Insight from the horsemeat scandal. Exploring the consumers’ opinion of tweets toward Tesco
| DOI | https://doi.org/10.1108/IMDS-10-2015-0417 |
| Date | 11 July 2016 |
| Pages | 1178-1200 |
| Published date | 11 July 2016 |
| Author | Ying Kei Tse,Minhao Zhang,Bob Doherty,Paul Chappell,Philip Garnett |
| Subject Matter | Information & knowledge management,Information systems,Data management systems |
Insight from the
horsemeat scandal
Exploring the consumers’opinion of
tweets toward Tesco
Ying Kei Tse, Minhao Zhang and Bob Doherty
The York Management School, The University of York, York, UK
Paul Chappell
Department of Sociology, The University of York, York, UK, and
Philip Garnett
The York Management School, The University of York, York, UK
Abstract
Purpose –Social media has become an important part of daily interpersonal communication in
contemporary society. The purpose of this paper is to explore the attitudes of UK consumers by
identifying the hidden information in tweets, and provide a framework which can assist industry
practitioners in managing social media data.
Design/methodology/approach –Using a large-scale dataset of tweets relating to the Horsemeat
scandal of 2013, a comprehensive data analysis framework, which comprises multidimensional scaling
and sentiment analysis, alongside other methods, was applied to explore customers’opinions.
Findings –Making jokes in social media was a main trend in the tweets relating to Tesco during
the Horsemeat scandal. Consumer sentiments were overall negative and burgers were the most
mentioned product in the week-long period after the story broke. The posting of tweets was correlated
with the timing of news coverage, which indicates that the traditional media is still crucial to public
opinion formation.
Practical implications –This paper presents a progressive tweet-mining framework that can serve
as a tool for academia and practitioners in crisis management. The proposed framework indicates the
significant importance of timely categorising the topics, identifying the sentiment of tweets and
understanding the changes of consumer opinions over time in a crisis.
Originality/value –The research presented in this paper is one of the limited social media research to
focus on a UK food fraud issue and adds to the limited body of literature investigating consumer social
media use from the side of industry practitioners.
Keywords Social media, Crisis management, Food fraud incident, Twitter mining
Paper type Research paper
Introduction
Using social media to manage relationships with customers has been a focus of much
attention within industry (Agnihotri et al., 2012). According to Blackshaw and Nazzaro
(2004), in the business context, “social media is a variety of new sources of online
information that are created, initiated, circulated and used by consumers intent on
educating each other about products, brands, services, personalities, and issues”.
Practitioners strive to identify ways in which social media platforms, such as Twitte r
and Facebook, can be used to to increase the profitability (Kaplan and Haenlein, 2010).
For example, through adopting social media, companies can communicate with their
consumers for the purpose of sales and customer service (Safko, 2010), market
Industrial Management & Data
Systems
Vol. 116 No. 6, 2016
pp. 1178-1200
©Emerald Group Publishing Limited
0263-5577
DOI 10.1108/IMDS-10-2015-0417
Received 6 October 2015
Revised 4 January 2016
Accepted 17 January 2016
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/0263-5577.htm
This research is supported by BA/Leverhulme Small Research Grants SG141713.
1178
IMDS
116,6
promotion (Mangold and Faulds, 2009), and even product innovation (Berthon et al. ,
2012). In the tide of big data, the use of the social media data is also of interest within
academia. For example, Bollen et al. (2011) use aggregated Twitter data to predict the
changes in the stock market. Other researchers have tried to explore how social media
can be employed as part of an effective risk communication strategy (Veil et al., 2011).
Nevertheless, relatively few pieces of social media research have focused on public
relations in product recall crises, in particular food quality and safety issues, whic h are
extremely sensitive issues for the general public.
The horsemeat scandal (also known as Horsegate) began in January 2013 when the
Food Safety Authority of Ireland (FSAI) announced the presence of horsemeat in
burgers on the shelves of some well-known retailers, such as Tesco, Iceland, Aldi and
Lidl. The suspected horsemeat was found in extensive beef products in the European
Market, and the scandal severely dented consumer trust in the food industry. In order
to regain the consumers’trust, retailers guaranteed they would enhance scientific
testing and change their food procurement strategies. For example, Tesco announced
that it was introducing a world-class DNA testing system to respond the outbreak of
Horsegate (Tesco, 2013). Nevertheless, in the phase of post-crisis, communicating well
with consumers should be the priority for the companies. Therefore, understanding the
attitudes of consumers from social media could be the first step in establishing efficient
risk communication strategy for practitioners.
To fill the gaps in the literature and contribute to practitioners’insights into the
horsemeat scandal, this study focus on tweets about Tesco. Tesco is the largest
supermarket chain in the UK[1] and was constantly under the spotlight during the
scandal –239 newspaper articles[2] had Tesco in their headlines during the outbreak of
the horsemeat scandal. Faced with this immense negative news portrayal in both the
online and print media, the reaction of consumers towards the company at this time can
provide important insights for Tesco, and for companies that may face similar crises in
the future. In this research, we investigate consumers’opinions towards the horsemeat
scandal on Twitter. According to data from September 2015, Twitter is the ninth most
visited website in the world (ALEXA, 2015) so provides a way for us to explore the
opinions from customers. The body of data provided free of charge by consumers
provides a valuable opportunity to improve public opinion tracking and customer
relationship management (Mostafa, 2013). Jansen et al. (2009) argue that microblogging
(i.e. Twitter) allows people to share brand-affecting thoughts (i.e. sentiment) anytime
and anywhere. As Twitter requires users to control the length of a blog in a short
manner (like a title of newspaper headline), it is easy to read and create ( Jansen et al.,
2009). It is also open access, meaning researchers can capture the textual data for in
depth analysis However, it is an impossible mission to manually analys e a large
amount of text and extract useful information without text mining (TM) techniques
(Liau and Tan, 2014).
TM is a new research method to overcome the challenges in analysing overloaded
unstructured texts by complementarily adopting different techniques, such as data
mining, machine learning, natural language processing, information retrieval, and
knowledge management (Feldman and Sanger, 2007). For example, sentiment analysis
(SA) has been widely adopted to analyse microblogging using computational
linguistics (Liau and Tan, 2014; Mostafa, 2013; Ghiassi et al., 2013; Kontopoulos et al.,
2013; Paltoglou and Thelwall, 2012). By applying SA, the sentiment of a large number
of tweets can be automatically identified, classifying them as positive, neutral, and
negative. This technique has an important industrial application in that a given
1179
Insight from
the horsemeat
scandal
To continue reading
Request your trial