Sickness or Silence: Social Movement Adaptation to COVID-19
How have activists responded to the COVID-19 pandemic? While there have been many anecdotal
reports of the pandemic's impact, there has been little to no cross-national comparative research
examining how movements discouraged from protesting on the streets because of the risk of infection
have or have not continued their activities through the COVID-19 pandemic. In this paper we present
findings from a survey of 550 activists in 27 countries, reporting on how the pandemic has affected their
perceptions of tactical adaptation, public interest, and long-term strategic planning. We also present
results from a survey experiment testing the impact of COVID-19 risk and pandemic lockdown policies
on activists’ willingness to join a street protest. We find that while the pandemic has posed significant
challenges for activists, activists believe they have been able to respond with tactical adaptation and
innovations, primarily with a shift to digital activism. Most activists also perceived an increase in public
interest for their movements across various issue areas and were optimistic about their movement’s
ability to advance its goals in the future. These findings speak to the long-term impact of COVID-19 on
the potential for social mobilization and the short and projected long-term effects of the pandemic on
A wave of nonviolent movements emerged in 2019, as activists took to the streets en masse in
places such as Chile, Iraq, Hong Kong, Ecuador, Iran, and Lebanon to express growing discontent with
their governments and demand greater democracy, economic equality, and social justice. In Sudan and
Algeria, movements overthrew longtime presidents whose regimes were characterized by corruption
and repression. This rise in nonviolent uprisings has been a striking global trend over the last decade
(Chenoweth 2020). People dissatisfied with the status quo are more commonly using peaceful but extra-
institutional methods to pursue social, economic, and political change.
This upward trend in nonviolent resistance, as with every other recent global trend, has been
transformed by the novel coronavirus. As COVID-19 grew into a global pandemic, movements that
United States Institute of Peace.
United States Institute of Peace.
heavily relied on street protests and other tactics requiring mass turnout struggled to respond (Pinckney
and Rivers 2020). April 2020 saw a more than 60 percent drop in public protests according to the ACLED
data project (Raleigh et al. 2010). At the same time, activism around the coronavirus spiked in some
places as health care workers and ordinary citizens demanded better government responses amid
shortages of personal protective equipment and rising death tolls.
Social movements have helped citizens build power and usher in major societal change, paving the
way for more democratic and peaceful societies (Chenoweth and Stephan 2011; Celestino and Gleditsch
2013; Bethke and Pinckney 2019). Successful movements face numerous difficulties, including
frequently confronting hostile governments that seek to quash dissent (Davenport et al 2015; Kurtz et al
2018; Johnston 2011). COVID-19 and the risk of exposure bearing deadly consequences further
complicates these challenges and adds unique hurdles. The pandemic carries the potential to blunt
activism as movements struggle to adapt tactically, rely more heavily on digital activism and organizing,
and face increased government repression.
However, despite these challenges, movements are already adapting to their new operating
environments amid COVID-19. Activists have broadened their tactical repertoire, finding innovative ways
to take action that do not involve mass gatherings. Movements have also shifted to a virtual workspace,
launching online campaigns and engaging in digital organizing.
While there has been extensive media speculation about the pandemic’s impact on social
movements, for instance arguing that public health restrictions will stop momentum for change, or that
a shift to online activism may put movements at risk of increased government repression (Pinckney
2020), to date there has been little systematic research examining COVID-19’s impacts on a variety of
movements across multiple contexts. This paper is one of the first attempts to do so, using a cross-
national survey of activists asking about a variety of outcomes and testing the impact of COVID-19
deaths and public health restrictions on willingness to join public protest with a survey experiment.
The paper proceeds as follows. First, we discuss the literature on participation in social
movements and provide a brief overview of the sparse existing literature on pandemics and movements,
and media-based speculation on the specific impact of COVID-19. We then describe our survey design
and present results both from general questions and from the survey experiment. The final section
concludes with discussion on the implications of the research for the study of social movements during
Individual Micro-Foundations of Activism in COVID-19
How might COVID-19 affect the dynamics of activists’ participation in social movements? We
emphasize three micro-foundational dynamics highly likely to be impacted by the pandemic. The first,
drawing on a rational choice framework (Olson 1965), is the additional personal risk of participation due
to the danger of infection. While not all activism involves increased risk of infection, as we detail below,
many of the most prominent activist tactics involve gathering in large groups, for instance in public
marches or demonstrations. While this is typically a low-risk activity in most democratic countries that
protect freedom of expression, the danger of infection may shift it into “high-risk activism” (McAdam
1986), with the attendant consequences for the number and profile of activists willing to participate.
Yet the direct risk of infection is not the only relevant impact on dynamics of social movement
participation. While COVID-19 may increase the risks of personal participation, it may also increase
biographical availability (McAdam 1986) to participate, as potential movement participants may lose
employment, or even simply other opportunities for personal recreation due to lockdown policies.
Family responsibilities may have also been altered as social distancing and lockdown measures resulted
in people spending more time at home with their spouses, children, and other relatives.
In addition to these incentives based more in a rational choice framework, we also see strong
reasons to believe that the pandemic may affect psychological propensities to participate in activism
through quotidian disruption. As Snow and his co-authors (1998) argue, the actual or threat of a
disruption in the quotidian increases the prospects of collective action. The major social and economic
changes brought by COVID-19 have created the conditions Snow et al (1998) identify as causing
movements to emerge, throwing normal routines into doubt and opening psychological space to the
possibilities for more radical collective action.
Social Movements and the COVID-19 Pandemic
Since the start of the COVID-19 pandemic, there has been little attention paid to how social
movements have been impacted, despite the virus following a year defined by mass movements. One
initial study from Metternich (2020) uses data from the Integrated Crisis Early Warning System and
found a global decline in protest activity, with Europe and Asia having the largest drops. Other evidence
of COVID-19’s impact on movements is based largely on journalistic and anecdotal accounts that focus
similarly to these studies on protest activity amid social distancing measures and tactical adaption of
Despite initial reporting that movements from Venezuela to Iraq had abandoned their use of
street protests, several movements continued to take to the streets. Protesters from France’s “Yellow
Vest” movement held demonstrations across Paris despite a decree banning non-essential gatherings
(RT 2020). In Lebanon, the ongoing movement against government corruption defied lockdown orders
and continued its streets protests (AFP 2020). And the United States witnessed the emergence of a new
movement, which may be its largest ever, following the police killing of George Floyd in Minneapolis
despite soaring coronavirus cases. Surveys indicate that as many as 15 to 26 million people may have
protested in the U.S. during the summer of 2020 (Buchanan, Bui, and Patel 2020). The mass street
protests quickly escalated into a global movement against police brutality and anti-Black racism.
The pandemic has also sparked several movements focused directly on demanding improved
responses to the coronavirus from their governments. In Nepal, hundreds have participated in protests
calling for increased transparency and accountability in the spending of government pandemic response
funds (Chaudhury 2020). In Bogota, residents banged pots and pans from their windows in "cacerolazo"
(casserole dish) protests after the president revoked local authorities’ measures to combat the
coronavirus (Alsema 2020). And doctors and nurses in Malawi held a sit-in calling on the government to
hire more medical staff and supply equipment (Pensulo 2020).
Movements have been innovating and adapting their tactics to social distancing and lockdowns,
moving beyond public protests to advance their demands (Chenoweth et al. 2020). In Jerusalem,
thousands joined a socially distanced protest by standing on designated marks at least six feet apart and
wore masks while protesting against Prime Minister Benjamin Netanyahu (Serhan 2020). Women
activists in Poland defied a lockdown order and used their cars to block roads in the country’s capital as
they protested legislation that would tighten abortion laws (Dettmer 2020). The movement in Lebanon
also switched to using cars to protest, honking their horns and waving the national flag out of their
windows (Dettmer 2020).
Several studies of nonviolent action and social movements show the importance of tactical
adaptation and diversification. Movements have a range of tactics at their disposal: methods of protest
and persuasion, methods of intervention, and methods of noncooperation, such as boycotts and strikes
(Sharp 1973). While protests drawing mass participation are a contentious act that can be useful for
social movements to attract public attention and create a sense of urgency, an overreliance on any one
tactic may lead to the tactic becoming less effective (Tarrow 1989). Kurt Schock (2005) found that
movements that used diverse tactics were more effective than movements focused solely on protests.
He also found that tactical diversification makes movements less vulnerable to repression. This can be
important for movements operating in a COVID-19 environment, as authoritarian governments have
used the coronavirus as a pretext to punish dissent and curtail democratic freedoms (Pinckney and
Among the major tactical adaptations movements have made during COVID-19 has been a shift
to digital activism. In Chile, artists who took part in the 2019 mass protests created virtual murals on
social media depicting anti-government messages (McGowan 2020). Hong Kong activists used video
games to spread pro-democracy messages and stayed connected via Telegram, Facebook, and other
instant messaging platforms (Cellan-Jones 2020). Members of the global climate movement abandoned
public street protests for online campaigns to raise awareness and demand cuts in greenhouse gas
emissions (Lawal 2020).
Digital and social media have proven to be key means of communication for activists and social
movements (Barberá et al. 2015). Social media has made it easier “for the powerless to collaborate,
coordinate, and give voice to their concerns” (Gladwell 2010). ICTs are beneficial for collective action
because they help spread messages rapidly and reduce costs of participation. They can help movements
overcome spatial limitations, allowing geographically dispersed actors to participate together in online
campaigns (Garrett and Edwards 2007) and the creation of transnational identities for more widespread
impact (Pudrovska and Ferree 2004).
But digital activism also comes with many downsides. Bennett and Segerberg (2013) assert that
digital communication networks have contributed to a new “logic of connective action,” in which an
individual’s willingness to engage in political action is based almost solely on self-motivation. Without
the accountability structures of formal membership, it can be harder for movements to overcome
collective action problems. The ease of online participation can also lead to an overreliance on what
Jennifer Earl and her co-authors (2015) describe as “ephemeral” forms of participant engagement that
do not require long-term or sustained commitments from participants. Combined with media attention,
these actions may be effective at getting widespread attention and showing broad support for an issue,
they may be less likely than offline actions such as mass protests and boycotts to put pressure on
authorities and hold governments accountable (Tufekci 2017; Chenoweth 2020).
This existing research on the impact of COVID-19 led us to several research questions. Given the
nascent state of this literature, and a lack of strong theoretical priors, we considered most questions
appropriate solely for exploratory analysis rather than formal hypothesis testing. The main areas we
sought to examine through exploratory analysis were the following:
• How do activists perceive the impacts of the COVID-19 pandemic on their movements
• How do activists perceive the effects of the pandemic on public interest in their movements?
• What are activists’ perceptions of the likely long-term effects of the pandemic on social
The one area where we had strong theoretical priors was on the impact of the pandemic on
participation in public protests. The initial months of the pandemic saw a massive decline in the number
and size of public demonstrations, as we described above. While some countries, notably the United
States, have seen a resurgence of public demonstrations since then, we strongly suspected that the
pandemic continues to impact activists’ willingness to join large public gatherings. While early evidence
suggests that the impact of public demonstrations on actual COVID-19 infection risk may not be as
extreme as initially feared (Dave et al. 2020), we strongly suspect that continued public health
messaging regarding the riskiness of engaging in activities where social distancing may be impractical or
impossible will continue to make participants hesitant to join public demonstrations.
Following Klandermans and Oegema’s typology of social movement participation (1987) we focus on
the decision by social movement participants who already support a movement’s goals and have
received a mobilization attempt to choose to actively participate in movement activities. When such
conditions are already in place, the decision to participate can largely be reduced to personal motivation
to participate and the barriers that must be overcome to participate. As we describe above, the COVID-
19 pandemic’s most obvious impact is on those motivating steps and final barriers to participation. If the
potential protester fears for their own health or believes that their participation may put others at risk,
the personal perceived costs of protest are likely to prevent participation.
However, as we also describe above, the direct impact on public health is not the only potentially
relevant impact of the pandemic on the decision by activists to participate in demonstrations. Lockdown
policies intended to curb the impact of the pandemic have significantly reduced other options for
entertainment and recreation available to the population at large, including activists. Thus, lockdowns
are likely to decrease the opportunity cost of joining a public demonstration, particularly in situations
where a lockdown is in place but the public health risk is perceived as being relatively low. Some
scholars have speculated that this decreased opportunity cost will increase the likelihood of
participation in protests (Buchanan, Bui, and Patel 2020).
Given these expectations, we formulated the following hypotheses about the impact of the
pandemic on public demonstrations:
Hypothesis 1: Increased public health risk from COVID-19 will reduce activists’ likelihood of
participating in public demonstrations.
Hypothesis 2: Lockdown policies to curb the spread of COVID-19 that eliminate potential options for
recreation will increase activists’ likelihood of participating in public demonstrations.
While we examined our exploratory questions through open survey questions, we tested these two
hypotheses through a multifactorial vignette experiment, following a growing literature using survey
experiments in the study of nonviolent action (RezaeeDaryakenari and Asadzade 2020; Dahlum,
Pinckney, and Wig 2020). We describe both our general exploratory analysis and survey experiment in
the following section.
To gain broad insight into the impact of COVID-19 on social movements we deployed a survey to
a population of self-identified activists who otherwise varied significantly in terms of demographics,
specific mobilization issues, and national context. We recruited a convenience sample of activists using
the Prolific online survey-taking platform (www.prolific.co). Prolific is a company that recruits research
subjects to participate in online studies. It is a relatively young company but has been shown to be a
more accurate alternative to popular online research sites such as Amazon Mechanical Turk, with higher
response rates and lower failure of attention checks (Peer et al. 2017; Palan and Schitter 2018). Over
100,000 people from dozens of countries participate in the Prolific online panel.
We identified activists using a short screening survey, which was made available to Prolific
participants in June of 2020. The screening survey was open to all Prolific participants who were fluent
in English and who were not residents of the United States,
and included seven demographic questions
and an attention check.
The key demographic question asked participants if they were currently an
activist or a member of a social movement. We did not ask participants how long they had been an
activist, and thus our sample may include both those who began their activism during the pandemic and
those who have been active since before the pandemic.
We collected 7,000 responses to the screening survey. 1,500 respondents answered yes to the
activist question. Participants who answered yes, and who did not fail the attention check, were then
eligible to take our full survey. We collected a total of 573 responses to the full survey out of the total
1,500 eligible respondents. 23 responses to the follow-up survey failed an attention check, and one
respondent left the answers to nearly all questions blank. Excluding these left us with a final sample of
Our research instrument was a thirty-seven-question online survey. Six questions asked about
characteristics of the respondent or the respondent’s movement. Twenty-nine questions dealt with
different aspects of the broad impact areas we were interested in examining through exploratory
analysis to gauge activists’ perceptions on: how challenging movements are finding the COVID-19
environment, specific tactical shifts, public interest, government repression, collaboration between
organizations, and the challenges and opportunities of a move to online activism.
The final two questions were the dependent variables for the survey experiment testing our two
hypotheses on the impact of public health risk and COVID-19 lockdown policies on activists’ support of
members of their movement participating in public demonstrations and the likelihood of their own
participation. The questions followed a short vignette describing plans for a large public protest being
organized by the respondent’s movement. Respondents were randomly assigned to receive one of four
versions of the vignette that varied in public health risk (high or low) and pandemic lockdown policies
(present or absent), to capture the two treatments suggested by our hypotheses. For all respondents,
the vignette opened by saying: “Imagine that your movement has called on its members to join a
massive street demonstration in the nearest big city to where you live next week.” This text was then
Prolific also does not allow participants under 18 to use its platform. We excluded residents of the United States
because of USIP’s mandate as a US nonpartisan peacebuilding institute.
The additional demographic questions in the screening survey were to obscure the eligibility requirements for the
follow-up survey and thus reduce the likelihood of participants gaming the survey through making false claims.
We do not report answers to all questions from the survey in this paper, but rather selected answers that were
indicative of broader trends in the data. A complete report of answers to all the survey questions, with full
question and answer text, as well as the survey data itself, is available from the authors upon request.
followed by one of the COVID risk statements and one of the lockdown policy statements in Table 2
Table 2: Survey Experiment Vignette Versions
High Covid Risk: “Public health authorities have
said that due to sharply-rising local COVID-19
infections, participation in the protest will be
Low Covid Risk: “Public health authorities have
said that due to sharply declining local COVID-19
infections, participation in the protest will be
Other Options Available: “Many businesses in
the city have reopened recently, and many
people are going to restaurants and movie
Other Options Unavailable: “Due to the
pandemic, most businesses in the city remain
closed, and there are no other major public
We then asked survey respondents whether they supported members of their movement
participating in the protest and how likely it was that they themselves would join the protest, with
answers for both questions measured on a five-point Likert scale.
Hypothesis 1 suggested that the average support for protest and the likelihood of personally
joining a protest would be significantly lower in the “high public health risk” condition. Hypothesis 2
suggested that average support for protest and likelihood of participation would be significantly lower
when other options for recreation were available. Since treatment is randomly assigned our primary
hypothesis tests were simple two-sample t-tests of difference in means between treatment and control
for both treatments across both dependent variables. We determine statistical significance at a level of
α = 0.05, and apply Dunn’s extension of the Bonferroni correction for multiple hypotheses (Dunn 1961).
Since we are measuring the impact of two treatments on two dependent variables, we determine
results to be statistically significant at a level of 𝛼 = 0.05
2×2 = 0.0125.
Our survey experiment results come with several caveats. Our sample is a convenience sample,
and thus faces issues of selection bias since we are limited to people who choose to participate in
Prolific and who speak English. We have no ex ante reasons to suspect that our sample is systematically
different from the general activist population, but also are only able to verify the representativeness of
Due to a coding error on the Qualtrics platform, where we housed our survey, one of the four conditions (Low
COVID risk and Other Options Available) was incorrectly specified, leading all those respondents to instead receive
the High COVID risk and Other Options Available combination of treatments. This means that, while the Other
Options variable is precisely balanced between the two conditions, the COVID risk treatment has 416 “High Risk”
and only 133 “Low Risk” responses. The results we report below are from the full sample, but are robust to
comparisons run on random samples of equal size between the two COVID risk conditions.
our sample through examining in-sample demographics. Our results also only report activists’ believed
response to a fictional vignette and are an imperfect measure of real-world behavior. We expect, for
instance, that social desirability bias likely impacts activists’ answers, perhaps leading them to report
higher likelihood of participation in the fictional protest. There is evidence that survey respondent
reactions to fictional vignettes are at least somewhat reliable indicators of real-world behavior
(Hainmueller, Hangartner, and Yamamoto 2015). Yet this limitation on the inferences that can be made
through this type of research are important to keep in mind.
In the following section we present the findings of the survey, first describing the survey
demographics, then the distribution of answers to several of the exploratory questions, and then finally
the results of the vignette experiment.
Sample Descriptive Statistics
Twenty-seven unique countries of residence had at least one respondent in our final sample.
While, the largest proportion of responses came from the United Kingdom, eight countries had more
than twenty responses. While the numerous countries represented in the sample increase our
confidence in our findings’ generalizability, it is important to note that the countries in our sample are
almost all developed democracies. The few exceptions, such as Mexico or Hungary, have relatively few
responses. Thus, we have limited capacity to make inferences about activists’ perceptions in developing
countries or in semi-democratic or authoritarian regimes. Our sample was quite young, with a median
age of 23, and skewed male (54%).
Figure 1: Countries of Residence of Activists in Sample
We also did not pre-register the survey experiment due to time constraints, and thus primarily present the results
Survey participants reported high levels of social movement participation, but little formal
professional involvement. Just under half of the participants (46%) indicated that they participate in
movement activities daily or weekly, and almost all (97%) indicated that they participate at least a few
times a year. Relatively few are paid for their activism: only 43 participants in the sample indicated that
they are employed at least part-time by their movement. This pattern of primarily volunteer-based
participation with very small numbers of professional employees fits well with the increasing
decentralization and informality of 21st century social movements (Tarrow 2011).
Our findings below should thus be interpreted as the perspective of a typical activist, neither an
outside observer looking in at social movements or a movement leader both personally and
professionally invested in the success of their cause. We consider this a strength of our approach, with
the perspective of “street-level” activists being a strong middle ground between highly-invested
movement entrepreneurs and the outside public that may be aware of but is not actively involved in
movement activity. But it is important to note that these exploratory results are the perceptions of a
typical activist, not reflective of reports from movement organizers.
As expected, the vast majority of participants indicated that the COVID-19 pandemic has
significantly impacted their movements. Two-thirds reported that innovating tactically has been a
significant challenge since the beginning of the pandemic. However, a significant proportion (36%)
indicated that their movement has been able to continue their normal activities, and roughly 81% of
respondents indicated that their movement has adopted new tactics in response to the COVID-19
pandemic. Thus, while movements are certainly struggling to adapt our respondents indicated that they
are actively responding to this challenge with tactical and strategic innovations.
One of the challenges that we expected was a lack of interest in activist causes as public focus
on the pandemic swamped out other concerns. The opposite was true, as activists from movements in
almost all issue areas reported increased public interest since the beginning of the pandemic. The only
issue area where activists perceived a decreased public interest during the pandemic was animal rights.
The issue where activists reported the highest increase in public interest was anti-racism,
understandable given the simultaneous significant increase in activism around the world associated with
the Black Lives Matter (BLM) movement.
Figure 2: Change in Perceived Public Interest since COVID-19 Pandemic Across Issue Areas
Our survey confirmed that a shift to digital activism has been a crucial part of movements’
adaptations to COVID-19. 76% of respondents reported that online activism was either “extremely
The relatively small number of respondents (13) in this issue area meant that the decreased level of interest was
not statistically significantly different from zero in a two-tailed t-test.
important” or “very important” for their movement during the COVID-19 pandemic, and nearly all (97%)
reported that at least some of their activities had shifted from in-person to online since the beginning of
the pandemic. “Digital campaigns” were the most common form of tactical innovation in response to
COVID-19 that activists described. Activists were overall optimistic about this shift, with 64% saying that
their movement had adapted to its increased online activity either “well” or “very well.”
Views of the Future
How will COVID-19 shape social movements into the future? To gain insight into these questions
we asked activists a series of questions about their perspective on the future of their movements, given
the COVID-19 pandemic. The picture was strikingly optimistic. While a slight majority (51%) of activists
said that the pandemic had made achieving their goals more difficult, activists were overwhelmingly
optimistic about their movement’s ability to advance its long-term goals after pandemic subsided (See
Figure 3 below), and a slight majority agreed that the COVID-19 pandemic was an opportunity to
advance a more ambitious reform agenda.
Figure 3: Activist Optimism About Post-COVID Environment
We now move from our exploratory analysis to our formal hypothesis testing of the impact of
public health risk and COVID-19 lockdown policies on participation in public demonstrations. Across all
conditions, participants were on average indifferent to supporting members of their movement
participating in a large public protest and considered it moderately unlikely that they themselves would
participate. The average on the support question was almost exactly equal to the “Neither Support nor
Oppose” answer (3.04 on a 1-5 Likert scale) while the average on the likelihood of personally joining the
protest was slightly lower (2.67 on the same 1-5 Likert scale).
Public health risk from COVID-19 significantly reduced support for movement participation and
the likelihood of anticipated personal participation. When COVID-19 risk is low, 53.4% of respondents
say they support or strongly support others joining a public demonstration. When the risk is high, this
drops to 34.9%. Similarly, when COVID-19 risk is low, 45.1% of respondents say they would be likely or
extremely likely to join a public protest. When the risk is high, this drops to 32.2%. Figure 4 shows the
difference in distribution across the high and low COVID-19 risk condition for both dependent variables.
Figure 4: Support for Protest and Likelihood of Protest Across COVID-19 Risk Conditions
The difference in means between high and low risk conditions were similar on both dependent
variables, roughly -0.6 (p < 0.001) for the first and -0.5 (p < 0.001) for the second. In both cases this
effect size is equivalent to roughly half a standard deviation. These effects are robust to an OLS model
with age, gender, country of residence, movement issues, and frequency of participation included as
covariates, with similar effect sizes for supporting other movement members participating in protest (β
= -0.64, p < 0.001) and likelihood of personal participation in protest (β = -0.55, p < 0.001).
In contrast, the availability of other options for personal recreation due to lockdowns only very
slightly reduced respondents’ support for fellow movement members participating in protest or the
likelihood of respondents themselves participating. When other options were unavailable due to
pandemic lockdown policies, 41% of respondents said they would support or strongly support others
joining a protest. When other options for recreation are available, this drops to 37.8%. Similarly when
other options for recreation are unavailable due to lockdown, 36.2% of respondents said they would be
likely or extremely likely to join a protest. When other options for recreation are available, this drops to
34.5%. Figure 5 displays the distribution of responses for both dependent variables.
Figure 4: Support for Protest and Likelihood of Protest Across Other Options for Recreation Conditions
While these differences were in the expected direction, with other options for recreation
reducing participation in public protests, the difference in means was far from statistical significance,
both when measuring the difference in means and in an OLS regression including covariates. Thus we
cannot conclude that other options for recreation significantly decrease perceived likelihood of
participating in protest or support for others participating in protest. Figure 4 summarizes the results of
all four hypothesis tests.
Figure 4: Survey Experiment T-test Results
These results suggest that, while the initial significant decline in public demonstrations that
accompanied the beginning of the COVID-19 pandemic may have abated, and many countries are
beginning to see the return of protesters in the streets, significant public health risk is still a major factor
reducing activists’ support for public protests and willingness to participate in protests.
Discussion and Conclusion
We are only at the very beginning of understanding COVID-19’s impacts on the global social and
political order. Its effects on public protest and social movements similarly remains obscure. While there
are beginning to be certain conventional wisdoms about the impact of COVID-19, for instance arguing
that the pandemic has stifled movement momentum, forced activists into an uncomfortable shift to
online activism, and may prevent long-term realization of social change, this conventional wisdom has
little systematic empirical support.
Our research vindicates some aspects of this conventional wisdom while putting others into
question. Activists from across the countries in our sample do indeed say that COVID-19 is a major
challenge to which their movements have struggled to adapt. And, as shown in our survey experiment
results, the public health risk from the pandemic does have a statistically significant negative effect on
participation in public protest. On the other hand, pandemic lockdown policies that affect the other
options for recreation available to activists do not, on average, appear to impact activists’ likelihood of
participating in protest. Our research also shows the pandemic does not appear to have led to
decreased interest in the causes that activists are advocating for. Indeed, most activists report that they
are seeing increased levels of interest since the beginning of the pandemic.
Furthermore, most activists do not seem to share the general pessimism about the pandemic’s
impact on their movements’ long-term potential for success. Indeed, a majority see the pandemic as a
window of opportunity to increase mobilization and push for transformative change. Only time will tell if
their optimism will be vindicated. The worldwide uprisings seen in 2019 may have only gone on
temporary hiatus during the height of the pandemic. Yet as the rise of anti-racist protests in the United
States and around the world in recent months shows, movements are already mobilizing for their return
to the streets and preparing to push for change.
This research comes with important limitations both in scope and in character. Our main
limitation in scope has to do with the countries and activists included in our sample. One critical
dynamic of the COVID-19 pandemic for many social movements has been an increase in government
repression, as authoritarian governments deploy the language of public health to impose unreasonably
harsh restrictions on freedom of assembly and speech (See e.g. Daraghi 2020). Since the countries in our
sample were almost entirely democracies, it is beyond the scope of our research to speak to this critical
dynamic, or other dynamics that are primarily playing out in semi-democratic or authoritarian countries.
Our main limitation in character comes to the nature of survey research. We are measuring the
reports of activist impressions of their movements, not external measures independent of perception.
Activist perceptions are an important indicator of the current state of the social movement space. A
movement whose activists are disillusioned and feel that they have lost public interest is critically
different from one whose activists are confident and feel that they have the public behind them. Yet we
cannot definitively validate these attitudinal measures against concrete external indicators.
These limitations in scope and character provide opportunities for future research. Further
examination of how COVID-19 is impacting movements in less democratic spaces is increasingly crucial,
as more activists continue to challenge unjust systems of authority and demand transformative change.
As nations continue struggling to mitigate these effects with no end yet in sight for the pandemic, the
crucial importance of understanding how the pandemic has impacted protests and social movements
will only become more critical.
This research was funded by the program on nonviolent action at the United States Institute of Peace.
The opinions expressed in this paper are solely those of the authors and do not necessarily reflect the
positions of the United States Institute of Peace.
AFP. 2020. “Hundreds Protest Government Corruption in Lebanon despite Virus Lockdown.” The Times
of Israel. April 18, 2020. https://www.timesofisrael.com/hundreds-protest-government-
Alsema, Adriaan. 2020. “Colombia’s President Reignites Protests against Himself over Coronavirus
Controversy.” Colombia News | Colombia Reports (blog). March 19, 2020.
Barberá, Pablo, Ning Wang, Richard Bonneau, John T. Jost, Jonathan Nagler, Joshua Tucker, and Sandra
González-Bailón. 2015. “The Critical Periphery in the Growth of Social Protests.” PloS One 10
Bennett, W. Lance, and Alexandra Segerberg. 2013. The Logic of Connective Action: Digital Media and
the Personalization of Contentious Politics. Cambridge University Press.
Bethke, Felix S., and Jonathan Pinckney. 2019. “Nonviolent Resistance and the Quality of Democracy.”
Conflict Management and Peace Science Onlinefirst.
Buchanan, Larry, Quoctrung Bui, and Jugal K. Patel. 2020. “Black Lives Matter May Be the Largest
Movement in U.S. History.” The New York Times, July 3, 2020, sec. U.S.
Celestino, Mauricio Rivera, and Kristian Skrede Gleditsch. 2013. “Fresh Carnations or All Thorn, No Rose?
Nonviolent Campaigns and Transitions in Autocracies.” Journal of Peace Research 50 (3): 385–
Cellan-Jones, Rory. 2020. “Animal Crossing Removed from Sale in China.” BBC News, April 13, 2020, sec.
Chaudhury, Dipanjan Roy. 2020. “Protests Against the Nepal Government’s Handling of COVID Pick up
Pace in Kathmandu.” The Economic Times, June 16, 2020.
Chenoweth, Erica. 2020. “The Future of Nonviolent Resistance.” Journal of Democracy 31 (3): 69–84.
Chenoweth, Erica, Austin Choi-Fitzpatrick, Jeremy Pressman, Felipe G. Santos, and Jay Ulfelder. 2020.
“The Global Pandemic Has Spawned New Forms of Activism – and They’re Flourishing | Erica
Chenoweth, Austin, Choi-Fitzpatrick, Jeremy Pressman, Felipe G Santos and Jay Ulfelder.” The
Guardian. April 20, 2020. http://www.theguardian.com/commentisfree/2020/apr/20/the-
Chenoweth, Erica, and Maria J. Stephan. 2011. Why Civil Resistance Works: The Strategic Logic of
Nonviolent Conflict. New York, NY: Columbia University Press.
Dahlum, Sirianne, Jonathan Pinckney, and Tore Wig. 2020. “Strategic Virtues? Individual Logics of
Nonviolent Civil Resistance.” Working Paper. Peace Research Institute of Oslo.
Daraghi, Borzou. 2020. “How Algeria’s Regime Used Coronavirus to Crush a Revolution.” The
Independent. July 14, 2020.
Dave, Dhaval, Andrew Friedson, Kyutaro Matsuzawa, Joseph Sabia, and Samuel Safford. 2020. “Black
Lives Matter Protests, Social Distancing, and COVID-19.” w27408. Cambridge, MA: National
Bureau of Economic Research. https://doi.org/10.3386/w27408.
Dettmer, Jamie. 2020. “Protesting in the Age of Coronavirus | Voice of America - English.” April 24, 2020.
Dunn, Olive Jean. 1961. “Multiple Comparisons among Means.” Journal of the American Statistical
Association 56 (293): 52–64.
Earl, Jennifer, Jayson Hunt, R. Kelly Garrett, and Aysenner Dal. 2015. “New Technologies and Social
Movements.” In Oxford Handbook of Social Movements, edited by Donatella Della Porta and
Mario Diani, 355–66. New York, NY: Oxford University Press.
Garrett, R. Kelly, and Paul N. Edwards. 2007. “Revolutionary Secrets: Technology’s Role in the South
African Anti-Apartheid Movement.” Social Science Computer Review 25 (1): 13–26.
Gladwell, Malcolm. 2010. “Small Change: Why the Revolution Will Not Be Tweeted.” The New Yorker.
September 27, 2010. https://www.newyorker.com/magazine/2010/10/04/small-change-
Hainmueller, Jens, Dominik Hangartner, and Teppei Yamamoto. 2015. “Validating Vignette and Conjoint
Survey Experiments against Real-World Behavior.” Proceedings of the National Academy of
Sciences 112 (8): 2395–2400. https://doi.org/10.1073/pnas.1416587112.
Klandermans, Bert, and Dirk Oegema. 1987. “Potentials, Networks, Motivations, and Barriers: Steps
towards Participation in Social Movements.” American Sociological Review, 519–531.
Lawal, Shola. 2020. “Coronavirus Halts Street Protests, but Climate Activists Have a Plan.” The New York
Times, March 19, 2020, sec. Climate.
McAdam, Doug. 1986. “Recruitment to High-Risk Activism: The Case of Freedom Summer.” American
Journal of Sociology 57 (4): 64–90.
McGowan, Charis. 2020. “How Quarantined Chileans Are Keeping Their Protest Movement Alive.” Al-
Jazeera. April 14, 2020. https://www.aljazeera.com/indepth/features/quarantined-chileans-
Metternich, Nils W. 2020. “Drawback before the Wave?: Protest Decline during the Covid-19 Pandemic.”
Olson, Mancur. 1965. The Logic of Collective Action: Public Goods and the Theory of Groups. Cambridge,
MA: Harvard University Press.
Palan, Stefan, and Christian Schitter. 2018. “Prolific. Ac—A Subject Pool for Online Experiments.” Journal
of Behavioral and Experimental Finance 17: 22–27.
Peer, Eyal, Laura Brandimarte, Sonam Samat, and Alessandro Acquisti. 2017. “Beyond the Turk:
Alternative Platforms for Crowdsourcing Behavioral Research.” Journal of Experimental Social
Psychology 70: 153–163.
Pensulo, Charles. 2020. “Malawi Health Workers Protest against Lack of Protective Gear.” Al-Jazeera.
April 14, 2020. https://www.aljazeera.com/news/2020/04/malawi-health-workers-protest-lack-
Pinckney, Jonathan. 2020. “Amid Coronavirus, Online Activism Confronts Digital Authoritarianism.”
United States Institute of Peace. April 28, 2020.
Pinckney, Jonathan, and Miranda Rivers. 2020. “Nonviolent Action in the Time of Coronavirus.” United
States Institute of Peace. March 25, 2020.
Pudrovska, Tetyana, and Myra Marx Ferree. 2004. “Global Activism in ‘Virtual Space’: The European
Women’s Lobby in the Network of Transnational Women’s NGOs on the Web.” Social Politics:
International Studies in Gender, State & Society 11 (1): 117–143.
Raleigh, Clionadh, Andrew Linke, Håvard Hegre, and Joakim Karlsen. 2010. “Introducing ACLED: An
Armed Conflict Location and Event Dataset: Special Data Feature.” Journal of Peace Research 47
(5): 651–60. https://doi.org/10.1177/0022343310378914.
RezaeeDaryakenari, Babak, and Peyman Asadzade. 2020. “Learning about Principles or Prospects for
Success? An Experimental Analysis of Information Support for Nonviolent Resistance.” Research
& Politics 7 (2): 2053168020931693. https://doi.org/10.1177/2053168020931693.
RT. 2020. “Yellow Vests DEFY Covid-19, Lockdown & TEAR GAS to Protest Macron Government
(VIDEOS).” RT International. March 14, 2020. https://www.rt.com/news/483129-yellow-vest-
Schock, Kurt. 2005. Unarmed Insurrections: People Power Movements in Nondemocracies. Vol. 22.
Minneapolis, MN: U of Minnesota Press.
Serhan, Yasmeen. 2020. “Israel Shows Us the Future of Protest.” The Atlantic. April 23, 2020.
Sharp, Gene. 1973. The Politics of Nonviolent Action. Boston, MA: Porter Sargent.
Snow, David, Daniel Cress, Liam Downey, and Andrew Jones. 1998. “Disrupting the Quotidian:
Reconceptualizing the Relationship Between Breakdown and the Emergence of Collective
Action.” Mobilization: An International Quarterly 3 (1): 1–22.
Tarrow, Sidney G. 1989. Democracy and Disorder: Protest and Politics in Italy, 1965-1975. Oxford
University Press, USA.
———. 2011. Power in Movement: Social Movements and Contentious Politics. 3rd ed. New York, NY:
Cambridge University Press.
Tufekci, Zeynep. 2017. Twitter and Tear Gas: The Power and Fragility of Networked Protest. New Haven,
CT: Yale University Press.