Content uploaded by Liran Shan
Author content
All content in this area was uploaded by Liran Shan on Dec 22, 2014
Content may be subject to copyright.
13th International Public Communication of Science and Technology Conference
5-8 May 2014, Salvador, Brazil
Exploring online mechanisms of public engagement in risk communication: Insights
from the European FoodRisC project
Liran Christine Shan
University College Dublin
christine.shan@ucd.ie
Áine Regan
University College Dublin
aine.regan@ucd.ie
Patrick Wall
University College Dublin
patrick.wall@ucd.ie
Áine McConnon (Corresponding author)
University College Dublin
aine.mcconnon@ucd.ie
Abstract
Public engagement is advocated in risk communication and can occur at different levels
including public consultation and participation. Such engagement has traditionally been
achieved by mechanisms such as focus groups or citizen juries; however, the growth in online
interactive media platforms and tools has provided new opportunities. Drawing on research
outcomes from the EU-funded FoodRisC (Food Benefit & Risk Communication) project, this
article aims to demonstrate online opportunities for public engagement. First, the mining and
analysis of user-generated content online has provided a new mechanism for public
consultation, assisting communicators to understand public concerns, opinions and
information dissemination behaviours in different scenarios. Second, interactive information
deliberation tools exemplified by VIZZATATM have created a new avenue for public
participation, allowing researchers and practitioners to encourage and understand public
13th International Public Communication of Science and Technology Conference
5-8 May 2014, Salvador, Brazil
deliberations over risk information. Collectively, the findings from the FoodRisC project
highlight the opportunities and challenges for online mechanisms of public engagement in
risk communication.
Introduction
The process of communicating risks relating to food has traditionally assumed a top-
down approach, with information flowing from authorities and scientists to the public in a
one-way manner (Gurabardhi, Gutteling, & Kuttschreuter, 2005). This approach alone cannot
guarantee effective communication as it overlooks the public’s active role in receiving risk
information and fails to account for their information needs, concerns and opinions
(McCarthy & Brennan, 2009). To encourage a two-way, interactive communication process,
increased public engagement has been widely advocated. It is suggested that public
engagement can occur on a spectrum from lesser to greater engagement, with three levels
defined: public communication, public consultation, and public participation, highlighted in
Figure 1 (Rowe & Frewer, 2005). Public communication mirrors the traditional one-way
form of communicating to the public. With public consultation, information is conveyed from
the public to the risk communicator, and the information is believed to represent public
opinions’ on the risk in question. With public participation, information is exchanged
between the public and official communicators with dialogue and deliberation occurring.
Figure 1. Different levels of public engagement determine the flow of science and risk
information between the communicator and the public. Adapted from (Rowe & Frewer, 2005)
Whilst there is widespread encouragement for public engagement in risk
communication, it has been difficult to integrate into practice because of uncertainties over
the most appropriate techniques, processes, and instruments (i.e. mechanisms) (Rowe &
13th International Public Communication of Science and Technology Conference
5-8 May 2014, Salvador, Brazil
Frewer, 2005). The development of online communication platforms has opened new
avenues and mechanisms for public engagement (Min, 2007). Online platforms are viewed as
a valuable tool for informing, consulting, and deliberating with the public, and may provide a
more accessible and resource-effective way of engaging the public. This paper aims to
highlight the potential for online mechanisms of engagement by considering empirical
examples from the EU-funded FoodRisC (Food Benefit & Risk Communication) project
(www.foodrisc.org) (Barnett et al., 2011). This paper first considers online mechanisms for
public consultation, specifically, the mining and analysis of user-generated content online. It
then considers online opportunities for public participation focusing on the potential for
interactive information deliberation through an online tool called VIZZATATM.
Mining & analysis of user-generated content online – new mechanism for public
consultation
Web 2.0 technologies have enabled the general public to easily generate and share
information. The value of user-generated content lies in the fact that it reflects ecologically
valid, spontaneous, unsolicited opinions not affected by study demand or researcher bias
(Regan et al., 2014). The information can be used to inform risk communicators on public
opinion and societal discourse of a given topic.
In terms of data mining, many mainstream media agencies have incorporated
comments sections into their news sites, which offer an easily obtainable source for
communicators to assess public opinions and reactions. Example 1 below illustrates how to
draw on the value of online comments. Social media such as Twitter, Facebook and blogs is
another valuable source. However due to its heterogeneity and massive size, the mining and
analysing of social media postings is not that straight-forward, typically including several
steps as follows (Gaspar et al., 2014). First is the selection of keywords and keyword
combinations that are used as search phrases. These usually cover formal references of the
study subject, lay terms, and associated words to achieve the best balance between data
completeness and accuracy. The next step is data retrieval aided by online tools or software.
Examples of such tools are search services of popular sites (e.g. Twitter API, Facebook API),
self-developed open source information surveillance systems (Chew & Eysenbach, 2010),
and commercial social media monitoring tools (e.g. Hootsuite, Radian 6, MeltwaterBuzz)
(Shan et al., 2013). The data are then analysed using different techniques such as content
analysis, sentiment analysis and computer facilitated automatic analysis and visualization
13th International Public Communication of Science and Technology Conference
5-8 May 2014, Salvador, Brazil
depending on the research objectives. If undergoing manual analysis, sampling and the
removal of irrelevant postings is usually necessary to reduce the data size to a manageable
level. Example 2 & 3 below aim to illustrate the exploration of social media data.
Example 1: Understanding public resistance to dietary risk messages through the lens of
online comments It has been recognized that public communication around chronic diet
related health risks is a challenging task, since such risks are often met with a less than
expected level of concern (Perez-Cueto & Verbeke, 2012). To understand lay negotiations of
and resistance to dietary risks, an analysis was carried out on a dataset of 959 comments left
in reply to two articles reporting health risks associated with red meat, published on a
mainstream British news media website (Regan et al., 2014). The data underwent a manual
thematic analysis and results showed that the comments generally expressed resistance to the
risk message, with two strategies being used: 1) deploying rules of thumb (e.g. everything in
moderation, humans have always eaten red meat, etc.); 2) attributing risk intensification to
media, scientists and other stakeholders (e.g. irresponsible media amplifying low risk, flawed
science without acknowledging other risk factors etc.) By exploring this user-generated
content, the analysis provided an understanding of the types of arguments and strategies used
by internet users to resist dietary risk communications.
Example 2: Understanding social media postings during the 2008 Irish dioxin crisis The rise
of social media has made public discourse of food contamination incidents more visible than
it used to be. Using Radian 6, a social media information aggregator, mentions of the 2008
Irish dioxin crisis (Shan et al., 2013) were retrieved to help investigate internet users’
concerns related to this crisis as expressed on social media. Data was coded on publishing
date, sourcing practice, story topic and use of tone. It was found that on social media, people
mainly focused a limited range of topics including governments’ handling of the crisis, global
reactions, dioxin related health facts, and the impact on people’ lives, with less attention on
topics such as the cause of the incident, the influence on the Irish food industry, etc. 42% of
social media postings expressed negative sentiments. Results also indicated that social media
users’ attention faded very quickly. This implies the importance of timely communication if
one wants to capture attention on social media during food crises.
13th International Public Communication of Science and Technology Conference
5-8 May 2014, Salvador, Brazil
Example 3: A psychological analysis of Twitter messages during the 2011 European E.coli
outbreak. Using the same social media information aggregator, researchers tracked tweets in
Spain during the 2011 European E.coli outbreak to characterise citizens’ coping strategies. It
was found that public concerns peaked every time when the food sources of the pathogen
were updated (Gaspar et al., 2014). Furthermore, coping strategies expressed on Twitter were
dynamic, flexible and social, with a predominance of strategies including accommodation
(e.g. adjustment to the options that were already available), information seeking (e.g.
studying/reading information about the threat) and opposition (e.g. refusing to cooperate,
active non-compliance, or doing the opposite of what is requested or expected).
Opportunities and challenges in mining and analysing user-generated data online
The examples presented demonstrate that online data has opened a new path to understanding
public opinions and concerns. However, one needs to be aware that the information generated
by internet users is exponentially increasing. Manual sampling and analysis is still useful for
investigating specific research questions and gaining in-depth insights into data, however
there appears to be a trend towards computer assisted auto-analysis and data visualization. In
addition to the very basic content and sentiment analysis, computers now allow researchers to
more easily map and identify things such as the most influential speakers, the most
disseminated web-links, or interactions between social media users, etc. While keeping
oneself updated about new technologies and resources in this area, researchers and
communicators should keep in mind the limitation of online data in terms of representing
ideas from the general public.
VIZZATATM, an interactive online deliberation tool for public participation
To facilitate deliberation and engagement with risk issues, public participation
mechanisms have, until recently, largely taken the form of consensus conferences, citizens’
juries, and public debates. Online mechanisms of engagement differ from offline mechanisms
in that there is little opportunity for face-to-face interaction with the former. However, there
could be advantages to the written and asynchronous nature of the online communication
platform – it may support more reflexive, rational, and argumentative discussions (Halpern &
Gibbs, 2013). People can take more time to formulate a response and the quality of the reply
should reflect a more coherent argument and allow for a more rational-critical form of debate,
all important features necessary for public participation (Blom, Carpenter, & Bowe, 2011;
13th International Public Communication of Science and Technology Conference
5-8 May 2014, Salvador, Brazil
Halpern & Gibbs, 2012). The online deliberation platform VIZZATATM (www.vizzata.com)
offers interesting potential in this regard. This platform was developed in part within the
FoodRisC project to explore citizen engagement and deliberation of information in the form
of an asynchronous dialogue between participants and the research team (Barnett, Fife-Schaw,
Shepherd, Timotijevic, & Fletcher, 2008). The VIZZATATM process consists of several steps
(See Figure 2). In step 1, researchers invite participants to the VIZZATATM study webpage
and present information to them in bite-sized chunks (called ‘content testers’) to elicit
questions and comments. The content testers can be text or non-text based information, such
as news reports, images, video clips, audio clips, etc, and they are presented in a sequence of
information pages. Researchers can also include a survey questionnaire to ask specific
questions if they wish. In step 2, participants read the information and provide voluntary
questions and comments by using the ‘Ask a question’ and ‘Make a comment’ dialogue box
at the end of each content tester page. They are also asked to fill out the questionnaire if there
is one. In step 3, researchers respond to all questions participants have submitted and the
system sends responses to participants through email. In step 4, researchers invite participants
to VIZZATATM again to capture their final reflections. New content testers and
questionnaires can be added in this step as well. As VIZZATATM is designed as a dialogue
flow between two ends of the communication, options exist for further interactions.
Figure 2. The workflow of a standard VIZZATATM study
13th International Public Communication of Science and Technology Conference
5-8 May 2014, Salvador, Brazil
Example 1: Eliciting and influencing lay sense-making around synthetic meat. Public
engagement with emerging technology at its innovation stage is very important, because it
can help shape innovation processes in tune with wider public attitudes and minimize
disruptive public controversy (Macnaghten, Kearnes, & Wynne, 2005). In 2012, an online
VIZZATATM study and a number of focus groups were carried out in 3 European countries
(UK, Portugal, and Belgium) to explore how the public make sense of emerging food
technology, where synthetic meat was used as an example (Marcu et al., 2014). A 2-minute
long YouTube video, ‘Would You Eat Synthetic Meat?’, produced for the Royal Institution
of Australia was used to elicit questions and comments. Seventy consumers participated in
the online study showing various strategies to make sense of synthetic meat (Marcu et al.,
2014). Researchers found that the anchoring to GMOs, metaphors like ‘Frankenfoods’ and
common phrases like ‘playing God’ closed off debates around potential applications of
synthetic meat, whereas asking factual and rhetorical questions about it, weighing up its risks
and benefits, etc. enabled a consideration of synthetic meat’s possible implications for
agriculture, environment, and society.
Example 2: Consumer reflections on communications of the horse DNA contamination
incident. On 15th January 2013, the Food Safety Authority of Ireland (FSAI) announced that
they had found horse DNA in some beef burger products. The news quickly went viral and it
soon became a public issue throughout Europe. To understand public deliberations of and
reaction to different stakeholders’ communications in real time, a VIZZATATM study was
created and launched on the 19th January, 2013. The content testers included announcements
from food safety authorities in Ireland and the UK, an overview of the media reports at that
time, a YouTube video showing a government official explaining the incident, and a food
retailer’s public apology letter. Over the following 9 days, 44 meat consumers from the UK
and Ireland attended and finished the study. In total they left 348 questions and comments –
indicating a considerable level of engagement with and deliberation over the presented
material. Results suggested that consumers needed clarifications on issues in relation to
traceability in the food chain, and food labelling and testing (Barnett, 2013). There was very
little evidence of health risk concerns. Such knowledge gave food communicators a clearer
direction to meet public concerns. Regarding the study itself, participants found it engaging,
informative and interactive, and they valued the opportunity to freely express legitimate
concerns.
13th International Public Communication of Science and Technology Conference
5-8 May 2014, Salvador, Brazil
Advantages and implications of VIZZATATM. The presented examples illustrate the flexible
applications of VIZZATATM to encourage and explore deliberation on different risk topics.
The unique mechanism of VIZZATATM in terms of eliciting questions and comments without
facilitator bias and fostering a dialogic communication mode enables communicators to
observe and influence the process of sense making. This has strong implications in testing
and improving communications and engaging the public into information deliberation. Other
advantages of VIZZATATM include increased speed and reduced cost compared with
conventional methods, for example the horsemeat adulteration study was conducted at the
very time when the investigation was still ongoing, providing data on the immediate reactions
and sense-making processes of these consumers. This highlights its promising value in
emergent situations. Furthermore, VIZZATATM allows for continuous dialogue, and thus it
could be used for exploring how people’s reactions change over time. On the side of
limitation, due to the efforts needed to enable two-way conversation, available resources may
limit the number of participants taking part in a VIZZATATM study. That said, the aim in
such situations can be argued not to establish representativeness of the larger public, but to
provide an in-depth investigation of the reasoning and concerns of consumers on a given
topic. It’s believed that VIZZATATM is a promising online mechanism for public
participation.
Conclusion
Undoubtedly, public engagement involving bottom-up, two-way approaches are
essential to the success of communications of risk, science and technologies. The growth of
social media and online platforms has changed the landscape of communication and created
new opportunities for public consultation and participation – two important dimensions of
public engagement. Based on research outcomes from the FoodRisC project, the current
paper demonstrated some of the opportunities and challenges facing this area, with particular
focus on how to make the best use of user-generated content online, and how to foster an
online dialogue between information sponsors and the public around food risk topics. While
the examples illustrated in this paper relate to food risk scenarios, the methodologies and
tools discussed have application to the wider spectrum of science communication.
13th International Public Communication of Science and Technology Conference
5-8 May 2014, Salvador, Brazil
Acknowledgement
The work reported in this paper was supported by the FoodRisC project under the 7th
Framework Programme (CORDIS FP7) of the European Commission (grant agreement no.
245124). The authors would like to acknowledge all project partners for their excellent and
inspiring work.
References
Scientific periodic article
Barnett, J., McConnon, A., Kennedy, J., Raats, M., Shepherd, R., Verbeke, W., . . . Wall, P.
(2011). Development of strategies for effective communication of food risks and benefits
across Europe: Design and conceptual framework of the FoodRisC project. BMC Public
Health, 11, 308-316.
Blom, R., Carpenter, S., & Bowe, B. (2011). No Comment: The negative effects of online
discussion dominators on E-democracy. Paper presented at the International Political Science
Association-European Consortium for Political Research Joint Conference, Sau Paulo, Brazil.
Chew, C., & Eysenbach, G. (2010). Pandemics in the age of Twitter: Content analysis of
Tweets during the 2009 H1N1 outbreak. PLoS One, 5(11), e14118.
Gaspar, R., Gorjao, S., Seibt, B., Lima, L., Barnett, J., Moss, A., & Wills, J. (2014). Tweeting
during food crises: A psychosocial analysis of threat coping expressions in Spain, during the
2011 European EHEC outbreak. International Journal of Human-Computer Studies, 72(2),
239-254.
Gurabardhi, Z., Gutteling, J. M., & Kuttschreuter, M. (2005). An empirical analysis of
communication flow, strategy and stakeholders' participation in the risk communication
literature 1988-2000. Journal of Risk Research, 8(6), 499-511.
Halpern, D., & Gibbs, J. (2013). Social media as a catalyst for online deliberation? Exploring
the affordances of Facebook and YouTube for political expression. Computers in Human
Behavior, 29(3), 1159-1168.
13th International Public Communication of Science and Technology Conference
5-8 May 2014, Salvador, Brazil
Macnaghten, P., Kearnes, M. B., & Wynne, B. (2005). Nanotechnology, governance, and
public deliberation: What role for the social sciences? Science Communication, 27(2), 268-
291.
Marcu, A., Gaspar, R., Rutsaert, P., Seibt, B., Fletcher, D., Verbeke, W., & Barnett, J. (2014).
Analogies, metaphors, and wondering about the future: Lay sense-making around synthetic
meat. Public Understanding of Science, (in press). doi: 10.1177/0963662514521106
McCarthy, M., & Brennan, M. (2009). Food risk communication: Some of the problems and
issues faced by communicators on the Island of Ireland (IOI). Food Policy, 34(6), 549-556.
Min, S. J. (2007). Online vs. face-to-face deliberation: Effects on civic engagement. Journal
of Computer-Mediated Communication, 12(4), 1369-1387.
Perez-Cueto, F. J. A., & Verbeke, W. (2012). Consumer implications of the WCRF's
permanent update on colorectal cancer. Meat Science, 90(4), 977-978.
Regan, A., Shan, L., McConnon, A., Marcu, A., Raats, M., Wall, P., & Barnett, J. (2014).
Understanding disparate risk views: What's being said about dietary health risks in online
comments sections? Work paper.
Rowe, G., & Frewer, L. J. (2005). A typology of public engagement mechanisms. Science
Technology & Human Values, 30(2), 251-290.
Shan, L., Regan, A., De Brun, A., Barnett, J., van der Sanden, M. C., Wall, P., & McConnon,
A. (2013). Food crisis coverage by social and traditional media: A case study of the 2008
Irish dioxin crisis. Public Understanding of Science, (in press). doi:
10.1177/0963662512472315
Internet article:
Barnett, J. (2013). Consumer perceptions of the horse meat buger incident: A VIZZATA case
study. Available at: http://www.vizzata.com/docs/vizzata_horsemeat_burgers.pdf
[Accessed:19 March 2014]
13th International Public Communication of Science and Technology Conference
5-8 May 2014, Salvador, Brazil
Barnett, J., Fife-Schaw, F., Shepherd, R., Timotijevic, J., & Fletcher, D. (2008). Online
deliberative engagement: A pilot study. A report for the Wellcome Trust. London: The
Wellcome Trust. Available at: http://www.brooklyndhurst.co.uk/online-deliberative-
engagement-a-pilot-study-_28 [Accessed: 10 March 2014]
