Slacktivism and the social benefits of social video: Sharing a video to 'help' a cause
First Monday

Slacktivism and the social benefits of social video: Sharing a video to 'help' a cause by Cat Jones

Interest in online ‘slacktivism’ is high both within and outside academia, and the online sharing of cause-related campaigns is increasing, but research into the extent to which ‘slacktivism’ applies to the act of sharing online content to ‘help’ a cause is limited. Slacktivism, as used in much of today’s popular discourse, is defined as ‘feel-good online activism [with] zero political or social impact’ (Morozov, 2009). Here, a survey-based analysis investigates social videos’ impacts on U.K. and U.S. viewers. Results show that the stronger viewers’ motivations are to share to help a cause, the stronger their motivations are to find out more afterwards. Further, a case study shows that social videos that motivate viewers to share the video ‘because it’s for a good cause, and I want to help’ (cause-related sharing versus, for example, to appear knowledgeable about a subject), can also prompt real-life actions, including donations. These results indicate that cause-related video sharing does have an impact, and therefore is not slacktivism by Morozov’s definition. Preliminary analysis suggests that of viewers sharing to ‘help’ causes, those discriminating in their choice of sharing audience (narrow vs. broad) are more likely to further engage than indiscriminate sharers. Result patterns from U.S. narrowcast sharers differ from those of other groups, indicating that tie strength and cultural differences may play a role in modifying slacktivist behaviors.


Introduction and literature
Results and analysis



Introduction and literature


Social video — defined as online video content that gets people talking (Unruly, 2012) — is increasingly being used by charities, not-for-profits, and other causes to spread messages and engage audiences. Social videos that have contributed towards real-life progress against the objectives of the causes they promote certainly exist. A prominent example is ‘KONY 2012’ (KONY), created by the not-for-profit organization, Invisible Children, to pressure political leaders worldwide to end human rights abuses from Ugandan warlord Joseph Kony and his Lord’s Resistance Army (LRA). The video amassed 50 million views and six million shares within three days of its launch, with nearly one in seven viewers sharing the video online (Waterhouse, 2012). Invisible Children (2012) attribute a range of real world, meaningful progress towards their ultimate objectives to the Kony 2012 video including over 3.7 million petition signatures and legislation passed by the U.S. Congress authorizing a US$5 million reward for information leading to Kony’s arrest. This video’s unprecedented success at raising awareness and generating publicity caused it to be viewed by many as proof of the effectiveness of online activism (Keeter, 2012) — though for many others it was ‘slacktivism’ in action (Audette, 2013). Generally, ‘we know little about the benefits — and possible costs — of engaging in social activism via social media’ [1] and many commentators today believe that people engaging in light-touch (especially social media-based) support activities exhibit ‘slacktivist’ behaviors.

The Oxford English Dictionary (2014) defines a ‘cause’ as ‘a principle, aim, or movement ... one is prepared to defend or advocate’, and ‘slacktivism’ as ‘actions performed via the Internet in support of a political or social cause but regarded as requiring little time or involvement’. However, popular use of the term ‘slacktivism’ carries negative associations: Morozov’s 2009 definition, ‘feel-good online activism that has zero political or social impact’, is closer to current usage and is the definition used here.

It may be true that, as Nielsen argues [2], ‘Generalizations as to whether the net benefit is positive or negative are hard to make and probably not useful — since new technologies are increasingly ubiquitous and here to stay, rejecting them ... would mean a self-imposed exile from the shared built communications environment’. An understanding of the extent to which the inevitable act of sharing cause-related videos online contributes to — or detracts from — the ultimate objectives of those causes is essential in today’s world.

Online advocacy opportunities can mobilize those not previously involved in activism, increasing public awareness of issues if nothing else. However, the question of whether token support prevents meaningful support remains unanswered. Morozov (2009) states, ‘the only way to conclusively answer this question is a scientific one: we simply need to start constructing gigantic surveys’. This study does exactly that, analyzing survey responses to obtain a preliminary answer.


A social video is an online video that is, or can be, influenced by social media (Business Insider, 2013). Social video is a medium through which persuasive information can be conveyed at speed and scale, and is growing fast: Cisco (2013) expect there to be 1.5 billion Internet video users by 2016. Advertisers are already buying into its power — social videos from advertisers collectively achieved over 71 million shares from June to August 2013 (Unruly, 2013). Barrett and Leddy [3] assume, ‘if ads can sell products, visual imagery linked to a social justice narrative can sell social action, or political conviction’. Marketing technology company Unruly uses regression techniques to identify viewer responses correlated with sharing (Unruly, 2014), finding that a common motivation for sharing video content is ‘because it’s for a good cause and I want to help’, and that the extent to which viewers want to share a particular video to help a cause is well correlated with actual sharing of that video [4].

Popular interest in slacktivism and online sharing has increased recently; 2014 has seen significant popular commentary on the topic in the wake of several high profile and widely shared cause-related online campaigns, including #nomakeupselfies, the ALS Ice Bucket Challenge and #bringbackourgirls (e.g., Kosinski, 2014; Taylor, 2014).

Research aim

Does the act of sharing a video to ‘help’ a cause constitute a slacktivist act that does not contribute to — or indeed that damages — the ultimate objectives of a cause?

The aim of this study is to investigate, via quantitative survey responses and a qualitative and quantitative case study, the extent to which sharing a social video to ‘help’ a cause constitutes a slacktivist action. Statista’s (2014) ranking of countries by the percentage of Internet users watching online video has both the U.K. and U.S. in the top 10. The data used in this study are restricted to U.K. and U.S. survey responses — future research could investigate other markets to determine whether findings generalize.

Research question

Is sharing a cause-related video to help a cause associated with a reduced propensity to further engage with the cause?


Social video is a relatively new medium; there exists little research into how sharing a cause-related video may contribute to, or detract from, a cause’s ultimate objectives. This investigation hopes to address that gap. However, there is a body of research concerned with the impact of Internet and social media based actions on causes, particularly the role played by ‘slacktivism’.

Kristofferson, et al. (2014) define actions that allow people to show their support for causes with little associated cost (e.g., liking a Facebook page) as ‘token support’, and contributions requiring significant cost/effort/behavior change with tangible positive impacts for the cause (e.g., donating money) as ‘meaningful support’. Similarly, Karpf [5] differentiates between ‘tactical’ and ‘strategic’ metrics for success. The Internet has made it easier for people to show token support than historically through off-line channels (Van Laer and Van Aelst, 2010). Sharing a video to support a cause is a version of Kristofferson, et al.’s token support, and Morozov (2009) refers to demonstrations of token support as ‘slacktivist’ activities. The important question is: can token support lead to meaningful support? Opinions are divided.

McCaugley and Ayers’ overview of cyberactivism summarizes a pervasive concern: because people feel ‘they are doing something useful’ when sharing cause-related content online, ‘apathy grows with each new Internet connection’ [6]. Morozov (2009) agrees, suggesting light-touch activities are likely to result in organizational losses to traditional activists, as people use ‘slacktivist’ activities to replace ‘conventional (and proven) forms of activism’ (demonstrations, litigation, etc.). He warns of unanticipated negative effects of digital activism on ‘more effective’ forms of enacting change. This is important: if token support prevents meaningful support, it could actively damage the strategic objectives of causes. Finding research where results support this view is difficult, and Morozov does not cite any. Christensen’s (2011) literature review, which assesses whether Internet activism replaces traditional forms of off-line political participation, also finds no evidence for the substitution thesis. However, Morozov’s hypothesis makes intuitive sense, and has been widely adopted. The lack of conclusive research supporting or refuting the hypothesis (Morozov, 2009) frequently leads authors to conclude that social media’s effects on causes are unclear (e.g., Rotman, et al., 2011). There is certainly a need for further research in this area.

Kristofferson, et al. (2014) find that ‘consumers exhibit greater helping on a subsequent, more meaningful task after providing an initial private (vs. public) display of token support’. However, their research was carried out across a short time period and it is possible that people newly introduced to the aims of a particular cause may initially participate passively and become more active later, as the cause becomes more familiar. This time-delayed conversion to meaningful support was found in Neumayer and Schoßböck’s (2011) study of an Austrian student movement but not measured by Kristofferson, et al.

Many authors disagree with Morozov’s hypothesis. Shirky [7] observes that ‘the fact that barely committed actors cannot click their way to a better world does not mean that committed actors cannot use social media effectively’, and Van Laer and Van Aelst [8] argue that ‘the shift towards new internet-based actions and tactics ... has complemented [the old action forms]’. Feezel, et al. (2009) found through multivariate regression analyses that participation in online political groups predicts off-line political participation. Lee and Hsieh (2013) found that participants who signed an online petition were significantly more likely to donate their participation payment to a related charity, and that participants who did not sign were more likely to donate to an unrelated charity. They attribute this to a ‘moral balancing’ effect, suggesting that ‘being involved in effortless political activities online ... reinforce[s] off-line engagement’.

Georgetown University’s Center for Social Impact Communication (2011) delivered a survey to 2,000 Americans, showing that ‘social media cause promoters’ are more likely to participate in certain activist-style activities than ‘non-social media cause promoters’. The Center’s conclusion is that ‘social media is simply being added to [social media cause promoters’] range of engagement activities, not replacing the more historically prominent ways of supporting’ [9]. Problematically, the study does not include a control that permits this inference: no data relate to what the individuals would have done had they not been using social media to promote causes (perhaps those most likely to promote causes on social media belong to socio-economic/demographic profiles with greater propensities to engage with social causes in general). Despite the flaws in analysis, the data suggest that token support may complement meaningful support.

Katz and Lazarsfeld (2006) suggest that ideas flow from mass media to opinion leaders, and from them to ‘less active’ population members [10]. By sharing cause-related videos, people become opinion leaders, potentially influencing their connections’ opinions. Therefore, even if sharing a video constitutes a slacktivist activity for the sharer, the impact on their personal network could result in a net positive effect.




Unruly, an authority on social video, provided a large sample of raw data (n=82,759) from its Unruly ShareRank dataset, along with the original questionnaire, for use in this research. To test the slacktivist hypotheses, a cross-sectional comparison was conducted using the Unruly dataset. The strength of the viewers’ intent to ‘find out more’ was adopted as a proxy for ‘meaningful support’, since finding out more is a prerequisite for further, meaningful, action. Morozov’s ‘slacktivism’ hypothesis, applied to social video, provides Hypothesis 1; Kristofferson, et al.’s finding relating to slacktivism and social observability provides Hypothesis 2.

H1: The more likely a viewer is to share a video to help a cause, the less likely he/she is to want to find out more.
H2: Viewers that share cause-related videos to help a cause are more likely to want to find out more when they share with narrow (vs. broad) audiences.

The variables from the dataset shown in Table 1 permitted statistical hypothesis testing:


Table 1: Variables for testing.
VariableResearch hypothesesVariable type
Strength of viewers’ intent to share a video ‘because it’s for a good cause and I want to help’
[Abbreviation: ‘Social Good’]
H1; H2Ordinal:
0–10 Likert scale
Strength of viewers’ intent to ‘find out more’H1; H2Ordinal:
0–10 Likert scale
Audience with whom viewers intend to share the videoH2Categorical: 5 categories


Hypothesis tests were complemented by a case study video for which real-life strategic success measures were known, which contextualizes findings and links ‘find out more’ intent to meaningful support.

Though the Unruly dataset included qualitative responses, its size made comprehensive qualitative analysis impractical; quantitative analysis was prioritized to permit generalization in hypothesis testing. However, the case study’s relatively small sample permitted qualitative analysis, adding depth to quantitative findings.

Unruly’s structured survey questionnaire measured the same variables across many cases, permitting consistent analysis of data gathered from multiple surveys conducted at different times.

Within Unruly’s dataset, questionnaire respondents for each video formed a demographically balanced sample in line with the country’s online population. Participants were recruited from online panels using interlocking age/gender/social grade quotas to ensure representative samples. Responses were automatically recorded using an online tool. Since large online panels do not use probability-based recruitment, certain population classifications can be overlooked/over-represented according to their incidence in the sampling frame (Baker, et al., 2010). However, the probability-sampling method, the fact that large proportions of U.K. and U.S. populations are registered with online panels (ResearchNow’s March 2012 panel size data claims 10 percent), and Unruly’s strict data quality checks, minimized sampling bias. The purpose of this research was not to produce estimates of population values, but identify general associations between variables. As Baker, et al. observed [11], in situations like this, the ‘lower cost and unique properties of Web data collection are an acceptable alternative to traditional probability-based methods’.

A sample size calculation on the dataset showed that the minimum required for hypothesis testing is 1,010 responses. To permit generalization, data must span multiple videos. The secondary dataset contains 82,760 responses for 479 videos, of which 338 motivated (to some extent) at least 50 percent of respondents to share the video to help a cause. The number of respondents that saw each video is relatively evenly distributed (mean: 163; standard deviation: 46); the minimum is 100. The sample comprises data only for videos launched since July 2012. This dataset is relevant and extensive, with a sample size more than 80 times the minimum. The provision of raw data eliminated the chance of inheriting potential second-hand arithmetical errors.

Most videos in Unruly’s dataset were randomly sampled, however some were tested at the request of commercial clients, introducing some selection bias. Commercial, non-commercial, cause- and non-cause-related videos were included; while cause-related videos provided data relevant to the research aim and questions, the others likely introduced some bias. However, they also facilitated the exploration of differences between viewers motivated and not motivated to share to help a cause by providing data for significant numbers of respondents not motivated to share to help. The bias was minimized for hypothesis testing by excluding respondents whose ‘Social Good’ score was 0, removing most data from non-cause-related videos.

Since secondary data were used, human participants were not recruited and no risks were associated with data collection. The data provided by Unruly were anonymized. The only significant risk in conducting this research using secondary data was a potential breach of confidentiality relating to the confidential information of a third party, namely Unruly’s full questionnaire and the data sample provided. To mitigate this risk, data was encrypted while stored and in transit, the full questionnaire was not published, and explicit approval was obtained from Unruly in writing before publication.



Results and analysis

Differences were identified between U.S. and U.K. responses for key variables, so the two were analyzed separately. For hypothesis testing, the large sample (U.K. n=33,842; U.S. n=48,917) presented an opportunity to set the threshold p-value below which test results were considered significant to 0.01 (Borradaile, 2003), reducing the chance of a Type I error to one percent compared with the five percent chance of the 0.05 threshold common in social sciences (Calkins, 2005).

As discussed above, for hypothesis testing the dataset was restricted to respondents motivated to share a video to support a cause to some degree (Social Good intent > 0/10), dubbed ‘cause-related sharers’ for convenience (though low scores on the Social Good scale indicate only a modest intent to share and as such are unlikely to have prompted actual cause-related shares). Since no variables were normally distributed and all were ordinal, Kendall’s τ and Spearman’s ρ rank correlations were used to test for hypothesized relationships (and identify their directions). Although some authors argue that ordinal variables can be considered continuous interval scales for analysis (e.g., Norman, 2010), many do not (Knapp, 1990). Since all hypotheses could be tested using nonparametric techniques whose appropriateness could not be questioned, no parametric tests were used.

H1: The more likely a viewer is to share a video to help a cause, the less likely he/she is to want to find out more

A visual inspection of Figure 1 shows that high Social Good scores with low Find Out More scores are unusual, and vice versa: high/low scores for one variable often result in correspondingly high/low scores on the other. The slight vertical stripe pattern is an artifact of the dataset (Unruly initially measured social motivation intensities using a 1–5 scale; this was later converted to 1–10 and historical results multiplied by two) and will be visible throughout.


Viewer proportions experiencing  'social good' and 'find out more' motivations at different intensities
Figure 1: Heatmap: Viewer proportions experiencing ‘social good’ and ‘find out more’ motivations at different intensities (respondents motivated, to any degree, to share to help a cause).
Note: Larger version of image available here.


Null and alternative hypotheses were formulated:

H10: There is no relationship between Social Good and Find Out More intensities.
H1A: There is a relationship between Social Good and Find Out More intensities.

Both correlation tests, in both territories, yielded a statistically significant test statistic p<0.01. Therefore, the null hypothesis was confidently rejected: a relationship exists between the intensity of a viewer’s intent to find out more, and their intent to share to support a cause.

While H1 predicts a negative correlation, τ and ρ show moderate (Botsch, 2011; Cohen, 1988) positive correlations, 0.26 and 0.33 respectively in both territories. A modified alternative hypothesis was therefore accepted:

H1A2: The more likely a viewer is to share a video to help a cause, the more likely he/she is to want to find out more.

This finding contradicts Morozov’s slacktivist hypothesis, which predicts a negative correlation between cause-related sharing intent and further engagement with a cause.

H2: Viewers that share cause-related videos to help a cause are more likely to want to find out more when they share with narrow (vs. broad) audiences

In the Unruly ShareRank questionnaire, respondents motivated to share were asked which audiences they intended to share with. These audiences were categorized as broad or narrow (Table 2). Respondents intending to share with both broad and narrow audiences were excluded. Null and alternative hypotheses were formulated:

H20: There is no difference between the intent of cause-supporting sharers to find out more when they share with narrow audiences versus broad audiences.
H2A: There is a difference between the intent of cause-supporting sharers to find out more when they share with narrow audiences versus broad audiences.


Table 2: Categorization of sharing audiences.
Sharing audienceCategorization
Family membersNarrow
Best friends
Select friends
Work friendsNot categorized
(broadcast or narrowcast depending on work environment)
All Facebook friendsBroad
Everyone (public platform)


While tests for significant differences between Pearson’s parametric correlation coefficients for independent samples are common, tools for conducting the same analysis on nonparametric correlation coefficients are not. Therefore, H2 was tested by obtaining a threshold value (upper quartile) Social Good motivation intensity, above which respondents were classed ‘cause-supporting sharers’ and below, ‘non-cause-supporting sharers’. Find Out More motivation intensity distributions were compared between broadcast and narrowcast cause-supporting sharers; Mann-Whitney U Tests tested for differences between them. Rank correlations and heatmaps aided the interpretation of results. The upper quartile value for the U.S. sample was five, and the U.K. sample four, but for consistency and ease of analysis, the upper quartile value across both samples (five) was used for U.K. and U.S. tests.

For U.K. respondents, the Mann-Whitney U Test gave p>0.01 (Figure 2) and the null hypothesis was retained. However, the result was not clear-cut: had the significance level been set at 0.05, the null hypothesis would have been rejected. Box plots and histograms show similar distributions across both audience types, supporting the retention of the null hypothesis:

H20: In the U.K., there is no difference between the intent of cause-supporting sharers to find out more when they share with narrow audiences versus broad audiences.


U.K. 'Find out more' intent for broadcast vs. narrowcast cause-supporting sharers
Figure 2: U.K.: ‘Find out more’ intent for broadcast vs. narrowcast cause-supporting sharers.
Note: Larger version of image available here.


In the U.S., however, tests for difference displayed asymptotic significances with p<0.01 (Figure 3) and the null hypothesis was rejected; boxplots indicate that broadcast cause-supporting sharers were more likely to strongly intend to find out more than narrowcast cause-supporting sharers.


U.S. 'Find out more' intent for broadcast vs. narrowcast cause-supporting sharers
Figure 3: U.S.: ‘Find out more’ intent for broadcast vs. narrowcast cause-supporting sharers.
Note: Larger version of image available here.


Correlation coefficients (Table 3) showed significant positive correlations between Social Good and Find Out More motivation intensities for broadcast and narrowcast sharers in both territories. Correlation strengths were similar. In the U.K., the correlation appeared slightly stronger for narrowcast sharers; in the U.S., for broadcast sharers.


Table 3: Correlations: ‘Social good’ and ‘find out more’ motivation intensities (respondents motivated, to any degree, to share to help a cause).
Note: **Correlation significant at the 0.01 level (2-tailed).
  Kendall’s τSpearman’s ρ
U.K.Correlation coefficient0.350**0.345**0.443**0.437**
Sample size (n)10,2134,80210,2134,802
U.S.Correlation coefficient0.388**0.394**0.488**0.495**
Sample size (n)14,4586,47614,4586,476


A visual inspection of the distributions (Figure 4) reflects the positive correlations indicated by the coefficients. Similar relationships were seen between sharing audience types in the U.K., but material differences appeared between narrowcast and broadcast sharers in the U.S.


Broadcast/narrowcast heatmap: Proportions of viewers experiencing 'social good' and 'find out more' intensities
Figure 4: Broadcast/narrowcast heatmap: Proportions of viewers experiencing ‘social good’ and ‘find out more’ intensities (respondents motivated, to any degree, to share to help a cause).
Note: Larger version of image available here.


For U.S. broadcast sharers, there is a ‘backbone’ of high respondent proportions where Social Good and Find Out More motivation intensities are approximately equal, similar to the U.K. pattern and that for both territories in the H1 tests. For narrowcast sharers, the backbone is much weaker, with 61 percent of narrowcast sharers experiencing Social Good motivations at intensity=10. This indicates that U.S. viewers tend to share with narrowcast audiences when their intent to share to support a cause is 10/10 (perhaps when potential sharers can think of specific recipients the share event becomes more certain). However, Find Out More intensities are reasonably evenly distributed along the 1–10 scale for these viewers (with a distribution not far from normal and a positive skew), indicating some insensitivity of the strength of Find Out More intent to that of Social Good when Social Good intent=10/10. There may be two distinct sub-groups within the U.S. narrowcast sample: ~40 percent for whom the intensities of Social Good and Find Out More intents are well correlated, each increasing with the other (with a similar pattern to that of broadcast sharers); and ~60 percent whose Social Motivation intent intensity is 10/10, but whose Find Out More intent intensities vary.

The null hypothesis was rejected for the U.S., and a specific alternative hypothesis was formulated:

H2A2: In the U.S., a significant proportion of narrowcast sharers are strongly altruistically motivated regardless of their desire to find out more.

Interestingly, Kendall’s τ for broadcast and narrowcast audiences in the U.K. and U.S. showed a strong correlation (Botsch, 2011) compared with the moderate correlation in the larger sample tested for H1: excluding respondents intending to share with both broadcast and narrowcast audiences appears to have increased correlation strength. This suggests that cause-related sharing intent may be more strongly associated with find out more intent for viewers who discriminate in their choice of sharing audience. Statistically testing for differences between H1 and H2 correlation coefficients (dependent samples) was beyond the scope of this study, but a new hypothesis was generated:

H3: Altruistically motivated sharers are more likely to want to find out more when they choose to share with either broad or narrow audiences versus when they share with both.

Case study

Our Better World (OBW), a digital initiative of the Singapore International Foundation, uses social video to tell stories about socially worthwhile initiatives and promote a global culture of ‘doing good’. OBW provided data pertaining to the impact of one of its social videos, ‘Kitchen of Cheer’, largely distributed in Singapore and the U.S. Unruly tested this video against a nationally representative U.S. sample of 151 viewers using the Unruly ShareRank survey; variables measured are consistent with those investigated above. The data were analyzed to

a) Elucidate the link between intent to find out more and actual progress against strategic success measures due to the video; and,
b) Use qualitative response data to add color and context to quantitative findings.

‘Kitchen of Cheer’ can be watched online at The story explains how a defining moment leads a Singaporean piano teacher, Mavis, to set up a Cambodian soup kitchen for children, combatting hunger and malnutrition and helping them to learn at school. The strategic objectives of ‘Kitchen of Cheer’ were to secure donations and volunteers to help Mavis feed more Cambodian children, and to inspire viewers to initiate positive social change in their own communities.


Table 4: Video views for ‘Kitchen of Cheer’, distributed 3–10 September 2013.
 During campaignAfter campaign
Most views within four weeks)
Singapore (approx.)6,3183836,700
U.S. (approx.)17,8871,67419,561


OBW partnered with international video channels to enable viewers to discover the video. Total views are shown in Table 4: almost half came from the U.S. In the Unruly data, U.S. viewers’ ‘find out more’ intent scores were significantly higher than the norm for U.S. videos (p<0.01, Independent Samples Median Test; Figure 5). This permits the inference that a significant proportion of U.S. viewers to whom ‘Kitchen of Cheer’ was distributed were motivated to find out more.


'Find out more' and 'social good' boxplots
Figure 5: ‘Find out more’ and ‘social good’ boxplots.
Note: Larger version of image available here.


Prior to the video’s launch, cash donations to Mavis’ kitchen were limited to her social circle. Following the launch, donations from outside that circle commenced, raising US$6,500 between the video’s launch and 21 July 2014. Donations from Southeast Asia were also received, resulting in US$16,270 total (Figure 6). The video’s distribution was the only real connection that Mavis’ project had to the U.S. and therefore it is reasonable to assume that the US$6,500 donated directly resulted from the 19,561 U.S. views of ‘Kitchen of Cheer’. To the knowledge of OBW, no U.S. donations occurred before the video’s launch, so zero pre-video U.S. donations have been assumed. The total donation increase enabled Mavis’ kitchen to feed an extra 50 people per week and extend her kitchen and storage space. Other positive impacts of the video on strategic objectives were indicated (e.g., increased volunteer numbers; physical item donations), however the data available limited the extent to which these could be quantified. On 6 June 2014, R. Lim confirmed by an e-mail message that, “Mavis received more than 100 genuine offers of help (donations, volunteers, supplies delivered), from Singapore (majority), as well as from the U.S. [and other countries]”.


Donations to Mavis' kitchen
Figure 6: Donations to Mavis’ kitchen (‘Kitchen of Cheer’ launched September 2013; most views delivered 3–10 September).
Note: Larger version of image available here.


These data suggest that videos with high ‘Find Out More’ and ‘Social Good’ intent scores do in fact lead to real world actions (like donations) and genuine progress against the strategic objectives of the cause the social video is used to promote.

Verbatim qualitative responses exist for those respondents that wanted to share for ‘other’ reasons than those tested. Qualitative responses are summarized in Table 5 and Figure 7. The most common ‘other’ reasons for wanting to share centered on informing, inspiring or prompting action from others — trends reflected in the word cloud summary of commonly used words. Several reasons for sharing to inform were specific to an American context, for example “for plps to realize how good of a life they have,” indicating that some viewers were motivated to share by the difference between the lives in the video and their own. The American context was also seen in some of the ‘prompting action’ motivations: “... maybe it would inspire [my friends] to make a program similar to this one in their communities”. ‘Kitchen of Cheer’ appeared not only to have contributed towards Mavis’ objectives, but also to have furthered the OBW strategic objective of inspiring viewers to contribute to positive social change in their communities.


Classification of qualitative responses to 'Other -- please specify' question
Table 5: Classification of qualitative responses to ‘Other — please specify’ question; most common categories only, not all responses were classified.
Note: Larger version of image available here.



Word cloud summary of qualitative responses
Figure 7: Word cloud summary of qualitative responses.
Note: Larger version of image available here.





Results imply that, when Morozov’s (2009) definition of slacktivism is used, sharing social videos to ‘help’ a cause is not a slacktivist act; indeed, the stronger a viewer’s motivation to share a video to help a cause, the stronger their motivation to find out more afterwards. However, meaningful support actions do not necessarily follow from a motivation to find out more. So, the extent to which the existence of Morozov’s slacktivism for altruistic video sharing could be categorically disproven was limited by the variables available and by the survey-based methodology, which restricted findings to viewers’ reported motivations, rather than actions. In the case of ‘Kitchen of Cheer’, strong ‘find out more’ intent was linked to meaningful support: social video viewing and sharing can result in meaningful support for small, relatively unknown cause-related videos as well as for well-known examples like KONY 2012.

This finding contradicts Morozov’s (2009) hypothesis that when token support is present, meaningful support is absent. It shows that social videos can catalyze meaningful support. These results are consistent with Christensen’s (2011) finding that there is no evidence for Morozov’s substitution thesis and Van Laer and Van Aelst’s [12] conclusion that Internet-based actions complement rather than replace ‘old action forms’. The complementary interaction is well illustrated by one respondent’s comment regarding ‘Kitchen of Cheer’: “If I decide to share ... it would be to see what we could do in our area”. Here, the very act of sharing the video is used to begin a conversation about direct action. As discussed above, Lee and Hsieh (2013) found that petition signers donated to a related charity and non-petition signers donated to an unrelated one. A similar effect was seen for ‘Kitchen of Cheer’: some viewers shared to help Mavis’ efforts in Cambodia, while others shared not to help Mavis, but to kick off a similar project at home.

In light of these results, the nuances of the slacktivist concept must be explored a little further to elucidate the link between slacktivism and cause-related video sharing. Kwang-Suk [13] suggests that slacktivism be ‘defined locally ... instead of having a general definition with a negative nuance’. In doing so, he takes slacktivism back to its roots as a simple portmanteau of the words ‘slacker’ and ‘activism’, prior to its evolution into a more pejorative term (Skoric, 2012) linked to activities with ‘zero political or social impact’ (Morozov, 2009). Rotman, et al. [14] define slacktivism as ‘low-risk, low-cost activity via social media, whose purpose is to raise awareness, produce change, or grant satisfaction to the person engaged in the activity’. Here, the defining characteristic of ‘slacktivism’ is that it is low-risk and low-cost; unlike Morozov’s definition, slacktivist activities do not need to have zero impact. Harris [15] conceptualizes slacktivist actions as falling ‘along a spectrum of situations, somewhere in between fully-engaged, effective action and lazy, detached lip service’.

‘Pop media’ tends to implicitly use Morozov’s more negative definition. The central difference between the various conceptualizations of slacktivism is in whether ‘impact’ is encompassed within the definition itself (Morozov) or whether it is a separate question (Rotman, et al.). If slacktivism equals ‘zero impact’, as Morozov claims, this study indicates that cause-related social video sharing is not a slacktivist act. If slacktivism rather means ‘low-risk, low-cost activity via social media’, as Rotman, et al. contend, then cause-related social video sharing is clearly slacktivist — though this is not to say that it is necessarily not effective. It may ‘raise awareness’, ‘produce change’, simply ‘grant satisfaction to the person engaged in the activity’, or some combination. As soon as the definition includes a built-in assumption of ‘zero impact’, the slacktivist concept should no longer be applied to the act of cause-related video sharing.

Future research could make explicit the link between ‘find out more’ intent and strategic success measures to categorically disprove Morozov’s slacktivist hypothesis in the case of cause-related sharing. This could be done by extending the strategic impact analysis of a video in this paper’s case study to include many videos, permitting further interrogation of the link between ‘find out more’ intent and strategic success.

This research could not replicate Kristofferson, et al.’s (2014) finding that meaningful actions (‘find out more’ intent used a proxy) are more common after initial private (vs. public) displays of token support. This may be due in part to the fact that sharing, even to limited audiences, is by its very nature a socially observable action. A body of literature relates to the importance of tie strength for social actions (Marsden and Campbell, 1984) and the results of the present research indicate differences that cannot be well explained without further analysis, the interpretation of which could be aided by an understanding of the importance of tie strength for cause-related sharing. Whilst for most people, cause-related sharing motivation increases with intent to find out more, a majority of U.S. narrowcast sharers were strongly altruistically motivated regardless of their desire to find out more. In addition, altruistically motivated sharers seem more likely to want to find out more when they choose a specific audience to share with versus when they share with everyone. Differences between U.K. and U.S. broadcast/narrowcast behaviors indicate that tie strength and cultural differences may significantly influence altruistic video sharing behaviors — a topic for future research that has received limited academic attention to date.

In light of the U.K./U.S. differences, another pertinent factor may be that Kristofferson, et al.’s study was conducted on students at a Canadian university. It is possible that their finding would not be replicated if their study were repeated in the U.K. or U.S. using a nationally representative sample instead of students.

It is important to note that the reliability of the quantitative results of this study is somewhat limited by the likely presence of common-method variance. Variables between which correlations were tested shared common methods, and those methods may have exerted a systematic effect, at least partially posing a rival explanation for the observed correlations (Podsakoff, et al., 2003). It is also important to be aware that, because of the large sample size, small effects can still yield significant test results.




The aim of this study was to understand whether sharing a cause-related video to help a cause constituted a slacktivist action. Results show that, at least for U.K. and U.S. viewers, social videos that can successfully prompt cause-related sharing from viewers tend also to prompt viewers to find out more — a prerequisite for subsequent meaningful engagement with the cause a given video is promoting. ‘Kitchen of Cheer’ and KONY are examples of social videos that effectively furthered the ultimate objectives of the causes they promoted. An initial framework for understanding the extent to which sharing a social video to help a cause constitutes a slacktivist act has been identified, and the groundwork laid for further investigation. This study has acted as a proof-of-concept for a survey-based method’s appropriateness for investigating the slacktivist question. Future studies may compound and deepen the investigations above and synthesize new variables to broaden the understanding gained here, investigating differences between different cultures, demographic profiles, attitudes, online behaviors and more. End of article


About the author

Cat Jones is a Master’s student of sustainable development at the School of Oriental and African Studies (SOAS) at the University of London. This research was conducted in her capacity at that University. She is also Product Director for Unruly, the marketing technology company that provided the data sample used in this research. Data from Our Better World of the Singapore International Foundation was used alongside Unruly data for a case study. Dr. Ben Daly, SOAS and Dr. Rhiannon MacDonnell and Dr. Caroline Weirtz, Cass Business School, provided academic advice and input.
E-mail: cat [dot] jones [dot] research [at] gmail [dot] com



I would like to thank Dr. Ben Daly, SOAS; Dr. Rhiannon MacDonnell and Dr. Caroline Weirtz, Cass Business School; and Dr. Karen Nelson-Field, University of South Australia for their academic advice and input. I would also like to thank Unruly, particularly Dr. Sarah Wood and Mr. Scott Button, for providing the data sample used in this research, and Our Better World of the Singapore International Foundation, particularly Ms. Rebecca Lim and Ms. Eelin Ong, for providing case study data.



1. Rotman, et al., 2011, p. 1.

2. Nielsen, 2010, p. 196.

3. Barrett and Leddy, 2008, p. 2.

4. Jones, 2013, slide 13.

5. Karpf, 2010, p. 151.

6. McCaugley and Ayers, 2013, p. 6.

7. Shirky, 2011, p. 7.

8. Van Laer and Van Aelst, 2010, p. 235.

9. Georgetown University’s Center for Social Impact Communication, 2011, p. 8.

10. Katz and Lazarsfeld, 2006, p. 32.

11. Baker, et al., 2010, p. 3.

12. Van Laer and Van Aelst, 2010, p. 235.

13. Asia Research Institute (ARI), 2012, p. 95.

14. Rotman, et al., 2011, p. 3.

15. Harris, 2010, p. 4.



Asia Research Institute (ARI), 2012. Methodological and conceptual issues in cyber activism research, at, accessed 19 April 2015.

N. Audette, 2013. “KONY2012: The new face of citizen engagement,” Exchange, volume 4, number 1, article 5, at, accessed 19 April 2015.

R. Baker, S.J. Blumberg, J.M. Brick, M.P. Couper, M. Courtright, J.M. Dennis, D. Dillman, M.R. Frankel, P. Garland, R.M. Groves, C. Kennedy, J. Krosnik, P.J. Lavrakas, S. Lee, M. Link, L. Piekarski, K. Rao, R.K. Thomas and D. Zahs, 2010. “Research synthesis: AAPOR Report on Online Panels,” Public Opinion Quarterly, volume 74, number 4, pp. 711–781.
doi:, accessed 24 April 2015.

D. Barrett and S. Leddy, 2008. “Assessing creative media’s social impact,” at, 8 February 2014.

G.J. Borradaile, 2003. Statistics of Earth science data: Their distribution in time, space, and orientation. Berlin: Springer.

R.E. Botsch, 2011. “Chapter 12. Significance and measures of association,” at, 29 June 2014.

Business Insider, 2013. “The rise Of social video: How social media is creating new winners In online video” (17 June), at, accessed 8 February 2014.

K. Calkins, 2005. “Hypothesis testing” (25 July), at, accessed 13 July 2014.

H.S. Christensen, 2011. “Political activities on the Internet: Slacktivism or political participation by other means?” First Monday, volume 16, number 2, at, accessed 1 February 2014.

Cisco, 2013. “Cisco Visual Networking Index: Forecast and methodology, 2012–2017,” at, accessed 6 February 2014.

J. Cohen, 1988. Statistical power analysis for the behavioral sciences. Second edition. Hillsdale, N.J.: L. Erlbaum Associates.

Georgetown University. Center for Social Impact Communication, 2011. “Dynamics of cause engagement,” at, accessed 1 February 2014.

E.R. Harris, 2010. “Youth, the Internet, pop culture, and other frivolous things: How ‘slacktivist’ is today’s youth activism?” (7 June), at, accessed 19 April 2015.

Invisible Children, 2012. “Year in review: Results,” at, accessed 20 January 2014.

C. Jones, 2013. In: L. Stampler, “How to make a video go viral — Based on the variables in this algorithm,” Business Insider Australia (11 March), at, accessed 8 February 2014.

D. Karpf, 2010. “Measuring the success of digital campaigns,” In: M. Joyce (editor). Digital activism decoded: The new mechanics of change. New York: International Debate Education Association, pp. 151–164.

E. Katz and P.F. Lazarsfeld, 2006. Personal influence: The part played by people in the flow of mass communications. Second edition. New Brunswick, N.J.: Transaction Publishers.

B. Keeter, 2012. “Kony 2012: Lessons for the Obama administration,” Foreign Policy (18 April), at, accessed 16 February 2014.

B. Kosinski, 2014. “#IceBucketChallenge: Why you’re not really helping,” Huffington Post (7 August), at, accessed 21 December 2014.

K. Kristofferson, K. White and J. Peloza, 2014. “The nature of slacktivism: How the social observability of an initial act of token support affects subsequent prosocial action,” Journal of Consumer Research, volume 40, number 6, pp. 1,149–1,166.
doi:, accessed 24 April 2015.

Y.-H. Lee and G. Hsieh, 2013. “Does slacktivism hurt activism? The effects of moral balancing and consistency in online activism,” CHI ’13: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, pp. 811–820.
doi:, accessed 24 April 2015.

P.V. Marsden and K.E. Campbell, 1984. “Measuring tie strength,” Social Forces, volume 63, number 2, pp. 482–501.
doi:, accessed 24 April 2015.

E. Morozov, 2009. “The brave new world of slacktivism,” Foreign Policy (19 May), at, accessed 1 February 2014.

C. Neumayer and J. Schoßböck, 2011. “Political lurkers? Young people in Austria and their political life worlds online,” CeDEM11: Proceedings of the International Conference for E-Democracy and Open Government, pp. 131–143.

R.K. Nielsen, 2010. “Digital politics as usual,” In: M. Joyce (editor). Digital activism decoded: The new mechanics of change. New York: International Debate Education Association, pp. 193–196.

Our Better World (OBW), 2014. E-mail correspondence.

P.M. Podsakoff, S.B. MacKenzie, J. Lee and N.P. Podsakoff, 2003. “Common method biases in behavioral research: A critical review of the literature and recommended remedies,” Journal of Applied Psychology, volume 88, number 5, pp. 879–903.
doi:, accessed 24 April 2015.

D. Rotman, S. Vieweg, S. Yardi, E. Chi, J. Preece, B. Shneiderman, P. Pirolli and T. Glaisyer, 2011. “From slacktivism to activism: Participatory culture in the age of social media,” CHI EA ’11: CHI ’11 Extended Abstracts on Human Factors in Computing Systems, pp. 819–822.
doi:, accessed 24 April 2015.

C. Shirky, 2011. “Political power of social media: Technology, the public sphere, and political change,” Foreign Affairs, at, accessed 17 January 2014.

M.M. Skoric, 2012. “What is slack about slacktivism?” In: Asia Research Institute (ARI). Methodological and conceptual issues in cyber activism research, pp. 77–92, at, accessed 19 April 2015.

Statista, 2014. “Percentage of Internet users who watch online video content on any device in 2014, by country,” at, accessed 14 December 2014.

A. Taylor, 2014. “Is #BringBackOurGirls helping?” Washington Post (6 May), at accessed 21 December 2014.

Unruly, 2014. “Unruly ShareRank,” at, accessed 11 February 2014.

Unruly, 2013. “Unruly social video report — Q3 2013,” at, accessed 8 February 2014.

Unruly, 2012. “What is social video advertising?” at, accessed 8 February 2014.

J. Van Laer and P Van Aelst, 2010. “Cyber-protest and civil society: The Internet and action repertoires in social movements,” in: Y. Jewkes and M. Yar (editors). Handbook of Internet crime. Cullompton: Willan Publishing, pp. 230–254.

D. Waterhouse, 2012. “Kony 2012: The video that shook the world,” at, accessed 11 February 2014.


Editorial history

Received 1 February 2015; revised 20 April 2015; accepted 21 April 2015.

Copyright © 2015, First Monday.
Copyright © 2015, Cat Jones. All Rights Reserved.

Slacktivism and the social benefits of social video: Sharing a video to ‘help’ a cause
by Cat Jones.
First Monday, Volume 20, Number 5 - 4 May 2015

A Great Cities Initiative of the University of Illinois at Chicago University Library.

© First Monday, 1995-2017. ISSN 1396-0466.