioux Reservation near Cannon Ball.
222
Opponents saw the decision as having racial overtones. Bismarck is 95 percent Caucasian;
223
Cannon Ball is 93 percent Native American. In January 2016 the North Dakota Public Service
224
Commission unanimously approved the project (Bismarck Tribune, 2016).
225
In April 2016, LaDonna Bravebull Allard, whose land lies adjacent to the pipeline,
226
invited like-minded pipeline opponents to camp on her property. Standing Rock young people,
227
including a group calling itself “REZpect our Water” responded, as did celebrities including
228
Shailene Woodley, who live-streamed her own arrest on Facebook (Woodley, 2016). Over the
229
next few months the camp swelled as hundreds, then thousands of indigenous people and their
230
supporters arrived at Standing Rock. Calling themselves “water protectors,” they set up schools,
231
kitchens, and temporary housing. Other camps sprang up as well. CNN estimated the number of
232
camp dwellers at more than 10,000, representing some 400 indigenous communities (CNN,
233
2017). More than half a million people signed REZpect’s petition on Change.org asking the
234
Army Corps of Engineers to stop construction (Change.org, 2016). The group also organized a
235
RUNNING HEAD: Disruptive social media virality
12
successful 2,000-mile run from Standing Rock to Washington, D.C. to garner mainstream media
236
attention to the anti-pipeline cause (Alli, 2016).
237
Methods
238
Data Collection
239
We conducted the Twitter data collection and analysis using computer-assisted coding
240
through the Crimson Hexagon ForSight software. ForSight relies on machine learning content
241
analysis methods described by Hopkins & King (2010), in which algorithms track linguistic
242
patterns that correspond to underlying concepts embedded in social posts identified by human
243
coders. Researchers use the Crimson Hexagon ForSight software to train an algorithm to classify
244
a corpus of social data into statements, or sentiments, that were previously identified through
245
human coding. The unit of analysis in the software output is statements (see Pew Research,
246
2012). In the results section we report the results of the thematic analysis in terms of these
247
statements, or sentiments. We report the results of the comparison of Twitter post volume with
248
mainstream media and public attention in terms of tweet volume.
249
To use ForSight, researchers define a dataset by setting parameters of a date-range and
250
Boolean keywords. Specifics on how keywords for each of our data streams and codes were
251
generated are discussed below. Then, researchers identify mutually exclusive and exhaustive
252
coding categories, and ensure reliability of these codes by conducting inter-coder reliability tests.
253
Then, twenty relevant tweets are sorted into each category to train the algorithm. At that time,
254
the software analyzes the corpus of posts. With this combination of human and computer coding,
255
ForSight has high reliability and validity, and so has been used to analyze social media discourse
256
in a variety of contexts (e.g. Su et al., 2017).
257
RUNNING HEAD: Disruptive social media virality
13
In terms of ethical considerations, this study was deemed exempt from Institutional
258
Review Board (IRB) review by the relevant university’s IRB board. We limited our data
259
collection to public Twitter posts. For the example tweets provided, unless individuals are public
260
figures we do not identify any of the posts by individual author names, or Twitter user handles,
261
to protect their identities. The frames in the data are discussed in aggregate, which provides an
262
additional layer of de-identification in most cases.
263
Standing Rock and Dakota Access Pipeline data collection. If we had analyzed
264
Standing Rock and DAPL discourse in a singular dataset, this would have conflated the
265
differences between the codes with difference inherent to the conversation streams.
266
Acknowledging the role of researchers in actively constructing datasets, our two datasets allowed
267
us to compare what we expected to be more pro-#noDAPL movement conversation (Standing
268
Rock dataset) to conversation we expected to be focused on the pipeline infrastructure itself and
269
logistics (DAPL dataset). We considered that the DAPL dataset may include more industry
270
supporters, following findings by Hopke and Simis (2017) that anti-fracking activists and shale
271
gas industry supporters used different hashtags on Twitter.
272
To compare the two sets of conversation, we collected two datasets posted between April
273
1, 2016 and March 31, 2017. This time period includes a full year following the start of the first
274
protest encampment in April 2016, encompassing months of heightened tensions on-the-ground
275
and the resulting crackdown by law enforcement and security forces, as well as the eventual
276
abandonment of the camps.
277
We conceptualized “social media attention” as a measure of the fluctuating volume of
278
Twitter posting over a one-year period of time for the case studies. When sharing news via
279
Twitter, individuals are more likely to retweet, meaning share the posts of others, than to post
280
RUNNING HEAD: Disruptive social media virality
14
original content (Pew Research Center, 2015b). Thus, we included retweets within our corpus of
281
data. While not representative of all social media activity, Twitter remains a useful social media
282
outlet for a study such as ours, as Twitter users turn to the medium for breaking news (Pew
283
Research Center, 2015a, 2015b).
284
We developed sets of keywords used for data collection through inductive research,
285
including major activist and industry hashtags, along with the name of the largest water protector
286
encampment and the tribe. Twitter posts for the Standing Rock dataset were retrieved based on
287
the following set of Boolean operator keywords: “standing rock” OR #standingrock OR
288
#noDAPL OR #standwithstandingrock OR “Standing Rock Sioux” OR “Oceti Sakowin Camp”
289
OR #DefendtheSacred. A total of 13,596,975 tweets were collected, with a reach of 39 billion
290
potential impressions, a Crimson Hexagon measure that sums the follower counts of those
291
tweeting about the topic during the specified time period.
292
For the DAPL dataset, the following Boolean operator keywords were used:
293
#DakotaAccessPipeline OR “Dakota Access Pipeline” OR #DakotaPipeline. In order to isolate
294
posting about the pipeline, we excluded the terms used for the Standing Rock dataset: “standing
295
rock,” #standingrock, #noDAPL, #waterislife, #standwithstandingrock, “Standing Rock Sioux,”
296
“Oceti Sakowin Camp,” and #DefendtheSacred. A total of 1,571,896 tweets were collected, with
297
a reach of 15 billion potential impressions as measured by Crimson Hexagon.
298
As noted, we excluded the Standing Rock keyword operators from the DAPL data
299
collection. Given that the protest was integrally about the pipeline it would not have made sense
300
conceptually to exclude the DAPL search terms from the Standing Rock data collection.
301
Therefore, it is possible that there is a degree of overlap between the two data streams.
302
RUNNING HEAD: Disruptive social media virality
15
Elsipogtog data collection. To examine the Elsipogtog case, we collected a dataset of
303
tweets posted between January 1 and December 31, 2013. This period includes the height of
304
Elsipogtog First Nation opposition to shale gas exploration in New Brunswick and when the
305
company, SWN Resources Canada, ceased shale gas exploration in the province in December
306
2013 (APTN National News, 2013).
307
The Elsipogtog dataset utilized the Boolean operator keywords: (#cdnpoli AND
308
elsipogtog) OR (#rcmp AND elsipogtog) OR (#INM AND elsipogtog) OR (#IdleNoMore AND
309
elsipogtog) OR (#nbpoli AND elsipogtog) OR (#FirstNation AND elsipogtog) OR (#FN AND
310
elsipogtog) OR elsipogtog) OR (#elsipogtog). A total of 185,331 tweets were collected, with a
311
reach of 490 million potential impressions as measured by Crimson Hexagon.
312
Mainstream media attention data collection. To study the relationship between
313
disruptive social media virality and mainstream media attention, we compared the volume of
314
tweets to press mentions of Elsipogtog and Standing Rock. We operationalized “mainstream
315
media attention” as the volume of mainstream media stories mentioning: (1) “Standing Rock”
316
and the “Dakota Access Pipeline” for the Standing Rock case, and (2) “Elsipogtog” and “shale
317
gas” for the Elsipogtog case.
5
318
For the Standing Rock case, we used the LexisNexis Academic database to measure the
319
frequency of mainstream media mentions of “Standing Rock” combined with “Dakota Access
320
Pipeline” from April 1, 2016 through March 31, 2017. The variable was operationalized to
321
include the frequency of mentions in “U.S. Newspapers” including full-text, English language
322
5
Print mainstream media are used as a measure of media attention. Newspapers, while not the only form of
mainstream media, have been used by scholars to measure the volume of attention to climate change (see Boykoff &
Boykoff, 2004; Schmidt, Ivanova, & Schäfer, 2013).
RUNNING HEAD: Disruptive social media virality
16
newspapers published in the United States, such as local newspapers and national ones like USA
323
Today.
324
For the Elsipogtog case, we used the LexisNexis Academic database to measure the
325
frequency of mainstream media mentions of “Elsipogtog” combined with “shale gas” from
326
January 1 through December 31, 2013. LexisNexis does not provide a search category for
327
Canadian newspapers only. Therefore, the variable was operationalized to include the frequency
328
of mentions in full-text English language publications.
329
Finally, we include “Major World Publications” as a measure of international press
330
attention for each time period, including full-text news sources major international newspapers,
331
magazines and trade publications.
332
Google search data collection. Finally, we measured “public attention” to each issue
333
through the volume of worldwide Google search for each time period. This data was collected
334
through Google Trends. Google dominates U.S. and global search by a wide margin, with 63.5%
335
of U.S. desktop searches and an estimated 95% of global mobile search volume (Sterling, 2017).
336
Google Trends scales search data from 0-100, with 100 representing the highest point in search
337
volume for a specified time period. Past research suggests that Google search volume can
338
represent the level of public interest in a topic and may be correlated with issues high on the
339
mainstream media agenda (Stocking & Matsa, 2017).
340
Data Analysis
341
The goal of our content analysis was to measure the relative frequency of conversation
342
across Twitter hashtags along five dimensions of collective action frames and counter-frames
343
(see Table 1). Each issue has a culture demarcated through interpretive packages, or frames
344
(Gamson & Modigliani, 1989), which give them meaning (Entman, 1993; Goffman, 1974).
345
RUNNING HEAD: Disruptive social media virality
17
Movements generate collective action frames, which are shared meanings that provide a
346
framework for their activities (Benford & Snow, 2000; Gamson, 1992). Collective action frames
347
are adversarial. They depend on defining a clear target responsible for an injustice about which a
348
collective we can take action (Gamson & Meyer, 1996). Collective action frames are culturally-
349
bounded and contextually dependent (Snow, 2004). They have three attributes: injustice, agency,
350
and identity (Gamson, 1992).
351
We further grounded our coding schema in work on police and protestor interaction on
352
social network sites (Earl et al., 2013), as well as inductively determined frames by examining a
353
sub-sample of posts in order to ensure that the coding schema was exhaustive. Two coders (the
354
first and second author) iteratively developed the codebook through open coding of the data for
355
emergent frames and conducted inter-coder reliability in order to program the software with
356
mutually exclusive and exhaustive codes. Refer to Appendix A for code definitions and example
357
tweets for each code.
358
[INSERT TABLE 1 ABOUT HERE.]
359
The coding categories were assessed for their reliability through inter-coder reliability
360
tests on 500 randomly selected posts for each of the three datasets using ReCal software
361
(Freelon, 2013). At least 20 posts were coded for reliability in each content category for each
362
dataset. Posts that were irrelevant, and not fitting into any of the above-defined coding
363
categories, were categorized as “off-topic.When sensible, we refined the Boolean-string that
364
defined the parameters of the dataset to minimize off-topic posts. The Scott’s Pi for all coding
365
categories for each monitor were > .80. The average Scott’s Pi for each monitor is as follows:
366
Elsipogtog (.85), Standing Rock (.82), Dakota Access Pipeline (.85). The lowest item for each
367
dataset was .74, within acceptable ranges. These coding categories, once their reliability was
368
RUNNING HEAD: Disruptive social media virality
18
tested, were those used in computer-assisted analysis of the entire datasets using the ForSight
369
intelligent algorithm.
370
Collective action frames.
371
Movement response/presence. The first dimension captured conversation about
372
“movement response/presence.” Within this dimension, we coded for several collective action
373
frames. The first, water protection” includes mentions of water protectors, the term preferred by
374
demonstrators, as well as the need to protect water resources and “Mni Wiconi,” meaning “water
375
is life” in the Lakota language. In contrast, the counter-frame “protest and blockade action” is
376
conceptualized as references to the act of “protest” on-the-ground and blockade actions. The
377
second collective action frame “Native/treaty rights” includes mentions of honoring treaty and
378
Native rights, as well as explicit support from other indigenous nations. Lastly, “solidarity” as a
379
collective action frame includes messages of solidarity with water protectors that do not specify
380
they are from other Native communities or individuals, such as the support of celebrities (e.g.
381
Mark Ruffalo) and politicians (e.g. former presidential candidate Senator Bernie Sanders).
382
Calls for government intervention. The third dimension “appeals to politicians and
383
elected officials” includes calls for specific policy actions. The height of controversy over
384
completion of the Dakota Access Pipeline took place during the 2016 presidential election.
385
“Calls for government intervention” is conceptualized as a collective action frame calling for
386
governmental action favoring protestor goals (e.g. stop the pipeline), as well as mentions of
387
sitting and elected politicians (e.g. U.S. President Donald Trump pre-inauguration).
388
Counter-frames.
389
Use of force. This dimension captures “force and surveillance.” Within this category, the
390
counter-frame “use of force” is conceptualized as mentions of industry or state sanctioned force
391
RUNNING HEAD: Disruptive social media virality
19
against demonstrators, including arrests, dispensing crowds, and deployment of equipment and
392
weapons.
393
Against the movement.
6
For the Standing Rock and Dakota Access Pipeline datasets, we
394
also included a counter-frame “against the movement,” which captured statements in opposition
395
to the #noDAPL movement, often including the hashtag #MAGA, short for the Trump
396
presidential campaign slogan “Make America Great Again.”
397
Other.
398
Hashtag jumping. Lastly, a dimension unique to the breakout #noDAPL hashtag that has
399
become synonymous with the movement, was “hashtag jumping,” a phenomenon that has been
400
studied by other scholars (see Christensen, 2013). We conceptualize “hashtag jumping” as the
401
usage of the #noDAPL hashtag for an unrelated or irrelevant topic.
402
Chi-square tests for independence. Given that the data are nonparametric and
403
categorical, we conducted a series of seven chi-square tests for independence using the statistical
404
software package SPSS version 24. The data, as classified by the Crimson Hexagon software,
405
violates the assumption of independent observations between groups because the program can
406
apply more than one conversational topic code to an individual tweet. Therefore, we ran separate
407
chi-square test for each topic (e.g. “use of force”) with the presence of the frame coded “yes” and
408
the absence coded “no” (Pallant, 2013). The data were weighted by count.
409
To test the strength of the relationships, we examined Cramer’s V test statistics. A higher
410
value signifies a stronger association between the variables (Hayes, 2005). For “against the
411
movement,” comparing only two groups in a 2x2 contingency table, we report the Yates’
412
6
The Crimson Hexagon software requires 20 mutually exclusive posts for each category to machine classify the
data. This required us to omit the “against the movement” and “hashtag jumping” dimensions from the Elsipogtog
dataset because there were not enough exemplar posts to train the algorithm.
RUNNING HEAD: Disruptive social media virality
20
Correction for Continuity chi-square statistic and the phi coefficient for effect size (Pallant,
413
2013).
414
Results
415
Collective Action Frames and Counter-frames
416
For the Standing Rock dataset, 33% of statements within the corpus discussed “use of
417
force,” 14% focused on “protest and blockade action,” with “Native/treaty rights” and
418
“solidarity” both comprising 13% of the overall conversation. 11% of the Standing Rock
419
statements dealt with “water protection.” Conversation on unrelated topics hijacking the breakout
420
movement hashtag #noDAPL comprised 6% of posts. “Calls for governmental intervention,”
421
including appeals to both state and federal elected officials and agencies made up 5%, as did
422
statements “against the movement.”
423
For the Dakota Access Pipeline dataset, the highest volume of statements (25%) was
424
about “Native and treaty rights,” followed by 19% that addressed the “use of force” by Energy
425
Transfer Partners and governmental police agencies. Messages of “solidarity” for the Standing
426
Rock Sioux Tribe made up 16% and “calls for governmental intervention” 14%. Furthermore,
427
“protest and blockade action” comprised 10%, “water protection” 9%, and “against the
428
movement” 6%.
429
[INSERT FIGURE 1 ABOUT HERE.]
430
For the Elsipogtog dataset, the highest proportion of statements dealt with “protest and
431
blockade action,” followed by “use of force” at 20%, with “Native/treaty rights” and “solidarity”
432
both making up 19%. Lastly, 7% of the statements discussed “water protection” and less than 1%
433
included “calls for governmental intervention.”
434
RUNNING HEAD: Disruptive social media virality
21
Chi-square Tests for Independence
435
A series of five chi-square tests were conducted to explore our second research question
436
on the social movement collective action frames and counter-frames. All chi-square tests for
437
independence were significant (see Table 2). Our results show that the Standing Rock
438
conversation stream included the greatest degree of the collective action frame “water
439
protection.” In addition, results show that the Standing Rock conversation stream included the
440
greatest degree of “use of force” tweets.
441
Counter to our expectations, for the collective action frames of “Native/treaty rights” and
442
“solidarity,” the results showed Dakota Access Pipeline conversation stream included the
443
greatest degree of these frames.
444
[INSERT TABLE 2 ABOUT HERE.]
445
Results show that the Dakota Access conversation stream included the greatest degree of
446
“calls for government intervention” tweets. The results indicate that Elsipogtog conversation
447
included the greatest degree of “protest and blockade action” tweets.
448
The relationships between viral topic and each frame have statistically significant
449
differences between the three datasets, as indicated by effect size test statistics. The effect size,
450
as indicated by the Cramer’s V test statistic, for viral topic and calls for governmental
451
intervention was the highest difference between the three groups, though relatively small to
452
medium (Pallant, 2013).
453
Twitter Virality Compared with Media and Public Agendas
454
RUNNING HEAD: Disruptive social media virality
22
To examine the relationship between disruptive social media virality and traditional
455
mainstream media attention and public salience (research question one and hypothesis one), we
456
compared the Twitter volume for each issue with the volume of mainstream press mentions of
457
“Standing Rock” combined with “Dakota Access Pipeline.” We contrasted the Standing Rock
458
case with that of controversy over shale gas exploration in New Brunswick during 2013,
459
comparing tweet volume to the frequency of mainstream press mentions of “Elsipogtog”
460
combined with “shale gas.”
461
Standing Rock and the Dakota Access Pipeline. The Standing Rock dataset had the
462
highest number of tweets, with more than 13.5 million as compared to approximately 1.5 million
463
mentioning the pipeline alone. Standing Rock Twitter and mainstream media attention start to
464
increase in mid-to-late August 2016 at a similar time point. Whereas, the first bump in Google
465
search volume, as a measure of general public attention, was in early September (see Figure 3).
466
Standing Rock as a viral topic on Twitter peaked when law enforcement and company security
467
forces cracked down on demonstrators at the pipeline construction site.
468
Social media attention to #noDAPL spiked at points of heightened repression. Peaks in
469
Twitter post volume aligned with major events in the anti-pipeline protest. The highest peak was
470
the week of Nov. 20, 2016, when authorities deployed water cannons on demonstrators at night
471
in freezing temperatures (see Figure 2). In an incident Nov. 20, a reported 167 people were
472
injured, with seven taken to the hospital (Wong, 2016).
473
[INSERT FIGURE 2 ABOUT HERE.]
474
Although the largest spike in Twitter conversation about the protests occurred during the
475
violent week of November 21, 2016, the point of highest public attention to the Standing Rock
476
RUNNING HEAD: Disruptive social media virality
23
protests, as measured by Google search volume, was the week of October 28, 2016, when law
477
enforcement arrested more than 100 people. By November 1, following the spread of a rumor
478
that law enforcers were using Facebook to track demonstrators, more than a million Facebook
479
users employed the application’s “check-in” function at Standing Rock to show solidarity with
480
those on-the-ground in North Dakota (Levin & Woolf, 2016). Bystanders could learn about the
481
on-going protests were exposed to the “check-ins” of their Facebook friends and could turn to
482
Google for more information.
483
In contrast, mainstream media attention peaks occurred with major policy and legal
484
decisions (see Figure 3). The largest peak in U.S. newspaper coverage of the pipeline protests is
485
the week in December 2016 when the U.S. Army Corps of Engineers under the Obama
486
administration denied the necessary easement for Energy Transfer Partners to drill under Lake
487
Oahe, announcing that it would conduct an environmental impact statement review. Smaller
488
spikes in U.S. press coverage occurred in September when a judge temporarily stopped
489
construction at the Lake Oahe crossing until the Standing Rock Sioux Tribe’s earlier lawsuit
490
could be decided. Another peak occurred when President Trump reversed the Obama
491
administration’s decision and approved the pipeline’s completion in January 2017. The pipeline
492
protest was at its highest on the international media agenda with Trump’s January approval,
493
followed by the Obama administration delay in December 2016. The exception to spikes in
494
mainstream media attention is press coverage by the North Dakota statewide newspaper, the
495
Bismarck Tribune, which is more consistent throughout the time period we studied.
496
[INSERT FIGURE 3 ABOUT HERE.]
497
Twitter conversation mirrored events in Cannon Ball, North Dakota, with increases in
498
discussion overall, and specifically about law enforcement, spiking at points of heightened
499
RUNNING HEAD: Disruptive social media virality
24
tension, such as the week of November 21, 2016, when the U.S. holiday Thanksgiving was
500
celebrated that year, when authorities used water cannons and tear gas on demonstrators in
501
freezing temperatures, an event widely shared on social media. In our classification of Standing
502
Rock tweets, “use of force” was the highest ranked category (33%). The real-time functionalities
503
of a Twitter timeline or Facebook Live broadcast allow people far from the physical site of
504
protest to see images bypassing mainstream press coverage. The vast majority of Standing Rock
505
posts came from outside of North Dakota, with less than one percent of posts with an identifiable
506
location from the state. The conversation reached all 50 states and 214 countries and territories
507
globally, with 81% of posts from the United States.
508
In contrast, the DAPL dataset, which we expected to be neutral in tone, functioned more
509
like mainstream media attention, while public attention was similar to Standing Rock Twitter
510
conversation during peak periods. The highest spike in volume, which was also the highest point
511
of U.S. press attention, occurred in December 2016 when the Army Corps of Engineers denied
512
the easement to complete the pipeline (see Figure 2).
513
The most discussed conversation frame for DAPL was “Native/treaty rights,” as opposed
514
to “use of force.” It also included the highest share of “calls for government intervention.”
515
Similar to the Standing Rock dataset, less than one percent of the posts were from North Dakota.
516
Tweets came all 50 states, with California (18%) and New York (10%) ranking highest.
517
Increases in public attention fuel heightened participation in viral conversation, aiding
518
others to speak out who otherwise might not have done so (Nahon & Hemsley, 2013). The
519
second major peak in public attention was several weeks later during the week of November 21.
520
The data illustrate the fleeting nature of public attention, as witnessed by the much lower Google
521
search interest in early 2017 (similar in volume to “Dakota Access Pipeline” searches), when
522
RUNNING HEAD: Disruptive social media virality
25
President Trump reversed the Obama administration decision, green-lighting the pipeline’s
523
completion.
524
Another factor in the Standing Rock protests gaining virality on the public and media
525
agendas was celebrity retweets. The most highly retweeted Standing Rock posts on Twitter
526
include celebrities such as Chance the Rapper and actor Mark Ruffalo, and former presidential
527
candidate Senator Bernie Sanders.
528
Elsipogtog. The Elsipogtog dataset had the fewest Twitter posts (N=185,329), and the
529
highest proportion of “protest and blockade action” tweets, lowest frequency of “calls for
530
governmental intervention” and “water protection.” The potential for water contamination was
531
among the concerns voiced by the Elsipogtog First Nation, though demonstrators did not widely
532
share the collective identification as “water protectors.” While the issue also follows a sharp
533
acceleration curve of a spike in attention and drop-off, it did not generate widespread attention
534
outside of Canada.
535
Elsipogtog varies from the Standing Rock protests in that, while the issue was widely
536
discussed among anti-fracking activists (Hopke, 2015), it was limited in the extent to which it
537
broke into the media and public agendas. Press attention and worldwide Google search interest
538
closely mirror each other (see Figure 5) The issue went viral on Twitter as tensions escalated on-
539
the-ground.
540
[INSERT FIGURE 4 ABOUT HERE.]
541
Twitter was a much less visual medium in 2013 then in 2016, with fewer overall posts.
542
Still, images of militarised RCMP snipers, as well as the standoff with peaceful demonstrators
543
and burning police cars, were widely shared on social media in real-time. A number of the viral
544
images depict peaceful protesters, including women and children facing off with authorities.
545
RUNNING HEAD: Disruptive social media virality
26
[INSERT FIGURE 5 ABOUT HERE.]
546
Canadians, and as well as allies and bystanders internationally, rapidly saw first-hand
547
images of heavy-handed police tactics, which could bypass mainstream media coverage and
548
were circulated through online social networks. Of posts with an identifiable geographical
549
location, 59% came from Canada, followed by the United States (31%). Of tweets from Canada,
550
nearly half came from Ontario (47%), followed by British Columbia (12%). New Brunswick,
551
where the protest took place, made up 10% of posts.
552
Discussion
553
In this study, we examined the extent to which the diffusion of Standing Rock and
554
Elsipogtog movement collective action frames—along with counter-frames—on Twitter is
555
related to spikes in public and media attention to each issue. Counter to our expectations, the
556
DAPL conversation stream had the highest proportion of “calls for government intervention.”
557
This makes sense in that individuals making these calls to action would focus on the pipeline,
558
rather than the protest, in their appeals to governmental officials. An interest distinction in the
559
results is that the Standing Rock conversation stream had the highest proportion of “water
560
protection,” a major focus on demonstrators, while DAPL had the highest proportion of
561
“solidarity.” It could be that individuals tweeting with Standing Rock keywords were more tied-
562
in to the anti-pipeline movement, while those using DAPL keywords could have consistent of a
563
greater degree of bystanders, in other words those tangential to the movement. It is noteworthy
564
that the Elsipogtog conversation stream consisted of the highest degree of “protest and blockade
565
action” tweets. This could be attributable to the changing affordances of the platform between
566
the time periods of data collection.
567
RUNNING HEAD: Disruptive social media virality
27
Reminiscent of people around the country tuning-in to their television sets to see images
568
of water cannons trained on peaceful marchers during the Civil Rights Movement, in 2016 the
569
social media-enabled public could see images from Standing Rock on social media. The
570
networked social media landscape has heightened opportunities for movements to amplify their
571
collective action frames in the public sphere. There is a synergic relationship between social
572
media posting, media attention, and public interest. While we are not making causal claims, as
573
we show, there is a relationship between media and public attention, influenced by viral topics
574
on social media and protest events on-the-ground. Social media contribute to the ability to
575
connect with like-minded groups, as was the case with the Standing Rock Sioux Tribe and other
576
Native communities, as well as non-native supporters. As McAdam (1999) theorized about the
577
role of communication infrastructure during the Civil Rights Movement, social apps can serve as
578
mechanisms for the “cognitive liberation” of marginalized communities to join together and
579
break into media and public agendas.
580
We have explicated what we term the disruptive participatory uses of social media using
581
two indigenous-led protests against large-scale energy industry infrastructure as case studies.
582
Taking into account the historical background we provide earlier; political opportunities are
583
often limited for indigenous peoples. Therefore, it is even more noteworthy that the Standing
584
Rock #noDAPL protests grew into a widespread movement, achieving mainstream media and
585
public attention. In the case of Elsipogtog, the spikes in social media volume and media, as well
586
as public attention were shorter though in that case activists achieved their objective in stopping
587
shale exploration when the provincial government—along with a shift to liberal control—had
588
instituted a moratorium on hydraulic fracturing, or fracking (Passut, 2015). In the Standing Rock
589
case, anti-pipeline activists, while successful in drawing attention to their cause, were dealt a
590
RUNNING HEAD: Disruptive social media virality
28
major blow when shortly after taken office, U.S. President Donald Trump approved completion
591
of the pipeline. While disruptive social media virality can matter in amplifying social movement
592
goals and collective action frames, the political systems in which activists’ function are still
593
critically important.
594
To further explicate our theoretical model, we include a visualization (see Figure 6). We
595
consider disruptive social media virality to be internet-mediated activism taken by individuals, or
596
groups, who see themselves as outsiders to decision-making processes, including those who are
597
from historically marginalized communities. When individuals and groups make use of digital
598
and social media applications to amplify and document dissent they are engaging in disruptive
599
public participation. This concept brings together the concept of virality, as explicated by Nahon
600
and Hemsley (2013) with political sociology scholarship on social movements, mainstream
601
media and political opportunity structures (see Gamson, 2004; McAdam, 1999) fused with the
602
disruptive potential of social media (Tufekci, 2017). It is important to note that these processes
603
take place within increasingly open, fluid, what Chadwick (2017) terms hybrid media systems (p.
604
4). For historically and currently marginalized communities having their voices heard on any
605
issue, including climate change and energy development, is far from a given.
606
[Insert Figure 6 about here.]
607
Twitter affordances have evolved between our two case studies. Twitter of 2016/2017
608
was more of a visual social media platform, which could in part explain our finding of multiple
609
spikes in social media volume. We found that these were related to points of heightened tensions
610
on-the-ground during protest events, such as the use by law enforcement in North Dakota of
611
water cannons on demonstrators November 20, 2016, an event widely shared on social platforms.
612
RUNNING HEAD: Disruptive social media virality
29
Both cases garnered the support of progressive celebrities, which has been shown in past
613
scholarship to be a factor in social media diffusion (Thorson et al., 2016).
614
A limitation of this study is that Google search volume data is a measure of public
615
interest, or curiosity, not of public opinion (Gramlich, 2017). Future research should investigate
616
public opinion in relation to media coverage of the Standing Rock and Elsipogtog cases. More
617
research is needed into the evolution of such communicative tactics over time, particularly at
618
points of changing political opportunities. In the case of the Dakota Access Pipeline, activists
619
were successful, in the waning days of the Obama administration, in delaying its completion.
620
The limits of disruptive social media came into focus when President Trump green-lighted
621
DAPL’s completion after taking office in January 2017. Later in the time period we studied,
622
#noDAPL activists shifted tactics to targeting financial institutions bankrolling the project, with
623
a “Divest DAPL” campaign.
7
As Standing Rock and Elsipogtog opposition to extractive industry
624
projects shows, political opportunities are still critical for movements on climate change and
625
energy industry infrastructure projects to achieve their objectives.
626
Disruptive uses of social media can expand the political avenues open to social
627
movements. #NoDAPL and related hashtags formed a diffused collective identity, amplifying the
628
solidarity of other tribes and non-native supporters. The social media attention to Standing Rock
629
and Elsipogtog mobilized material resources to sustain the movements. Standing Rock saw
630
unprecedented solidarity, with 400 tribal flags posted at the Sacred Stone camp and an estimated
631
10,000 supporters visiting No DAPL camps. This solidarity also resulted in more material
632
7
For more on the use of fossil fuel divestment as a climate action tactic, and as a social movement, refer to the
contribution to the Oxford Encyclopedia of Climate Change Communication by Hopke and Hestres (2017).
RUNNING HEAD: Disruptive social media virality
30
resources for water protectors, with $3 million raised through GoFundMe and $2.8 million
633
contributed to a legal defense fund, as well as donations to Amazon supply wish lists.
634
The relationship between spikes in Twitter volume with press and public attention was
635
similar in the Elsipogtog case. In fact, heightened tensions over both extraction projects and the
636
resulting crackdowns by law enforcement that sparked social media attention came at points
637
nearing project deadlines. Elsipogtog peaked over a shorter period and media attention was
638
largely confined to the Canadian sphere, whereas Standing Rock and DAPL had a series of
639
spikes.
640
Climate activists are increasingly turning to focus on anti-pipeline activism to draw
641
attention to the devastating impact of climate change on indigenous peoples and their
642
environmental heritage (see Melchior Figueroa, 2011). As Dana Tizya-Tramm, tribal councilor
643
for the Vuntat Gwitchin Government First Nation, in Canada, said at the 2018 Global Climate
644
Action Summit in San Francisco, “Respect for indigenous rights is key to stemming and
645
reversing climate change. The disregard of our people is the disregard of this planet”
646
(Hernandez-Arthur, 2018).
647
RUNNING HEAD: Disruptive social media virality
31
Conflict of Interest Statement:
648
The authors declare that the research was conducted in the absence of any commercial or
649
financial relationships that could be construed as a potential conflict of interest.
650
651
Author Contributions Statement:
652
JH and MSW contributed conception and design of the study; JH and MSW coded data; JH
653
performed the statistical analysis; JH wrote the first draft of the manuscript; JH developed the
654
conceptual model; PL researched historical case background; JH, MSW and PL wrote sections of
655
the manuscript; JH revised the manuscript. All authors contributed to manuscript revision, read
656
and approved the submitted version.
657
RUNNING HEAD: Disruptive social media virality
32
References:
658
Alli, J. (2016, Sept. 12). Running for their lives: Native American relay tradition revived to
659
protest Dakota Access Pipeline. Salon. Retrieved from
660
http://www.salon.com/2016/09/12/running-for-their-lives-native-american-relay-
661
tradition-revived-to-protest-dakota-access-pipeline/
662
APTN National News. (2013, Dec. 6). SWN ending exploration work in NB, will be back in
663
2015: Elsipogtog War Chief Levi. APTN. Retrieved from
664
http://aptnnews.ca/2013/12/06/swn-ending-exploration-work-nb-back-2015-war-chief-
665
levi/
666
Benford, R.D., & Snow, D.A. (2000). Framing processes and social movements: An overview
667
and assessment. Annual Review of Sociology, 26, 611-639.
668
Bennett, W.L., & Segerberg, A. (2012). The logic of connective action: Digital media and the
669
personalization of contentious politics. Information, Communication & Society, 15(5),
670
739-768. doi: 10.1080/1369118X.2012.670661
671
Bismarck Tribune. (2016, Aug. 18). Timeline: Dakota Access pipeline. The Bismarck Tribune.
672
Retrieved from http://bismarcktribune.com/news/state-and-regional/dakota-access-
673
pipeline/collection_3ece8508-fce0-5f0f-a2b3-7092134bee07.html
674
Boykoff, M., & Boykoff, J. (2004). Balance as bias: Global warming and the U.S. prestige press.
675
Global Environmental Change, 14, 125–136
676
Briggs, F. (Reporter). (1973, February 28). NBC Evening News: American Indian Movement /
677
Wounded Knee [Television broadcast]. New York, NY: National Broadcasting Company.
678
Retrieved from https://tvnews.vanderbilt.edu/broadcasts/468929
679
Bureau of Indian Affairs. (2005). American Indian Population and Labor Force Report, US Dept.
680
of the Interior. Retrieved from
681
http://www.sdtribalrelations.com/new/tribalstatprofiles/srststatprofile2011.pdf
682
Chadwick, A. (2017). The hybrid media system: Politics and power. Oxford: Oxford University
683
Press.
684
Change.org (2016). Stop the Dakota Access Pipeline. Retrieved
685
from https://www.change.org/p/jo-ellen-darcy-stop-the-dakota-access-pipeline
686
Christensen, C. (2013). Wave-riding and hashtag-jumping: Twitter, minority ‘third parties’ and
687
the 2012 US elections. Information, Communication & Society, 16(5), 646-666. doi:
688
10.1080/1369118X.2013.783609
689
CNN (2017, Feb. 23). Dakota Access Pipeline protest site is cleared. Retrieved from
690
http://www.cnn.com/2017/02/22/us/dakota-access-pipeline-evacuation-order/index.html
691
Discover Canada. (2015). Canada’s history: Aboriginal peoples. Retrieved from
692
http://www.cic.gc.ca/english/resources/publications/discover/section-06.asp
693
RUNNING HEAD: Disruptive social media virality
33
Earl, J., McKee Hurwitz, H., Mejia Mesinas, A., Tolan, M., & Arlotti, A. (2013). The protest
694
will be tweeted. Information, Communication & Society, 16(4), 459-478.
695
doi:10.1080/1369118X.2013.777756
696
Entman, R.M. (1993). Framing: Toward clarification of a fractured paradigm. Journal of
697
Communication, 43(4), 51-58.
698
Freelon, D. (2013). ReCal OIR: Ordinal, interval, and ratio intercoder reliability as a web
699
service. International Journal of Internet Science, 8(1), 10-16.
700
Gamson, W.A. (2004). Bystanders, public opinion, and the media. In D. A. Snow, S. A. Soule,
701
and H. Kriesi (Eds.), The Blackwell companion to social movements (pp. 242-261).
702
Malden, MA: Blackwell Publishing.
703
Gamson, W.A. (1992). Talking politics. Cambridge: Cambridge University Press.
704
Gamson, W.A., & Meyer, D.S. (1996). Framing political opportunity. In D. McAdam, J.D.
705
McCarthy, & M.N. Zald (Eds.), Comparative perspectives on social movements: Political
706
opportunities, mobilizing structures, and cultural framings (pp. 275_290). New York,
707
NY: Cambridge University Press.
708
Gamson, W.A., & Modigliani, A. (1989). Media discourse and public opinion on nuclear power:
709
A constructionist approach. The American Journal of Sociology, 95(1), 1-37.
710
Gilbert, K. (2016, Nov. 20). Live by the blockade on HWY 1806 just north of Oceti Sakowin
711
camp. Facebook Live. Retrieved from
712
https://www.facebook.com/kevin.happychappy/videos/1806777989594705/
713
Gitlin, T. (1980). The whole word is watching. Berkeley: University of California Press.
714
Goffman, E. (1974). Frame analysis: An essay on the organization of experience. New York,
715
NY: Harper & Row.
716
Gramlich, J. (2017, April 27). Q&A: Using Google search data to study public interest in the
717
Flint water crisis. Pew Research Center. Retrieved from
718
http://www.pewresearch.org/fact-tank/2017/04/27/flint-water-crisis-study-qa/
719
Hayes, A.F. (2005). Statistical methods for communication science. London: Lawrence Erlbaum
720
Associates.
721
Hernandez-Arthur, S.. [SimonHernandez]. (2018, September 13). Back to #GCAS2018:
722
"Respect for indigenous rights is key to stemming and reversing climate change. The
723
disregard of our people is the disregard of this planet," says Dana Tizya-Tramm. [Tweet].
724
Retrieved from https://twitter.com/SimonHernandez/status/1040282153052004353
725
Himelboim, I., Hansen, D., & Bowser, A. (2013). Playing in the same Twitter network: Political
726
information seeking in the 2010 US gubernatorial elections. Information, Communication
727
& Society, 16(9), 1373-1396. doi: 10.1080/1369118X.2012.706316.
728
Hopke, J. E. (2015). Hashtagging politics: transnational anti-fracking movement Twitter
729
practices. Social Media + Society, 1(2), doi: 2056305115605521.
730
RUNNING HEAD: Disruptive social media virality
34
Hopke, J. E., & Hestres, L. E. (2017). Communicating about fossil fuel divestment. Oxford
731
Encyclopedia of Climate Science. Retrieved from
732
http://climatescience.oxfordre.com/abstract/10.1093/acrefore/9780190228620.001.0001/a
733
crefore-9780190228620-e-566?rskey=QuU2Tv&result=2
734
Hopke, J. E., & Simis, M. (2017). Discourse over a contested technology on Twitter: A case
735
study of hydraulic fracturing. Public Understanding of Science, 26(1), 105-120.
736
Hopke, J. E., Simis-Wilkinson, M., Loew, P. A. (2018). Social media in agenda-setting: The
737
Elsipogtog First Nation and Standing Rock Sioux Tribe. In A. Gruzd et al. (Eds.),
738
SMSociety '18 Proceedings of the 9th International Conference on Social Media and
739
Society (pp. 310-314). Copenhagen, Denmark — ACM New York, NY, USA ISBN: 978-
740
1-4503-6334-1 doi>10.1145/3217804.3217935.
741
Hopkins, D.J., & King, G. (2010). A method of automated nonparametric content analysis for
742
social science. American Journal of Political Science, 54(1), 229-247. doi:
743
10.1111/j.1540-5907.2009.00428.x
744
Howe, M. (2015). Debriefing Elsipogtog: The anatomy of a struggle. Halifax; Winnipeg:
745
Fernwood Publishing.
746
Indigenous and Northern Affairs Canada. (2017, May). Registered population: Elsipogtog First
747
Nation. Retrieved from http://fnp-ppn.aadnc-aandc.gc.ca/fnp/
748
Johnston, E. (2017, Nov. 24). I shut down an oil pipeline – because climate change is a ticking
749
bomb. Guardian. Retrieved from
750
https://www.theguardian.com/commentisfree/2017/nov/24/oil-pipeline-valve-turner-
751
protest-climate-change
752
Kelley, G. (2014, January 23). The Elsipogtog crisis in wake of “Idle No More” [web log
753
comment]. Retrieved from http://piloteconseil.com/en/blogue/the-elsipogtog-crisis-in-
754
wake-of-idle-no-more/
755
Koopmans, R. (2004). Protest in time and space: The evolution of waves of contention. In D. A.
756
Snow, S. A. Soule, and H. Kriesi (Eds.), The Blackwell companion to social movements
757
(pp. 19-46). Malden, MA: Blackwell Publishing.
758
Levin, S. & Woolf, N. (2016, Oct. 31). A million people 'check in' at Standing Rock on
759
Facebook to support Dakota pipeline protesters. Guardian. Retrieved from
760
https://www.theguardian.com/us-news/2016/oct/31/north-dakota-access-pipeline-protest-
761
mass-facebook-check-in
762
Manjoo, F. (2015, June 11). For Twitter, future means here and now. New York Times. Retrieved
763
March 22, 2018, from https://nyti.ms/2xWBuWC
764
McAdam, D. (1999). Political process and the development of Black insurgency, 1930-1970
765
(second edition). Chicago: University of Chicago Press.
766
RUNNING HEAD: Disruptive social media virality
35
McCay-Peet, L. & Quan-Haase, A. (2017). What is social media and what questions can social
767
media research help us answer? In L. Sloan and A. Quan-Haase, The SAGE handbook of
768
social media research methods (pp. 13-26). London: SAGE.
769
McKibben, B. (2016, Nov. 29). Standing Rock is the civil rights issue of our time – let's act
770
accordingly. Guardian. Retrieved from https://www.theguardian.com/us-
771
news/commentisfree/2016/nov/29/standing-rock-civil-rights-act-accordingly
772
Melchior Figueroa, R. (2011). Indigenous peoples and cultural losses. In J. S. Dryzek, R. B.
773
Norgaard, & D. Schlosberg (Eds.), The Oxford Handbook of Climate Change and Society
774
(pp. 232-247). Oxford: Oxford University Press.
775
Meraz, S., & Papacharissi, Z. (2013). Networked gatekeeping and networked framing on #Egypt.
776
The International Journal of Press/Politics, 18, 138–166. doi:
777
10.1177/1940161212474472
778
Meyer, R. (2016, September 9). The legal case for blocking the Dakota Access Pipeline. The
779
Atlantic. Retrieved from https://www.theatlantic.com/technology/archive/2016/09/dapl-
780
dakota-sitting-rock-sioux/499178/
781
Miller, R. (Reporter). (1973, February 28). ABC Evening News: American Indian Movement /
782
Wounded Knee [Television broadcast]. New York, NY: American Broadcasting
783
Company. Retrieved from https://tvnews.vanderbilt.edu/broadcasts/24701
784
Nahon, K. & Hemsley, J. (2013). Going viral. Cambridge: Polity Press.
785
Pallant, J. (2013). SPSS survival manual: A step-by-step guide to data analysis using IBM SPSS
786
(5th edition). Maidenhead, UK: Open University Press.
787
Papacharissi, Z., & de Fatima Oliveira, M. (2012). Affective news and networked publics: The
788
rhythms of news storytelling on #Egypt. Journal of Communication, 62, 266–282.
789
doi: 10.1111/j.1460-2466.2012.01630.x
790
Passut, C. (2015, March 27). New Brunswick, Ontario take divergent paths on fracking. NGI’s
791
Shale Daily. Retrieved from http://www.naturalgasintel.com/articles/101806-new-
792
brunswick-ontario-take-divergent-paths-on-fracking
793
Pew Research Center, Journalism and Media Staff. (2012). Winning the Media Campaign 2012:
794
Methodology. Retrieved from http://www.journalism.org/2012/11/01/methodology-26/
795
Pew Research Center. (2015a). The Evolving Role of News on Twitter and Facebook. Retrieved
796
from http://www.journalism.org/files/2015/07/Twitter-and-News-Survey-Report-
797
FINAL2.pdf
798
Pew Research Center. (2015b). How do Americans use Twitter for news? Retrieved from
799
http://www.pewresearch.org/fact-tank/2015/08/19/how-do-americans-use-twitter-for-
800
news/
801
Piven, P. F. & Cloward, R. A. (1979). Poor people’s movements. New York: Vintage Books.
802
RUNNING HEAD: Disruptive social media virality
36
Schmidt, A., Ivanova, A., & Schäfer, M. S. (2013). Media attention for climate change around
803
the world: A comparative analysis of newspaper coverage in 27 countries. Global
804
Environmental Change, 23, 1233–1248. doi:10.1016/j.gloenvcha.2013.07.020
805
Schwartz, D. & Gollum, M. (2013, October 19). N.B. fracking protests and the fight for
806
aboriginal rights. CBC News. Retrieved from https://www.cbc.ca/news/canada/n-b-
807
fracking-protests-and-the-fight-for-aboriginal-rights-1.2126515
808
Slater, M. D. & Gleason, L. S. (2012). Contributing to theory and knowledge in quantitative
809
communication science. Communication Methods and Measures, 6, 215-236.
810
Snow, D. A. (2004). Framing processes, ideology, and discursive fields. In D. A. Snow, S. A.
811
Soule, and H. Kriesi (Eds.), The Blackwell companion to social movements (pp. 380-
812
412). Malden, MA: Blackwell Publishing.
813
Sterling, G. (2017, Feb. 9). Data: Google monthly search volume dwarfs rivals because of
814
mobile advantage. Search Engine Land. Retrieved from
815
http://searchengineland.com/data-google-monthly-search-volume-dwarfs-rivals-mobile-
816
advantage-269120
817
Stocking, G. & Matsa, K.E. (2017, April 27). Using Google Trends data for research? Here are 6
818
questions to ask. Medium. Retrieved from https://medium.com/@pewresearch/using-
819
google-trends-data-for-research-here-are-6-questions-to-ask-a7097f5fb526
820
Su, L.Y.-F., Cacciatore, M.A., Liang, X., Brossard, D., Scheufele, D.A., & Xenos, M.A. (2017).
821
Analyzing public sentiments online: Combining human- and computer-based content
822
analysis. Information, Communication & Society, 20(3), 406-427.
823
doi: 10.1080/1369118X.2016.1182197
824
Tarrow, S. (1994). Power in movement: Social movements, collective action and politics.
825
Cambridge: Cambridge University Press.
826
Thorson, K., Edgerly, S., Kligler-Vilenchik, N., Xu, Y., & Wang, L. (2016). Seeking Visibility
827
in a Big Tent: Digital Communication and the People’s Climate March. International
828
Journal Of Communication, 10, 23. Retrieved
829
from http://ijoc.org/index.php/ijoc/article/view/4703/1802
830
Tufekci, Z. (2017). Twitter and tear gas: The power and fragility of networked protest. New
831
Haven: Yale University Press.
832
Vowel, C. (2013, Nov. 14). The often-ignored facts about Elsipogtog. Toronto Star. Retrieved
833
from
834
http://www.thestar.com/opinion/commentary/2013/11/14/the_oftenignored_facts_about_
835
elsipogtog.html
836
West Film Project. (2001). Lakota accounts of the massacre at Wounded Knee. Retrieved from
837
http://www.pbs.org/weta/thewest/resources/archives/eight/wklakota.htm
838
Wong, J.C. (2016, Nov. 21). Standing Rock protest: hundreds clash with police over Dakota
839
Access Pipeline. Guardian. Retrieved from https://www.theguardian.com/us-
840
news/2016/nov/21/standing-rock-protest-hundreds-clash-with-police-over-dakota-access-
841
pipeline
842
RUNNING HEAD: Disruptive social media virality
37
Woodley, S. (2016, Oct. 10). The riot police are arriving. Facebook Live. Retrieved from
843
https://www.facebook.com/ShaileneWoodley/videos/624178791076838/
844
RUNNING HEAD: Disruptive social media virality
38
Table 1: Social movement collective action frames and counter-frames.
845
Collective Action Frames
Counter-Frames
Water protection
Protest and blockade action
Solidarity
Against the movement
Native/treaty rights
Calls for government intervention
Use of force
Other
Hashtag jumping
846
847
Table 2: Conversational frames by viral topic.
848
Frame
Viral Topica
Chi-square Resultsc
Standing
Rock
DAPL
Elsipogtog
Collective Action Frames
Protest/blockade
action
14%
10%
35%
χ2(df=2, N=15,354,200)=83,614.93,
p=.000. Cramer’s V=.074, p=.000.
Water
protection
11%
9%
7%
χ2=9,690.67, p=.000.
Cramer’s V=.025, p=.000.
Native/treaty
rights
13%
25%
19%
χ2=162,743.85, p=.000.
Cramer’s V=.103, p=.000.
Solidarity
13%
16%
19%
χ2=14,710.95, p=.000.
Cramer’s V=.031, p=.000.
Calls for gov’t
intervention
5%
14%
<1%
χ2=261,724.78, p=.000.
Cramer’s V=.131, p=.000.
Mainstream, Counter Frames
Use of force
33%
19%
20%
χ2=139,139.04, p=.000.
Cramer’s V=.095, p=.000.
Against the
movement
5%
6%
--
χ2(df=1, N= 15,168,871)=5,005.06,
p=.000. Phi=.018, p=.000.
Other
Hashtag
jumping
6%
--
--
--
Total
13,596,975
1,571,896
185,329b
a The proportions included in the table are the percentages within each dataset by category. The category with the
849
highest proportion for each group is bolded.
850
b For Elsipogtog, the analyses exclude two not applicable posts.
851
c The chi-square analyses with all three groups have the same degrees of freedom and sample size. For hashtag
852
jumping, the analysis is a 2x2 table. The Yates’ Correction for Continuity chi-square statistic and phi coefficient are
853
reported.
854
855
RUNNING HEAD: Disruptive social media virality
39
Figure 1: The proportions of conversational themes by viral topic.
856
857
RUNNING HEAD: Disruptive social media virality
40
Figure 2: Standing Rock and Dakota Access Pipeline tweet volume, April 1, 2016 to March
858
31, 2017.
859
860
RUNNING HEAD: Disruptive social media virality
41
Figure 3: Standing Rock and DAPL mainstream media and public attention, April 2016 to
861
March 2017.*
862
863
*Mainstream media attention is measured by the frequency of Lexis Nexis mentions of “Standing Rock” and
864
“Dakota Access Pipeline.” Public attention is measured by the worldwide Google search volume through Google
865
Trends. Search volume values are normalized where the highest point is 100.
866
RUNNING HEAD: Disruptive social media virality
42
Figure 4: Elsipogtog tweet volume, January 1 to December 31, 2013.
867
868
RUNNING HEAD: Disruptive social media virality
43
Figure 5: Elsipogtog mainstream media and public attention, Jan. to Dec. 2013.*
869
870
*Mainstream media attention is measured by the frequency of Lexis Nexis mentions of “Elsipogtog” and “shale
871
gas.” Public attention is measured by the worldwide Google search volume through Google Trends. Search volume
872
values are normalized where the highest point is 100.
873
RUNNING HEAD: Disruptive social media virality
44
Figure 6: Proposed model of disruptive social media virality processes.*
874
875
*This model expands on work presented at the 9th International Conference on Social Media and Society (Hopke,
876
Simis-Wilkinson & Loew, 2018).
877
... reach new audiences). Second, we use a theoretical framework for understanding the utility of social media for environmental activists, disruptive social media virality, to examine two case studies of environmental activists using social media to amplify their demands in the public sphere (Hopke, Simis-Wilkinson, & Loew, 2018a;2018b). ...
... A major area of research within social media studies is looking at factors which contribute to individual social posts, as well as issues, going viral on social platforms and correspondingly their ability, or not, to break through into traditional media (see Nahon & Hemsley, 2013). In this chapter, we take a theoretical perspective, proposed elsewhere by the first author and colleagues, termed disruptive social media virality, which enables scholars to explore factors that underlie virality among environmental protest movements and causes (Hopke et al., 2018a, Hopke et al., 2018b. Applied to environmental action and social movement organizing, disruptive social media virality consists of the potential for marginalized, or otherwise underrepresented populations, such as Indigenous communities, youth, and others, to turn to digital and social media apps and platforms to connect with like-minded individuals and collectives. ...
... Applied to environmental action and social movement organizing, disruptive social media virality consists of the potential for marginalized, or otherwise underrepresented populations, such as Indigenous communities, youth, and others, to turn to digital and social media apps and platforms to connect with like-minded individuals and collectives. As Hopke et al. (2018b) write, the concept involves: … stakeholders perceiving themselves to be outsiders in the decision-making processes, or who are from historically marginalized groups, who make use of social media to amplify and document dissent when traditional modes of public participation have been exhausted or are perceived to be ineffective for achieving their goals. ...
Chapter
Full-text available
Environmental movements are increasingly turning to social media apps and platforms, along with other digital organizing tools, to reach bystanders and supporters alike. We overview literature on how environmental activists are using social media for both internal movement functions (e.g. coordinate a transnational day of action) and external ones (e.g. reach new audiences). Second, we expand on a theoretical framework for understanding the utility of social media for environmental activists, disruptive social media virality. Disruptive social media virality consists of a set of stakeholders who consider themselves to be outsiders to traditional environmental and political decision-making systems who turn to social media for organizing when conventional public participation methods are perceived to be ineffective to reach their collective goals, and to amplify social movement activities and objectives. We illustrate this framework through two case studies: the 2016 #NoDAPL protests which brought the concerns of the Standing Rock Sioux Tribe in North Dakota in their fight against the Dakota Access Pipeline (DAPL) worldwide attention and the #FridaysForFuture school strike for climate youth movement. In the concluding section of the chapter, we provide methodological reflections for best practices for studying environmental movements using social media.
Article
Full-text available
Recent technological developments have created novel opportunities for analyzing and identifying patterns in large volumes of digital content. However, many content analysis tools require researchers to choose between the validity of human-based coding and the ability to analyze large volumes of content through computer-based techniques. This study argues for the use of supervised content analysis tools that capitalize on the strengths of human- and computer-based coding for assessing opinion expression. We begin by outlining the key methodological issues surrounding content analysis as performed by human coders and existing computational algorithms. After reviewing the most popular analytic approaches, we introduce an alternative, hybrid method that is aimed at improving reliability, validity, and efficiency when analyzing social media content. To demonstrate the usefulness of this method, we track nuclear energy- and nanotechnology-related opinion expression on Twitter surrounding the Fukushima Daiichi accident to examine the extent to which the volume and tone of tweets shift in directions consistent with the expected external influence of the event. Our analysis revealed substantial shifts in both the volume and tone of nuclear power-related tweets that were consistent with our expectations following the disaster event. Conversely, there was decidedly more stability in the volume and tone of tweets for our comparison issue. These analyses provide an empirical demonstration of how the presented hybrid method can analyze defined communication sentiment and topics from large-scale social media data sets. The implications for communication scholars are discussed.
Article
Full-text available
I examine a 2-week window into an environmental movement trying to gain traction in the public sphere, centered on a transnational day of action calling for a ban on the drilling technology, high-volume hydraulic fracturing, the Global Frackdown. Twitter serves a different purpose for the anti-fracking Global Frackdown movement than other Internet-based communications, most notably email listservs. Findings show that Global Frackdown tweeters engage in framing practices of movement convergence and solidarity, declarative and targeted engagement, prefabricated messaging, and multilingual tweeting. In contrast to Global Frackdown tweeters’ use of the platform for in-the-moment communication, Global Frackdown activists report in in-depth interviews that they place more emphasis on private (i.e., listservs) communication channels for longer term, durable movement building. The episodic, crowdsourced, and often personalized, transnational framing practices of Global Frackdown tweeters support core organizers’ goal of promoting the globalness of activism to ban fracking. This research extends past scholarship on socially mediated activism by providing a case study of how environmental activists use Twitter for ephemeral movement communication during a pre-planned transnational day of action, blurring internal movement collective identity-building and affirmation with publicly enacted strategic framing.
Article
The September 2014 “People’s Climate March” was reportedly the largest climate change mobilization in history. The coalition of organizations behind the march chose a strategy of inclusion: They sought to create a “big tent” for a climate movement. Building on theoretical developments in the literature on digital media and social movements, we used Twitter as a window to observe how march organizers and participants attempted to (a) create a digital space of shared attention intersecting with the on-the-ground event, and also (b) thread together diverse orientations to the climate issue.
Article
Statistical Methods for Communication Science is the only statistical methods volume currently available that focuses exclusively on statistics in communication research. Writing in a straightforward, personal style, author Andrew F. Hayes offers this accessible and thorough introduction to statistical methods, starting with the fundamentals of measurement and moving on to discuss such key topics as sampling procedures, probability, reliability, hypothesis testing, simple correlation and regression, and analyses of variance and covariance. Hayes takes readers through each topic with clear explanations and illustrations. He provides a multitude of examples, all set in the context of communication research, thus engaging readers directly and helping them to see the relevance and importance of statistics to the field of communication. Highlights of this text include: * thorough and balanced coverage of topics; * integration of classical methods with modern "resampling" approaches to inference; * consideration of practical, "real world" issues; * numerous examples and applications, all drawn from communication research; * up-to-date information, with examples justifying use of various techniques; and * a CD with macros, data sets, figures, and additional materials. This unique book can be used as a stand-alone classroom text, a supplement to traditional research methods texts, or a useful reference manual. It will be invaluable to students, faculty, researchers, and practitioners in communication, and it will serve to advance the understanding and use of statistical methods throughout the discipline. © 2005 by Lawrence Erlbaum Associates, Inc. All rights reserved.
Article
This article introduces ReCal OIR, an extension to the existing ReCal suite of online intercoder reliability modules that accommodates ordinal, interval, and ratio levels of measurement. It includes a discussion of the currently available options for calculating reliability for the ordinal, interval, and ratio levels; explains what ReCal OIR does; validates its calculations; and discusses its usage data. The process of validating its output reveals ReCal OIR to be slightly more accurate than one of its major competitors.
Article
This article discusses a variety of experiences and philosophical reflections on cultural loss under an environmental justice framework, wherein 'environmental justice' is broadly construed as the conceptual connections, causal relationships, and strong correlations that exist between environmental issues and social justice. Environmental justice frames social issues as environmental issues. Social and environmental issues are inseparable, co-causally related, and always in a context that requires a political interpretation; in particular, such a consideration of justice accounts for power dynamics and socio-environmental practices that maintain historical relations, as well as the remedies for injustices. The primary argument of this article is that the environmental justice framework is a proper theoretical and practical approach to understanding the cultural loss among indigenous peoples caused by climate change. It emphasizes restorative justice philosophies and procedures that can address the future consequences of cultural loss.