Content uploaded by Ray Qing Cao
Author content
All content in this area was uploaded by Ray Qing Cao on Feb 20, 2018
Content may be subject to copyright.
1 23
Information Systems Frontiers
A Journal of Research and Innovation
ISSN 1387-3326
Inf Syst Front
DOI 10.1007/s10796-017-9762-2
Using sentiment analysis to improve supply
chain intelligence
Ajaya Kumar Swain & Ray Qing Cao
1 23
Your article is protected by copyright and all
rights are held exclusively by Springer Science
+Business Media New York. This e-offprint is
for personal use only and shall not be self-
archived in electronic repositories. If you wish
to self-archive your article, please use the
accepted manuscript version for posting on
your own website. You may further deposit
the accepted manuscript version in any
repository, provided it is only made publicly
available 12 months after official publication
or later and provided acknowledgement is
given to the original source of publication
and a link is inserted to the published article
on Springer's website. The link must be
accompanied by the following text: "The final
publication is available at link.springer.com”.
Using sentiment analysis to improve supply chain intelligence
Ajaya Kumar Swain
1
&Ray Qing Cao
2
#Springer Science+Business Media New York 2017
Abstract Analysis of comments and opinions expressed in
social media can be used to gather additional intelligence via
market research information to better predict consumer behav-
ior. The area of Bopinion mining^, particularly sentiment anal-
ysis, aims to find, extract, and systematically analyze people’s
opinions, attitudes and emotions towards certain topics.
Performance of a supply chain is closely associated with the
level of trust, collaboration, and information sharing among
its members. In this paper, using textual Bsentiment
analysis^, we explore the relationship between elements of
social media content generated by supply chain members
and performance of supply chain. In particular, we identify
specific elements of member generated supply chain related
content on social media such as: information sharing, col-
laboration, trust, and commitment to determine their associ-
ation with supply chain performance. We find information
sharing and collaboration to be positively associated with
supply chain performance, and these findings are consistent
with previous reports in supply chain literature. In addition,
ours is one of the first attempts to use sentiment analysis to
analyze social media content in a supply chain context. The
findings indicate that supply chain members value the shar-
ing of relevant information and collaborative contents on
social media as such efforts improve individual and overall
supply chain performance. The results of this study should
prove useful to other studies that utilize social media in a
supply chain context, and to improve supply chain manage-
ment strategies.
Keywords Social media .Intelligent decision making .Social
network analysis .Sentiment analysis .And supply chain
management
1 Introduction
Advantages of social media over traditional media lie in its:
magnitude of usage and reach, ease of use, and pace of trans-
mission of information (Gallaugher and Ransbotham 2010;
Harris et al. 2013; Luo et al. 2013). Given these advantages,
social media has magnified the power of online communica-
tion by facilitating multi-way, real-time, convenient, and easy
information sharing between producers, consumers, and users.
Decision makers use social media for gathering information or
opinions about products and services (Ramos and Young
2009; Zhang and Piramuthu 2016), understanding how stake-
holders perceive their actions and performance and how these
perceptions are reflected in various social media channels.
Consumers consider social media generated product/service
related information to be more reliable than those disseminat-
ed by corporations using traditional media (Mangold and
Faulds 2009; Chang and Jung 2015). For obvious reasons,
social media channels have experienced explosive growth in
recent years (van Dam and van de Velden 2015; Harris et al.
2013).
The use of social media channels transcends disciplinary
boundaries. Social media applications can be found in mar-
keting (identifying new marketing channels, and increasing
*Ajaya Kumar Swain
aswain@stmarytx.edu
Ray Qing Cao
caoq@uhd.edu
1
Finance and Quantitative Management Department, Greehey School
of Business, St. Mary’s University, San Antonio, Texas 78228, USA
2
Management, Marketing, and Business Administration Department,
Davies College of Business, University of Houston-Downtown,
Houston, Texas 77002, USA
Inf Syst Front
DOI 10.1007/s10796-017-9762-2
Author's personal copy
customer loyalty); supply chain management (improving
product-service quality, delivery, flexibility); and information
systems (gathering new sources of business intelligence)
(Madison 2012). Social media can also be used for informa-
tion-sharing, relationship-building, and improving communi-
cation, coordination, and performance (Madison 2012). Best
Buy and Dell have successfully used social media to reduce
costs, motivate employees, improve internal communications,
and stimulate innovation (Shipilov 2012). However, to date,
there is scant empirical literature on how social media affects
communication/coordination between organizations (Aral et al.
2013;Douetal.2013; Miller and Tucker 2013;Madison2012;
Bharati et al. 2014). Efficient communication/coordination is
critical in supply chain management (SCM) as they are directly
linked to supply chain performance.
As researchers from the McKinsey Global Institute predict,
use of socialmedia within large companies could contribute as
much as $1.3 trillion in annual value to the U.S. economy,
more and more organizations are trying to leverage the power
of social media technologies to aid organizational knowledge
sharing (Leonardi 2015).Social media allows organizations to
improve performance by monitoring and analyzing consumer
conversations. However, as an HBR report (2010) indicated,
three out of four companies were unaware of where their most
valuable customers were talking about them. In addition, ap-
proximately 33% of firms did not measure the effectiveness of
social media, while less than 25% were using social media
analytic tools. Even more telling, less than 10% of companies
integrate social media into their marketing activities and over-
all operations. This may be due in part to the relative novelty
of social media and the availability of few established methods
to gauge its effectiveness, and on how they integrate/align
social media activities with corporate profit strategies (HBR
Report 2010). Many companies overinvest in social media
and in unused software generating wasted resources (Baker
2009). Furthermore, as social media datasets are vast, noisy,
distributed, unstructured, and dynamic in nature, the cost of
collecting, mining, and deriving meaningful information from
this data remains a challenge (Gundecha and Liu 2012;HBR
report 2010;SwainandCao2013,2014a). Until very recently,
the situation has not quite improved. While most experts be-
lieve that there is considerable potential for firms to utilize
social media data to refine their corporate strategies, the
Data & Analytics Report by MIT Sloan Management Review
and SAS (Ransbotham et al. 2016) finds that Banalytics is still
a mainstream idea, but not a mainstream practice. Only a
handful of companies have a strategic plan for analytics or
are executing a strategy for what they hope to achieve with
analytics^. A C-Suite Study (Quesenberry 2016) recently re-
ported that almost half of all chief marketing officers believed
that they are not prepared to manage the current challenges of
social media citing reasons of uncertainty about management,
strategies, and integration of social media. To date, social
media research in the area of supply chain management is
minimal and need further investigation (Lam et al. 2016;
Casemore 2012;O'leary2011).
While social media provides opportunities for users to
build valuable Bsocial capital^that can be used to improve
supply chain performance, they entail significant investment
that need to be justified financially (Deans 2011). Social me-
dia content has been defined as B…amixture of fact and opin-
ion, impression and sentiment, founded and unfounded tid-
bits, experiences, and even rumor…^(Blackshaw and
Nazzaro 2006, p. 4). In this study, we explore the relationship
between social media and supply chain performance by em-
pirically testing whether social media usage by supply chain
partners in terms of frequency, volume, polarity, and content,
is associated with supply chain performance. As most extant
studies use social media data that are either cross sectional in
nature or are based on self-reports, the objectivity and gener-
alizability of results are compromised (e.g. Gretzel and Yoo
2008). Hence in this study, we analyze the word of mouth
(WOM) data related to supply chain management and collect-
ed from various social media outlets to address the proposed
research question. By examining opinionated information on
social media generated by supply chain partners from a cross
section of industries, we provide empirical evidence on the
links between social media use and supply chain performance.
We find that WOM effects (social contagion) are positively
correlated with supply chain performance.
The paper is organized asfollows. In Section 2, the relevant
literature on social media and supply chain are discussed.
Section 3presents the theoretical basis for the study, and is
followed by research methodology and testable hypotheses in
Section 4. Results are presented and discussed in Section 5,
implications are presented in section 6followed by limita-
tions, further research directions, and conclusions, in
Section 7.
2 Literature review
2.1 Social capital and social media
The term Bsocial capital^has its roots in the field of sociology
and political science. Hanifan (1916) coined this term in his
study of rural school community centers. Over the years, the
social capital concept has found its acceptance and applica-
tions in social sciences, economics, and organizational and
management sciences (Chang et al. 2016). Social capital can
be gained through social networks of a person, but the advan-
tage is generated to the individual from the network (Portes
1998). Social capital is linked to (and derived from) social
relations. It is not fully interchangeable (Coleman 1988)and
also not easily transferable (Nahapiet and Ghoshal 1998). It
does not exist without social interactions and networks. Social
Inf Syst Front
Author's personal copy
capital derives its novelty and heuristic power from two
sources: the positive consequences of sociability and its link
to a larger discussion of capital (Portes 1998).
Social capital has a wide range of definitions focusing on
two aspects: social networks and the resources embedded
within those networks. The main proposition of social capital
theory is that networks of relationships are a valuable resource
for social action. The basic premise of social capital theory is
the benefit that an actor, individual, or collective, can reap
from those social relations and their embedded resources.
1
Within the context of social capital theory, the social proper-
ties applicable to social media are reach, engagement, and
influence (Sofia et al. 2012). Reach refers to the degree of
effective dissemination of certain content or potential spread
of the network, operationalized by the number of individuals
reached, proximity, and propagation speed. Engagement cor-
responds to the extent of participation and involvement of a
specific individual in the network. Engagement can be mea-
sured by time spent with a specific individual, inter-contact
times, or reciprocity of contacts. Finally, influence refers to the
degree of attention/mobilization that the individual can gener-
ate within other network members.
2.2 Supply chain and social media
Supply chains are complex socio-technical systems where
both technical and social factors play important roles. The
technical factors are system dominated and deal with techno-
logical and supply chain structural issues such as logistics,
information systems, and supply chain performance (Li et al.
2015) whereas social factors are human focused and deal with
social relationships among various supply chain partners
(Burgess et al. 2006). While the technical factors are dealt with
by formal mechanisms set up by organizations, the social fac-
tors such as reciprocity and mutual trust relate to the social
system of the supply chain. Social factors and technical factors
are equally important in the context of supply chains (Burgess
and Singh 2012). Historically, the supply chain process has
emphasized the Bhard^topics such as supply chain modeling
issues that generally relate to the incorporation of the latest
available technology in this area. Snub and Stonebraker
(2009) argue that not enough attention has been focused on
using Bsoft^human resource activities and organization vari-
ables to enhance supply chain strategies/success. These soft
constructs of supply chain management (SCM) are Bmore
vulnerable to misalignment, because they are more closely
embedded in the socio-cultural context of the recipient
culture^(Brannen 2004, p. 597). Although many researchers
stress the importance of trust and collaboration within supply
chains, the social aspects of SCM has not been investigated in
depth (Burgess et al. 2006). Moreover, with opportunities for
improvement in supply chain performance via break through
technical improvements are on a decline, the focus on enhanc-
ing supply chain performance by improving relationships
among supply chain partners is on the rise (Burgess and
Singh 2012). Therefore, exploring the links between social
media content and various functions of supply chain can lead
to a profitable line of research (Aral et al. 2013; Benbya and
Van A l st y ne 2010). For instance, such research can unearth
interesting links between social media content and supply
chain performance, and can be useful to academics and prac-
titioners alike (Swain 2016a).
Extant social media data has traditionally been used by the
business to customer (B2C) segment for brand promotion and
marketing products (Markova and Petkovska-Mircevska
2013), and by the business to business (B2B) segment
(Rodriguez et al. 2012) to improve sales performance.
Unfortunately, there is scant literature on the social media
usage –supply chain links. Some studies have examined the
impact of social media usage on supply chain events with the
aim of developing contexts for these events and to build rela-
tionships between supply chain partners (e.g. O’leary 2011).
In this study, and unlike prior studies, the focus is on supply
chain performance. In particular, we investigate whether so-
cial media usage of various supply chain partners is associated
with supply chain performance.
2.3 Social capital and supply chain
Several studies have used social capital and social network
theories to analyze supply chain issues. Drawing from the
social network perspective, Bernardes (2010)exploredfactors
associated with the relational embeddedness of social capital
and investigated the role of supply management on the
process. Using empirical data, he confirmed that the
relational embeddedness aspect of social capital should be
treated as a critical antecedent to performance. While this
study focused on the network of actors, our study focuses on
the ties within the network, a premise of social capital theory.
Borgatti and Li (2009) provided an overview of social net-
work analysis in a supply chain context by discussing impor-
tant concepts (such as structural holes) and the links between
centrality and the generic explanatory mechanisms. Based on
the concepts of social capital, financial capital, and relational
capital, Hammervoll (2011) studied recently formed supply
chain relationships (SCR). He extended the concept of rela-
tional capital in SCR to include financial capital and
psychological commitment and also suggested new
propositions that relate relational capital and the length of
the time that has elapsed immediately following the SCR
formation. Villena et al. (2011) evaluated how social capital
in its cognitive, relational, and structural forms affects value
creation within buyer-supplier relationships (BSR). Using
1
Some researchers (e.g. Granovetter 1983) used the term social network the-
ory instead of social capital theory. Both terms are used interchangeably.
Inf Syst Front
Author's personal copy
both primary and secondary data, they showed that there is an
inverted curvilinear relationship between social capital and
performance, implying that too little or too much social capital
can hurt performance. This study also confirmed that building
social capital in a collaborative BSR positively affects buyer
performance.
2
Finally, drawing on social capital theory,
Lawson et al. (2008) developed a model linking positive rela-
tional capital and its antecedents, supplier integration and sup-
plier closeness, to buyer performance improvements. They
provided evidence that structural capital was also positively
related to buyer performance improvements. However, this
study only considered structural and relational aspects of so-
cial capital, not the cognitive aspect which is a point of depar-
ture from our study.
2.4 Sentiment analysis
The volume of user generated content on social media has
grown exponentially in recent years. In turn, this growth has
evoked increased interest from academics and practitioners on
how to analyze and use this information creatively. However,
as manual processing of this voluminous information is cost
prohibitive, the literature has developed several automated
techniques for detection of opinion from social media gener-
ated free text. In this context, opinion mining technique refers
to the process of using the opinions, attitudes and emotions of
social media users towards firms, brands, companies, prod-
ucts, etc. (e.g. Liu 2012; Pang and Lee 2005). Sentiment anal-
ysis aims to automatically detect and classify opinions, senti-
ments and attitudes from natural language text and has been
successfully applied in the areas of data mining, information
retrieval, and natural language processing. More specifically,
sentiment analysis is an automated computational technique
that can detect and classify sentiment from text by polarity or
orientation (Balahur and Jacquet 2015). It can broadly be cat-
egorized into: knowledge-based sentiment analysis which
uses unsupervised learning techniques such as ontologies
and semantic networks to detect sentiments from text and
statistics-based sentiment analysis that uses supervised learn-
ing techniques such as machine learning and clustering to
analyze text (Xia et al. 2016). Sentiment analysis is a special
case of text mining where the text to be mined is represented
by a bag-of-words model and then classified by using several
statistical machine learning algorithms. Classification is a ma-
chine learning task of inferring a function from labeled train-
ing data, where statistical methods are applied to construct
prediction rules. Typical supervised learning algorithms in-
clude: naïve bayes classifiers, maximum entropy, support vec-
tor machines and K-Nearest neighbor learning (Poria et al.
2014;Swain2016b). Use of several semi-supervised learning
methods incorporating auxiliary unlabeled data have also been
reported (Zhang et al. 2012).
Of late, the focus of sentiment analysis research is shifting
towards social media platforms such as Twitter, Facebook,
Flickr, Blogs, Forums, Wikis, and LinkedIn. However, reports
on use of sentiment analysis in supply chain research are
scant. Hence, this research adopts a statistics-based sentiment
analysis approach that uses a Naïve Bayes classifier super-
vised learning algorithm to test the proposed hypotheses.
3 Theory and hypotheses
Social capital is the collective useful actual/potential relational
resources embedded in personal ties of individuals in social
organizations (Adler and Kwon 2002;Bourdiue1985).
Supply chain can be viewed as an organization itself since it
shares many similarities with an organization, such as overall
vision and shared mission. The concept of social capital can
then be defined within the context of the supply chain organi-
zation. It can refer to the reservoir of social capital generated
by the network of each member in a supply chain. The knowl-
edge exchanged or gained through interaction with specific
members or partners can then be viewed as a social resource
that can be used to maximize organizatonal performance.
Social capital of supply chain members can influence sup-
ply chain performance while also serving as a sustainable
source of competitive advantage (Min et al. 2008). In addition,
social capital in the supply chain plays a key role in the de-
velopment and management of buyer-supplier relationships
(Bernardes 2010) and can serve as the main source of value
creation in a supply chain (Lawson et al. 2008). Krause et al.
(2007) present evidence of how three forms of social capital,
namely, structural, relational, and cognitive capital, can bring
about performance improvements in a supply chain partner. In
turn, a partner’s socialcapital can enhance knowledge creation
in the process (Wei and Ju 2010). Next, Yim and Leem (2013)
suggest that supply chain social capital can enhance supply
chain integration which in turn improves firm performance.
Social capital also serves as a source of physical and informa-
tional resource for service supply chains (Avery and Swafford
2009). A common thread of all these studies is that social capital
can increase supply chain performance. Social profiles created
within a network can provide a collection of shared networks
that can be further strengthened for member benefit (Ellison et al.
2007;Shahetal.2001), can improve weak ties (Kenski and
Stroud 2006), and enable collective action (Valenzuela et al.
2009). This in turn leads to better supply chain performance.
Social issues arising from concerns for product and human
safety, welfare, and reputation, can potentially pose significant
supply chain related operational risks thereby affecting its re-
liability and performance (Fombrun et al. 2000;Marcusand
Goodman 1991;Rothetal.2008). Social supply chain
2
While this study considered all of the three social capital dimensions, the
context and the goal of the study is different from ours.
Inf Syst Front
Author's personal copy
relationships developed over time through formal and informal
interactions can prove important for supply chain performance.
Burgess and Singh (2012) provide case-based evidence of a
large organization achieving significant improvement in its op-
erating performance by effective use of the social system of its
key supply chain. They found that instead of focusing exces-
sively on formal governance and information technology (IT)
systems, firms can improve their operating performance in
terms of cost and time frames by working through existing
social systems. Godes (2016) observed that social interactions
of a firm can affect product quality. In sum, these studies high-
light the importance of the social system, social issues, and
social factors, in determining supply chain performance.
Recently, there has been a growing interest in understand-
ing how WOM, particularly online WOM or eWOM (which is
a form of social media) impacts sales, diffusion, and other
business performance measures (e.g. Chevalier and Mayzlin
2006; Godes and Mayzlin 2004;Trusovetal.2010;
Villanueva et al. 2008). The literature also documents evi-
dence of social media usage leading to greater product
innovation and enhanced firm performance. For example, Di
Gangi and Wasko (2009) and Gallaugher and Ransbotham
(2010) present evidence of successful use of social media by
firms such as Dell and Starbucks for developing online
communities to solicit and evaluate ideas from customers for
product development. Luo et al. (2013) argue that social me-
dia may be used to monitor and estimate customer feedback
and brand buzz leading to improved firm performance.
Rodriguez et al. (2012) report a positive relationship between
social media usage and sales processes and performance. The
basic conclusion that can be derived from this segment of the
literature is that social media use can affect a product’ssuccess
in the marketplace. As such, we posit hypothesis 1:
&H1: Social media usage by supply chain members is
positively associated with supply chain performance.
Coleman (1988) defines social capital as a set of entities that
are part of a specified social structure. These entities facilitate
some interaction among individuals within the structure. For a
supply chain, each member is part of a bigger network of mem-
bers with a common shared goal who share views, information,
and resources, and interact with other members to achieve this
shared common goal. There is a strong positive correlation be-
tween interaction and transparency levels and positive outcomes.
In this study, we examine two aspects of the usage of social
media (e.g. frequency and volume) by supply chain members.
WOM is a social phenomenon and degree of strength of social
relations can influence WOM behavior (Brown and Reingen
1987). In a network, social relations of a member with other
members could range from a Bstrong primary^contacts (such
as close contacts) to Bweak secondary^(such as lower contact
levels) contacts with acquaintances (Granovetter 1973).
Members with relatively frequent interactions between them-
selves for sharing information and resources are likely to pos-
sess stronger ties (Reingen et al. 1984). Given that individuals
tend to seek and interact with other like-minded individuals,
the strengthof these ties is alsodirectly related to the degree of
this similarity (Laumann 1966). Frequent interactions also re-
inforce similarities (Granovetter 1973) and lead to better in-
formation flow between members (Brown and Reingen 1987).
Peng and Luo (2000) provide evidence that interpersonal ties
of managers and top executives of firms improved firm per-
formance. In the study, we also expect that stronger ties gen-
erated from frequent interactions between supply chain mem-
bers will lead to better supply chain performance.
The volume of social media usage (eWOM) can be viewed
as a proxy for the intensity of the eWOM effect (Duan et al.
2008), and has been shown to positively influence product and
firm level performance. Duan et al. (2008) found weekly mov-
ie box office sales to be significantly (positively) influenced
by the volume of online posting. The literature also provides
of a positive relationship between these two variables (e.g.
Godes and Mayzlin 2004). Hence, we propose to test the
following hypotheses:
&H1a: The frequency of social media usage by supply chain
members is positively associated with Supply chain
performance.
&H1b: The volume of social media usage by supply chain
members is positively associated with Supply chain
performance.
Sharing of information among supply chain partners on
key supply chain processes such as: point of sales, adaptation
of collaborative practices such as vendor managed inventory,
and collaborative planning, forecasting and replenishment, are
positively correlated with supply chain performance (Angulo
et al. 2004; Aviv 2001). Information sharing among members
can also provide competitive advantages to members by cre-
ating customer values and reducing supply chain
costs (Sharma et al. 2013). As Inderfurth et al. (2013)note,
information sharing can reduce supply chain inefficiencies
due to information asymmetry and can control non-
cooperative behavior of supply chain members.
Information sharing is one of the essential factors that can
enhance channel-wide collaboration across the supply chain
(Ballou et al. 2000). Asymmetric information
3
between mem-
bers can negate value creation through poor collaborative ef-
forts, difficulty in dealing with market uncertainty, suboptimal
decisions, and opportunistic behavior (Simatupang and
3
Asymmetric information refers to the situation where different players in a
supply chain having differing levels of specific information on various re-
sources such as capacity, demand, and inventory status, related costs, supply
chain operations and performance.
Inf Syst Front
Author's personal copy
Sridharan 2001). These factors either as stand-alone factors or
in combination can significantly (negatively) impact supply
chain performance. In addition, inter-organizational ca-
pabilities such as information visibility and supply chain
flexibility could be augmented from a strengthened ex-
change relationship between supply chain partners
(Sahin and Robinson 2002). Increased sharing of infor-
mation together with other firm level resources can im-
prove a member’s decision-making capabilities and sup-
ply chain performance (Mabert and Venkataramanan
1998). Relational links between team members have
been found to be a significant contributor to the effec-
tiveness of information sharing (Warkentin et al. 1997).
Social media provides opportunities for real-time, conve-
nient, quality information sharing between members because
of its reach, ease of use, and fast paced transmission of infor-
mation. Hence, we expect supply chain performance to im-
prove if the supply chain partners share more information
regarding supply chain management issues on social media:
&H2: Amount of information shared by supply chain mem-
bers in social media is positively associated with supply
chain performance.
Collaboration allows parties to leverage their differ-
ences in interests, concerns, and knowledge. Online col-
laboration includes collaboration on social media facili-
tating sharing, transfer and reuse of information, exper-
tise and knowledge among online communities
(Leonardi 2015). Collaboration in a supply chain con-
text refers to the sharing of supply chain information
related to product design, product development, produc-
tion processes, logistics and distribution strategies, and
all forms of planning (Balakrishnan and Geunes 2004;
Lejeune and Yakova 2005). Collaboration encompasses
factors such as coordination, communication, relation-
ship management, trust and structure (Lin et al. 2010).
Successful collaboration between supply chain partners
can generate improved supply chain performance
through improvements in efficiency, effectiveness, prof-
itability, and stronger and long-lasting ties between part-
ners (Min et al. 2008). Supply chain management is a
combination of approaches and efforts that support effi-
cient collaboration of suppliers, producers, and cus-
tomers with the ultimate aim of achieving customer sat-
isfaction (Simchi-Levi, Kaminsky, & Simchi-Levi 2003).
The introduction of new internet technologies has
allowed partners from diverse background and locations
to collaborate effectively to coordinate their activities
and improve individual supply chain functions. Social
media can increase these benefits significantly as they
offer convenient modes of communication, and informa-
tion sharing.
Collaboration through delicate, complex social relations
requires a medium where these relations can be used effec-
tively (Suchman 1987). Socially related factors that contribute
to collaboration include: formal and informal communica-
tions, trust, motivation, and social ties (Kotlarsky and Oshri
2005; Charband and Navimipour 2016). Social media usage
can influence successful online collaboration via social rela-
tions between supply chain partners. In this study, we suggest
that greater levels of collaboration on social media between
partners will generate better supply chain performance:
&H3: Level of collaboration by supply chain members in
social media is positively associated with supply chain
performance.
Putnam (2001) defines social capital as a set of fea-
tures that assist in facilitating and coordinating actions
in structures. From a social networking perspective,
these include levels of trust, or reciprocity. Trust and
commitment are the two fundamental components of
improving social relationships that generate better coop-
eration between partners (Morgan and Hunt 1994). Trust
is often developed by individual contact, recurrent ex-
changes of information, and socialization among groups
and individuals (Child 2001). Social media and social
networking sites in particular, can strengthen weakening
ties and promote collective action based on common
interest, activities, and goals (Kenski and Stroud 2006;
Shah et al. 2001). As social capital is a positive func-
tion of increased online and offline socialization
(Kobayashi et al. 2006), social media can help build
user social capital by building trust relationships within
the network (Holsapple and Wu 2008).
Trust plays a vital role in information exchange and knowl-
edge integration, as it allows individuals to justify and evalu-
ate their decision to provide or attain more useful information.
Trust has been found to be essential to virtual community
members’intention to exchange information with other mem-
bers (Chu and Kim 2011). From social networking perspec-
tive, trust serves as an important means for consumers to eval-
uate the source and value of information, and thus has a crit-
ical influence on social media content. Thus, when users in
social media trust their social connections, their willingness to
rely on those connections is enhanced because of the connec-
tions’perceived reliability and trustworthiness (Cao et al.
2015; Burgess et al. 2011;ChuandKim2011).
Recent online and offline surveys have indicated that
internet users have more generalized trust and extensive
social networks (Cole 2000). Although there is no direct
evidence, it is conjectured that higher levels of trust/
bigger networks will also work within social media.
The within network trust is important as sharing of re-
sources generates more common resource, and this
Inf Syst Front
Author's personal copy
premise forms the cornerstone of social capital theory.
4
Hence there is a direct relationship between network
trust/reciprocity levels and the availability of common
resources/social capital for use.
5
In the current context,
trust is the degree to which partners possess the inten-
tion and capacity to work towards improvement of the
overall supply chain performance (Morgan and Hunt
1994). Trust is also viewed as the degree of buyer’s
confidence and reliance on supplier’s expertise required
to perform an activity effectively (Ganesan 1994).
Online relationships and communications are positively
associated with an individual’s social trust (Kavanaugh
et al. 2005). Social media usage has been found to be
strongly associated with maintaining or strengthening
existing offline relationships of communities (Ellison
et al. 2007). In this study, and based on the above
literature, we postulate that higher levels of partner so-
cial media interactions related to trust will generate
greater mutual trust and consequently lead to better sup-
ply chain performance:
&H4: Level of trust among supply chain members in
social media is positively associated with supply chain
performance.
Supply chain integration is positively associated with sup-
ply chain performance (Quesada et al. 2008). Stronger trust
and collaboration based relationships between partners may
lead to integration of supply chain functions such as design,
purchasing, production, distribution. Wu et al. (2004) also
document evidence that commitment of supply chain partners
can enhance the integration of supply chain management pro-
cesses. Allen and Meyer (1990) describe commitment from
three aspects: affective commitment; normative commitment;
and continuance commitment. While affective commitment
refers to the feeling of belongingness and the sense of attach-
ment to the firm, normative commitment is related to the ob-
ligation that members feel to remain with a certain firm and
builds on generalized cultural expectations. Continuance com-
mitment is perceived from originating in a lack of suitable
alternatives. Normative commitment can increase coordina-
tion and integration leading to mutually shared goals and
values. Identifying with others, gaining a sense of belonging-
ness and insight into the circumstances of others, are some of
the major reasons stated for using social networking sites.
Social media (in particular, several social networking sites)
can help create personal identity by providing multiple chan-
nels for relational feedback and peer acceptance. In this man-
ner, social media can facilitate development of affective and
normative commitment of users and can promote trust to im-
prove supply chain performance.
Commitment is a key to understand why some relation-
ships persist and other do not. Empirical evidence (Meyer
et al. 1999) suggest that the greater an individual’sneedto
maintain a relationship, the more the individual will be com-
mitted to the relationship. Relationships on socialmedia are an
extension of the users’relationships in the physical world. The
network of affiliations an individual has, the stronger the in-
dividual’s perceived relationship commitment to that commu-
nity (Ma and Chan 2014).
Based on the above reasoning, we hypothesize that higher
levels of partner talk on social media related to commitment to
common supply chain goals and objectives can improve sup-
ply chain performance:
&H5: Level of commitment of supply chain members in
social media is positively associated with supply chain
performance.
4 Methodology
This study adopts a sentiment analysis approach to test all
hypotheses. Of late, sentiment analysis has been very useful
for companies in understanding customer opinions about
products, brand perception, new product perception, and rep-
utation management (Gundecha and Liu 2012). This method-
ology is operationalized by extracting related sentiments or
opinions of customers, suppliers, and employees, from thou-
sands of unstructured documents available on social media.
For this study, we collected social media data from over six
hundred randomly sampled companies from a variety of in-
dustries. The main three categories of social media data were
obtained from Forum, Blog, and Micro blog. For Forum, we
used a forum search engine (Boardreader)
6
to locate and dis-
play information contained on web’s forums and message
boards. For blog, we chose Google Blog Search, which pro-
vides real time, relevant search results from millions of feed-
enabled blogs. In other words, users can search for blogs or
blog posts, and can narrow their searches by dates, etc. Twitter
was selected for micro blog data.
We created four different web crawler algorithms to auto-
matically download relevant data from conventional media,
blog, forum, and twitter. A customized Python based HTML
4
As stated earlier, social capital theory states that the amount of social capital
available to member depends on the strength and quantity of the network
connections that the individual member can enlist, and the sum of the amount
of capital that each network member possesses.
5
Trust is a core concept in Putnam’s(2001) view of social capital. He claims
that although social capital is created through an individual’s active participa-
tion in a collective, it is a set of features of social organizations –like trust,
norms, and structures –that can help in creating a better society through
coordinated actions.
6
This search engine addresses the shortcomings of other currently used search
engine technology.
Inf Syst Front
Author's personal copy
(Hyper Text Markup Language) parser was designed and
imported as a Bnoise^filter to remove noisy information such
as sidebars, headers, and footers, and recognize useful text
paragraphs from large chunks of HTML code. The filter uses
information on the density of text vs. HTML code to deter-
mine if a line of text is relevant to the study. The advantage of
this filter (relative to the common html parser) resides in that it
can be applied to an arbitrary html code without specific
knowledge of page layouts or noise tags used. For each blog,
forum and conventional media, we obtained the title, date,
author, source domain, and main content of the selected arti-
cle. For each tweeter content, we obtained Twitter identifier,
the date-time of the submission (GMT + 0), submission type,
and the text content of the tweet.
7
Tweets that included URLs
(Uniform Resource Locator) were filtered to eliminate spam
messages and other advertising tweets.
A conventional text classification framework consists of
stages that include (in order), preprocessing, feature extrac-
tion, feature selection, and classification. The major steps of
a sentiment analysis approach include: finding relevant con-
tents, finding the overall sentiment, quantifying the sentiment,
and aggregating all sentiments to form an overview. Sentiment
analysis was conducted in three stages, namely, pretreatment,
SCM dimension classification, and quantitative evaluation or
sentiment polarity classification.
The aim of pretreatment isto produce clean texts for further
analyses. This stage typically includes the identification and
correction of spelling errors, elimination of arbitrary se-
quences of whitespaces between words, detection of sentence
boundaries, elimination of arbitrary use of punctuation marks,
and capitalization. This stage is usually executed in a pipeline
(Kernighan et al. 1990). The preprocessing stage also contains
tasks such as tokenization (a form of text segmentation and
removal of stop words that are usually assumed to be irrele-
vant in text classification studies), and stemming (obtaining
stem or root forms of derived words). In this pretreatment
stage, the raw social media data (WOM) was cleaned up by
deleting blanks and duplicated records, and stored as refined
corpus in a computable format. Next, sentences were recog-
nized from review segments. All words in each sentence were
normalized before storing.
The feature selection stage usually employs filtering
techniques such as, document frequency, mutual infor-
mation, information gain, chi-square, Gini index, and
distinguishing feature selector (Uysal and Gunal 2014).
In the third stage, a decision was made on whether
sentences refined in the first stage fell into one of spec-
ified SCM categories/dimensions. During this process,
both domain knowledge and machine learning algo-
rithms were used to process natural language in an in-
terpretative format. At this stage, we created the
sentiment classifier and assigned sentiment polarity for
each sentence. During this process, the associated senti-
ment training corpus was also integrated into the sup-
port set. Using a text classification algorithm, we mined
WOM toward SCM dimensions (e.g., information shar-
ing, collaboration, trust, and commitment). Finally, a
SCM opinion matrix was derived to show firms’senti-
ment on each dimension (a score from −1to1),where
a score of 1 means the supply chain partner has the
most positive view of that respective dimension, and
−1 denotes the most negative sentiment. The overall
sentiment score on each dimension was calculated
by the formula, S
i
=(N
pi
-N
ni
)/N
pi
+N
ni
), where N
pi
denotes the number of positive sentences in dimension
iand N
ni
denotes the number of negative sentences in
dimension i.
We next explain the process by which the required
social media (WOM) metrics were obtained directly
from online sources. Specifically, data on activity (fre-
quency of interaction), tone (sentiment), participation
(comments, feedbacks, opinions), and qualitative attri-
butes (meaning of the contents of participation) were
obtained directly from online sources. Frequency is fre-
quency of usage of social media by supply chain part-
ners. Vol u m e is the total number of reviews, comments,
or postings, by supply chain partners. A higher volume
is indicative of a greater intensity of member opinions
or reactions to products, or on other issues faced by
chain members. Contents used words and phrases to
determine any possible information sharing, collabora-
tion, trust, and commitment between partners. Finally,
supply chain performance was measured by inventory
turnover, a measure commonly used to measure perfor-
mance in supply chain management research
(Schönsleben 2003).
The classification stage uses well-known and successful
pattern classification algorithms such as: support vector ma-
chines, decision trees, artificial neural networks, and naïve
bayesian classifier (Uysal and Gunal 2014). The sentiment
analysis in this study is conducted using the naive bayes
(NB) process, which is a simple but effective classifier that
is very popular in the literature. This algorithm uses several
information processing techniques such as image recognition,
natural language processing (NLP), information retrieval
based on the open-source natural language toolkit (NLTK)
(Escudero et al. 2000;Pedersen2000). After the first stage,
cleanup and sentence tokenization, a total of 544,072
sentences covering 83,657 pieces of WOM was stored. We
then populated the classification system with external infor-
mation related to the four key words list that represents the
four dimensions of SCM. These keywords were selected by
manually reading 1000 comments, randomly chosen from the
83,657 available comments. The criterion to select words to
7
Limited to 140 characters by design.
Inf Syst Front
Author's personal copy
capture each dimension was based on prior research and ex-
perts’domain knowledge. We used two experienced coders
(one supply chain manager and one senior SCM researcher) to
develop codes (bag-of-words). In addition, we imported the
Cornell movie-review dataset for sentiment signals in the third
process.
8
After computing the accuracy of the selected classi-
fier on a test set, we used the F-statistic measure statistical
significance of performance measured based on precision
and recall (Pang and Lee 2005). The proposed four-
dimension classification algorithm retrieved a 0.70 F-measure
and positive-negative sentiment classification algorithm re-
trieved a 0.91 F-measure on the test set. On average, F-
measure of quaternary (four-dimension classification case)
and binary (positive-negative sentiment classification case)
is around 0.60 and 0.80 respectively (Pang and Lee 2005).
These findings suggest the statistical validity of our classifi-
cation methodology.
5 Results and discussion
The following multiple linear regression models were utilized
to measure and predict supply chain performance. In the first
model (Table 1), the input variables used were respectively:
information sharing, collaboration, trust, commitment, and
volume and frequency of social media use. As indicated by
the R-square value in the table below, almost 63% of variance
in supply chain performance can be explained by the selected
input variables. The high predictive ability of the model is also
substantiated by a high statistical significance of the overall
model (F-statistic of 39.096). The Standardized Beta
Coefficients give a measure of the contribution of each vari-
able to the model. A large value indicates that a unit change in
this predictor variable has a large effect on the criterion vari-
able. The t- and Sig (p) values give a rough indication of the
impact of each predictor variable. A large absolute t value and
small pvalue suggests that a predictor variable is having a
large impact on the criterion variable. As shown in the table
below, two of the SCM dimensions: information sharing with
β=0.538,t= 7.027 and collaboration with β=0.24,t=3.953
are significant, as such H2 and H3 are supported. However,
H1 H1a, H1b, H4, and H5 were not supported.
SCP ¼β0þβ1Inf oSentiiþβ2CollS entiiþβ3T rusSentii
þβ4CommSentiiþβ5V olumeiþβ6Frequencyi
þ∈i
A second regression was run with all inputs used in model
1, with the exception of volume and frequency. The results are
shown in Table 2. It was found that all four variables (infor-
mation sharing, collaboration, volume and frequency) were
significant at the p < = 0.05 level. Based on these results, we
conclude that H2, H3, H4, and H5 were supported.
A firm proof of the cause and effect relationship between
independent and dependent variables is always a challenge in
management research. We recognize this study might suffer
from the issue of endogeneity. Endogeneity occurs when a
regressor is correlated with the error term and can lead to
biased and inconsistent parameter estimates (Corbett 2013).
Both social media activities and supply chain performance
may be influenced by various unobserved factors, e.g. com-
pany culture, governance structure, suggesting that a possibil-
ity of endogeneity of our variables. Moreover, the choice of
firms to have social media presence and sharing relevant sup-
ply chain related content is not random. However to rule out
the possibility of endogeneity, we took these following steps.
As suggested by Hitt et al. (2016), the best way to avoid
endogeneity and common method bias is inthe initial research
design. Our design relied on collecting our dependent variable
and independent variables from two separate sources, the for-
mer from WRDS database and the latter from naturally occur-
ring social media data from: Forums, Blogs, and Micro blogs.
This avoids common source bias and rules out sources of
endogeneity such as reverse causality (Bapna et al. 2017).
Moreover, in this research we used panel data with time lags
between dependent and independent variables. We test for
muticolliniarity in both of our models, and while the highest
variance inflation factor is reported in Model 1, the score is
still well below the threshold of 10, indicating the
multicollinearity is not obviously problematic.
Literature suggests that eWOM can be a powerful social
channel if people are socially independent and the level of
commitment in decision making is low (Standing et al.
2016). In the current context, it can be particularly relevant
when consumers are discussing and sharing opinions about
brands and products (e.g. Daugherty and Hoffman 2014).
Our findings suggest that social media outlets can serve as a
powerful communication tool when consumers share content,
information, experience and recommendations. From among
the four supply chain dimensions (information sharing, col-
laboration, trust, and commitment) analyzed, information
sharing was found to be most significantly associated with
supply chain performance. While these findings are consistent
with extant literature (e.g. Tan et al. 2016), a particular contri-
bution of this study is that it sheds light on the importance of
social media usage in managing the supply chain. Social me-
dia offers a convenient, easy, and relatively inexpensive mode
for information sharing. Hence, this outlet can facilitate effi-
cient exchange of information between various elements of
the supply channels such as product demand and quality,
8
Cornell University is a pioneer in sentiment analysis, and has maintained a
vocabulary of sentiment words (e.g., dislike, enjoy) even though it is originally
derived from movie reviews. It is now regarded as the de facto sentiment
vocabulary list for sentiment analysis.
Inf Syst Front
Author's personal copy
inventory levels (Popovičet al. 2016). In addition, customer/
supply chain partners’reaction to these important factors that
can enhance individual and chain performance can also be
extracted to aid in decision making. For example, the infor-
mation gathering process related to product attributes and de-
velopment (and the collective opinion developed using social
capital) used by supply chain members can be utilized to assist
new product development and enhance performance. An
added benefit is that shared mutual goals and objectives be-
tween supply chain members and key stakeholders developed
on social media can enhance the reputation of a chain mem-
ber’s supply chain reputation. Finally, by tracking/translating
buzz in WOM (such as demand variation of a product), the
detrimental bullwhip effect in the supply chain can be re-
duced, thereby facilitating more efficient inventory manage-
ment and forecasting processes.
The findings indicate that social media content may be
useful for information sharing and collaboration within the
supply chain, not just because they simply make the commu-
nication visible to others throughout the chain, but also visi-
bility allows the members access to signals through which
they can gather relevant information from other members.
For example, the contents available on social media can pro-
vide useful information to consumers considering purchases
that require a high involvement level (Standing et al. 2016).
Involvement is a key to effective collaboration. We found
collaboration to be positively related to supply chain perfor-
mance. This finding should not occasion surprise since effec-
tive collaboration can reduce the bullwhip effect in supply
chain that is generally attributed to uncertainty in demand
and inaccurate customer demand forecasts. The social media
buzz, opinions, voices, and experiences of stakeholders can
serve as an advanced indicator of potential demand, and in
turn, allows firms to develop efficient supply chain strategies
to handle the demand.
These findings also suggest that social media content can
help supply chain members to more accurately identify and
collaborate with appropriate other chain members who have
correct and trustworthy relevant information (Jackson and
Klobas 2008). Supply chain managers can improve accuracy
of demand forecasts by analyzing WOM in terms of numbers
and sentiment of mentions for a particular product (or type of
products). Supply chain intelligence derived from WOM on
social media can improve network level collaboration, leading
to quicker responses to changes in demand and supply, there-
by improving chain profitability. In addition, because WOM
data on social media provides extensive social and interactive
user experiences and issues tracked in real time, faster resolu-
tion of issues can be forthcoming.
The study data could not find support for the hypothesis
that higher amount of discussion of trust in the content on
social media leads to improved supply chain performance.
Tabl e 1 Regression model 1 Effects Unstandardized
coefficients
Standardized coefficients t Sig. Collinearity
statistics
B Std. Error Beta Tolerance VIF
(Constant) −.140 .432 −.325 .746
Info Sharing .606 .086 .538 7.027 .000 .470 2.130
Collaboration .205 .052 .241 3.953 .000 .738 1.355
Trust .081 .071 .080 1.131 .260 .547 1.828
Commitment .114 .076 .122 1.510 .133 .422 2.370
Volume .054 .076 .049 .718 .474 .595 1.682
Frequency −.056 .076 −.057 −.740 .461 .470 2.130
Dependent variable: SCP, Adjusted R
2
= .629, F value = 39.096, p = .000
Tabl e 2 Regression model 2 Effects Unstandardized
coefficients
Standardized coefficients t Sig. Collinearity
statistics
B Std. Error Beta Tolerance VIF
(Constant) −.094 .409 −.230 .819
Info Sharing .437 .055 .623 5.661 .004 .701 2.236
Collaboration .331 .049 .118 7.375 .001 .482 1.881
Trust .108 .012 .108 3.344 .039 .354 1.649
Commitment .325 .093 .086 3.594 .046 .634 1.113
Dependent variable: SCP, Adjusted R
2
= .632, F value = 59.038, p = .000
Inf Syst Front
Author's personal copy
Our hypothesis was based on the argument that the higher the
level of trust supply chain members have in their network, the
greater the likelihood they will engage in information sharing and
collaborating behavior on social media. The study findings do
not corroborate earlier reports of a positive association between
perceived trust and members’intention to exchange information
via virtual communities (Ridings et al. 2002). One possible ex-
planation is that trust may have indirectly impacted supply chain
performance, since sharing of resources in the supply chain can
lead to a greater common resource base. In contrast, absence of
trust between partners can inhibit resource sharing. This conjec-
ture is consistent with social capital theory that postulates that
greater trust/reciprocity among members facilitates the creation
of a bigger base of social capital (Nahapiet and Ghoshal 1998).
Trust helps create stronger networks through coordinated actions
(Putnam 2001) with an eventual increase in information sharing/
collaboration between members that positively impact supply
chain performance. Although it was not clearly evident from
the results of this study, a higher level of trust among supply
chain members can lead to higher levels of information sharing
and collaboration, which in turn can positively impact supply
chain performance. This warrants further investigation as trust
impacts attitude and social media can influence attitudes and
choices of firms leading to brand switching (Gruen et al. 2006).
Putnam (2001) suggests that social capital is created
through an individual’s active participation in a network.
Greater levels of interaction/trust between partners lead to
better sharing of common resources. Social capital, being a
cooperative resource, increases with usage. Hence, commit-
ment by a member to enhance social capital benefits the entire
supply chain network. In this study, commitment was not
found to be significantly associated with supply chain perfor-
mance. However, commitment can indirectly influence perfor-
mance as it is closely linked to the sharing of resources among
its members. As an added benefit, content on social media can
enhance commitment of supply chain members because of its
potential reach, ease of use, and convenience.
In order to assess the social capital of a member generated
through content on social media, it is essential to measure and
analyze attributes such as sentiment, velocity, perception,
comments and trackbacks of other firms that include compet-
itors. The results of this study do not provide evidence of a
significant impact of volume and frequency on supply chain
performance. This is consistent with established research (Liu
2006; Basuroy et al. 2003). In fact, measuring WOM by fre-
quency, volume, and reach fails to address WOM’s power and
scope. Invariably, this method produces mixed results
(Sweeney et al. 2012). Furthermore, researchers (Gabbott
and Hogg 2000;MasonandDavis2007) insist that it is ex-
tremely important to examine the social dimensions, words,
phrases, language, and expressiveness of the WOM for deter-
mining the scope of its use. Investigation of the WOM or
partner generated content on social media can enhance the
effectiveness of simple volume measures of these contents
(Sweeney et al. 2012).
We used advanced sentiment analysis tools to identify sub-
jective information such as opinions and views of firms from
voluminous social media data. This information combined
with other data streams can provide firms with an in-depth
understanding of market trends. In addition, they can receive
real-time business intelligence that can be used in operations,
planning, production, and control processes across a supply
chain (Ahn et al. 2016). Setting up a specific supply chain
Facebook page or establishing a supply chain group on
LinkedIn can improve collaboration between members.
Member generated content on social media can also be used
as multimedia content areas where supply chain members can
post useful links and materials to share for developing their
internal supply chain network.
6 Implications
The specific aim of this study was to analyze the impact of
social media content generated by supply chain members on
supply chain performance. Drawing upon social capital theory
to explain the impact of social media content, the findings of
this research provides both theoretical and managerial contri-
butions. Although previous studies have analyzed supply
chain activities from a social capital perspective, this study
represents one of the earliest attempts to explore the impact
of social media on supply chain performance viewed from a
social capital perspective. Contents on social media have great
potential to build online communities through social media
design, and can facilitate online communication and knowl-
edge sharing (Standing et al. 2016). Supply chain members
not just can effectively leverage their social capital to commu-
nicate value to other members in the digital market place but
also can provide/obtain useful feedback. This, in turn, can
prove valuable in the design and co-creation of new products
and services.
Risk monitoring in a supply chain network is a key research
topic in the literature. Risk monitoring can prove to be expen-
sive since firms cannot easily forecast unexpected events.
Firms also rarely examine risk parameters in a systematic
manner, insofar as the supply chain and logistics management
area is concerned. In this regard, the availability of volumi-
nous social media data, cloud computing, and sophisticated
predictive analytics technology, can provide a cost effective
way for firms to analyze and handle risk in the supply chain
process (Lee et al. 2016; Swain 2016a; Di Gangi et al. 2016).
The so called BBullwhip effect^is one of the most
researched topics in supply chain management (Wang and
Disney 2016). This phenomenon has primarily been attributed
to uncertainty in demand and inaccurate customer demand
forecast. Many companies, particularly those servicing
Inf Syst Front
Author's personal copy
consumers directly, have indicated that social media is a
source for early indicators of actual consumer demand.
Social media can allow for predicting demand with greater
precision than other methods. Specifically, data generated
from social mentions for a particular product, or variant of a
product, and their response to current demand forecasts, can
be used to anticipate demand. Such social demand intelligence
can improve network collaboration and lead to faster response
rates and generate increased revenue. Despite our findings,
there is scant research on the use of social media to examine
the bullwhip effect. Examining this link would serve as an
interesting area for future research.
To manage social media content, supply chain members
need to understand the specific dimensions of these content
areas. Doing so aids building social capital and/or an online
community. The findings of this research reinforce the bene-
fits of this strategy by identifying the important elements that
form social media content. In addition, as our findings sug-
gest, integrating social networking into the firm’s interactions
with employees and suppliers, managers can better influence
and control performance throughout the chain. Supply chain
managers should consider the advantages of using social me-
dia networking to identify and track relevant information to
design appropriate strategies/solutions to improve the chain
performance while mitigating supplier risk. Greater under-
standing of the elements of social media content valued by
partners can lead to better exploitation of social media to en-
hance performance.
Despite the exploratory nature of this study, it offers many
interesting insights. For instance, the evidence suggests the
existence of a significant relationship between several key
supply chain dimensions (information sharing and collabora-
tion) and supply chain performance (e.g. Barkataki and
Zeineddine 2015). The results of this study can be used to
develop a predictive model to forecast supply chain perfor-
mance. In addition, linking social networking with the net-
works of employees and suppliers can enable managers to
improve chain performance. Information gathered from social
media combined with other data streams can provide firms
with an in-depth understanding of market trends and allow
for real-time business intelligence data that can be used in
the operations, planning, production, and control processes,
across the supply chain spectrum (Swain and Cao 2014b).
7 Limitations, future research directions,
and conclusions
Despite several interesting findings of this study, a few limi-
tations may be noted. First, the research question can be con-
sidered as broad and exploratory. The explorative nature of the
study was motivated from the dearth of adequate literature on
social media research in supply chain. We assumed it would
be appropriate to conduct this type of study as a first step.
Future research can concentrate in depth on impact of social
media on specific issues in supply chain management.
Second, in terms of external validity of this study, we sampled
firms from a population of publicly traded companies as data
for our analysis were publicly available. This may have been a
limited approach to data collection. Future studies with
sample from a wide range of firms from more industries
would increase the generalizability of the findings. Third,
we attempted to explore the relationship between social
media usage and supply chain performance. The relation-
ship cannot be concluded as a phenomenon of causality as
there may be other confounding factors affecting the rela-
tionship. For example, individual differences such as self-
presentation and motivational variables such as voluntary
self-disclosure could influence user engagement in social
media (Lee et al. 2008). It is suggested that future re-
search could use an experimental study on our concept
to examine the direction of causal effects among the key
variables which may improve the internal validity of the
findings. Fourth, we considered only a few key supply
chain factors that affect supply chain performance in this
study. Further studies might include other drivers of sup-
ply chain performance in our research model. Also it
would be interesting to explore the mediating or moderat-
ing effects between the variables in the model. Last, an-
other important extension of this research could be to
conduct this study in different cultural contexts.
Researchers (Li et al. 2009, p. 126) suggest that although
Bthe Internet is a global medium, but its content is local to
each country^. Investigating the hypotheses presented in
this study under different cultural contexts would add fur-
ther support to the current findings.
A few have argued that social media generated data are
irrelevant and lack purpose (e.g. Standing et al. 2016). This
study provides both theoretical and empirical evidence to
counter this argument by demonstrating that a detailed analy-
sis of user generated content on social media can be very
valuable for the supply chain. The more firms realize the use-
fulness of social media the more they will be able to harness its
potential.
Talking about social media analytics, opinion mining is an
emerging area of research, although, admittedly, it is still in its
infancy in the supply chain area. Future research can focus on
the impact of social media on specific issues in supply chain
management. Data can be gathered from a broader spectrum
of industries and from other areas to see if our findings are
generalizable. Finally, sentiment is an important and insightful
indicator of public opinion. However, it is often critiqued as a
single metric that cannot address fundamental policy and de-
cision making related questions. Combining sentiment analy-
sis with other available methods to extend this research could
be another fruitful avenue for research.
Inf Syst Front
Author's personal copy
References
Adler, P. S., & Kwon, S. W. (2002). Social capital: Prospects for a new
concept. Academy of Management Review, 27(1), 17–40.
Ahn, J., Ma, K., Lee, O., & Sura, S. (2016). Do big data support TV
viewing rate forecasting? A case study of a Korean TV drama.
Information Systems Frontiers,19(2),411–420.
Allen, N. J., & Meyer, J. P. (1990). The measurement and antecedents of
affective, continuance and normative commitment to the organiza-
tion. JournalofOccupationalPsychology,63(1), 1–18.
Angulo, A., Nachtmann, H., & Waller, M. A. (2004). Supply chain in-
formation sharing in a vendor managed inventory partnership.
JournalofBusinessLogistics,25(1), 101–120.
Aral, S., Dellarocas, C., & Godes, D. (2013). Introduction to the special
issue-social media and business transformation: A framework for
research. Information Systems Research, 24(1), 3–13.
Avery, S. L., & Swafford, P. M. (2009). Social capital impact on service
supply chains. Journal of Service Science, 2(2), 9.
Aviv, Y. (2001). The effect of collaborative forecasting on supply chain
performance. Management Science, 47(10), 1326–1343.
Baker, S. (2009). Beware social media snake oil. Business week
(December 3). http://www.Businessweek.Com/magazine/content/
09_50/64159048693735.Htm.
Balahur, A., & Jacquet, G. (2015). Sentiment analysis meets social me-
dia–challenges and solutions of the field in view of the current in-
formation sharing context. Information Processing & Management,
51(4), 428–432.
Balakrishnan, A., & Geunes, J. (2004). Collaboration and coordination in
supply chain management and E-commerce. Production and
Operations Management, 13(1), 1–2.
Ballou, R. H., Gilbert, S. M., & Mukherjee, A. (2000). New managerial
challenges from supply chain opportunities. Industrial Marketing
Management, 29(1), 7–18.
Bapna, R., Gupta, A., Rice, S., & Sundararajan, A. (2017). Trust and the
strength of ties in online social networks: An exploratory field ex-
periment. MIS Quarterly, 41(1), 115–130.
Barkataki, S., & Zeineddine, H. (2015). On achieving secure collabora-
tion in supply chains. Information Systems Frontiers, 17(3), 691–
705.
Basuroy, S., Chatterjee, S., & Ravid, S. A. (2003). How critical are critical
reviews? The box office effects of film critics, star power, and bud-
gets. JournalofMarketing,67(4), 103–117.
Benbya, H., & Van Alstyne, M.W. (2010). How to find answers within
your company?
Bernardes, E. S. (2010). The effect of supply management on aspects of
social capital and the impact on performance: A social network
perspective. Journal of Supply Chain Management, 46(1), 45–55.
Bharati, P., Zhang, C., & Chaudhury, A. (2014). Social media assimila-
tion in firms: Investigating the roles of absorptive capacity and in-
stitutional pressures. Information Systems Frontiers, 16(2), 257–
272.
Blackshaw, P., & Nazzaro, M. (2006). Consumer-generated media
(CGM) 101: Word-of-mouth in the age of the web-fortified
consumer. New York: Nielsen BuzzMetrics.
Borgatti, S. P., & Li, X. (2009). On social network analysis in a supply
chain context. Journal of Supply Chain Management, 45(2), 5–22.
Bourdiue, P. (1985). The forms of capital. In handbook of theory and
research for the sociology of education. NY: Greenwood,
Richardson publisher.
Brannen, M. Y. (2004). When Mickey loses face: Recontextualization,
semantic fit, and the semiotics of foreignness. Academy of
Management Review, 29(4), 593–616.
Brown, J. J., & Reingen, P. H. (1987). Social ties and word-of-mouth
referral behavior. Journal of Consumer Research, 14(3), 350–362.
Burgess, K., & Singh, P. J. (2012). Using the social system of a supply
chain to improve a focal organization’s operating performance.
Operations Management Research, 5(1–2), 57–68.
Burgess, K., Singh, P. J., & Koroglu, R. (2006). Supply chain manage-
ment: A structured literature review and implications for future re-
search. International Journal of Operations & Production
Management, 26(7), 703–729.
Burgess, S., Sellitto, C., Cox, C., & Buultjens, J. (2011). Trust perceptions
of online travel information by different content creators: Some
social and legal implications. Information Systems Frontiers,
13(2), 221–235.
Cao, X., Guo, X., Liu, H., & Gu, J. (2015). The role of social media in
supporting knowledge integration: A social capital analysis.
Information Systems Frontiers, 17(2), 351–362.
Casemore, S. (2012). Social media and the coming supply-chain revolu-
tion. CFO. February, 29, 2012.
Chang, W.L., & Jung, C.F. (2015). A hybrid approach for personalized
service staff recommendation. Information systems Frontiers, 19(1),
149-163. doi:10.1007/s10796-015-9597-7
Chang, H. H., Tsai, Y. C., Fu, C. S., Chen, S. H., & De Peng, Y. (2016).
Exploring the antecedents and consequences of technology and
knowledge integration mechanisms in the context of NPD.
Information Systems Frontiers, 18(6), 1165–1189.
Charband, Y., & Navimipour, N. J. (2016). Online knowledge sharing
mechanisms: A systematic review of the state of the art literatureand
recommendations for future research. Information Systems
Frontiers, 18(6), 1131–1151.
Chevalier, J. A., & Mayzlin, D. (2006). The effect of word of mouth on
sales: Online book reviews. Journal of Marketing Research, 43(3),
345–354.
Child, J. (2001). Trust—The fundamental bond in global collaboration.
Organizational Dynamics, 29(4), 274–288.
Chu, S. C., & Kim, Y. (2011). Determinants of consumer engagement in
electronic word-of-mouth (eWOM) in social networking sites.
International journal of Advertising, 30(1), 47–75.
Cole, J. (2000). Surveying the digital future. UCLA Center for commu-
nication policy, University of California–Los Angeles.
Coleman, J. S. (1988). Social capital in the creation of human capital.
American Journal of Sociology, 94,S95–S120. doi:10.1086/228943
Corbett, J. (2013). Using information systems to improve energy efficien-
cy: Do smart meters make a difference? Information Systems
Frontiers, 15(5), 747–760.
van Dam, J. W., & van de Velden, M. (2015). Online profiling and clus-
tering of Facebook users. Decision Support Systems, 70,60–72.
Daugherty, T., & Hoffman, E. (2014). eWOM and the importance of
capturing consumer attention within social media. Journal of
Marketing Communications, 20(1–2), 82–102.
Deans, P. C. (2011). The impact of social media on C-level roles. MIS
Quarterly Executive, 10(4), 187–200.
Di Gangi, P. M., & Wasko, M. (2009). Steal my idea! Organizational
adoption of user innovations from a user innovation community:
A case study of Dell IdeaStorm. Decision Support Systems, 48(1),
303–312.
Di Gangi, P.M., Johnston, A.C., Worrell, J.L., & Thompson, S.C. (2016).
What could possibly go wrong? A multi-panel Delphi study of or-
ganizational social media risk. InformationSystems Frontiers, 1–20.
doi:10.1007/s10796-016-9714-2
Dou, Y., Niculescu, M. F., & Wu, D. J. (2013). Engineering optimal
network effects via social media features and seeding in markets
for digital goods and services. Information Systems Research,
24(1), 164–185.
Duan, W., Gu, B., & Whinston, A. B. (2008). Do online reviews mat-
ter?—An empirical investigation of panel data. Decision Support
Systems, 45(4), 1007–1016.
Ellison, N. B., Steinfield, C., & Lampe, C. (2007). The benefits of
Facebook Bfriends:^social capital and college students’use of
Inf Syst Front
Author's personal copy
online social network sites. Journal of Computer-Mediated
Communication, 12(4), 1143–1168 .
Escudero, G., Màrquez, L., & Rigau, G. (2000). Naive Bayes and
exemplar-based approaches to word sense disambiguation revisited.
arXiv preprint cs/0007011.
Fombrun, C. J., Gardberg, N. A., & Barnett, M. L. (2000). Opportunity
platforms and safety nets: Corporate citizenship and reputational
risk. Business and Society Review, 105(1), 85–106.
Gabbott, M., & Hogg, G. (2000). An empirical investigation of the impact
of non-verbal communication on service evaluation. European
JournalofMarketing,34(3/4), 384–398.
Gallaugher, J., & Ransbotham, S. (2010). Social media and customer
dialog management at Starbucks. MIS Quarterly Executive, 9(4),
197–212.
Ganesan, S. (1994). Determinants of long-term orientation in buyer-seller
relationships. The Journal of marketing, 58(2), 1–19.
Godes, D. (2016). Product policy in markets with word-of-mouth com-
munication. Management Science, 63(1), 267–278.
Godes, D., & Mayzlin, D. (2004). Using online conversations to study
word-of-mouth communication. Marketing Science, 23(4), 545–
560.
Granovetter, M. S. (1973). The strength of weak ties. American Journal
of Sociology,1360–1380.
Granovetter, M. (1983). The strength of weak ties: A network theory
revisited. Sociological Theory, 1(1), 201–233.
Gretzel, U., & Yoo, K. H. (2008). Use and impact of online travel re-
views. Information and communication technologies in tourism,
2008,35–46.
Gruen, T. W., Osmonbekov, T., & Czaplewski, A. J.(2006). eWOM: The
impact of customer-to-customer online know-how exchange on cus-
tomer value and loyalty. Journal of Business Research, 59(4), 449–
456.
Gundecha, P., & Liu, H. (2012). Mining social media: A brief introduc-
tion. Tutorials in Operations Research, 1(4), 1–17.
Hammervoll, T. (2011). Honeymoons in supply chain relationships: The
effects of financial capital, social capital and psychological commit-
ment. The International Journal of Logistics Management, 22(2),
264–279.
Hanifan, L. J. (1916). The rural school community center. The Annals of
the American Academy of Political and Social Science, 67,130–
138.
Harris, J. K., Snider, D., & Mueller, N. (2013). Social media adoption in
health departments nationwide: The state of the states. Frontiers in
Public Health Services and Systems Research, 2(1), 5.
Hitt, M. A., Carnes, C. M., & Xu, K. (2016). A current view of resource
based theory in operations management: A response to Bromiley
and Rau. Journal of Operations Management, 41(10), 107–109.
Holsapple, C. W., & Wu, J. (2008). Building effective online game
websites with knowledge-based trust. Information Systems
Frontiers, 10(1), 47–60.
Inderfurth, K., Sadrieh, A., & Voigt, G. (2013). The impact of information
sharing on supply chain performance under asymmetric informa-
tion. Production and Operations Management, 22(2), 410–425.
Jackson, P., & Klobas, J. (2008). Transactive memory Systems in
Organizations: Implications for knowledge directories. Decision
Support Systems, 44(2), 409–424.
Kavanaugh, A. L., Reese, D. D., Carroll, J. M., & Rosson, M. B. (2005).
Weak ties in networked communities. The Information Society,
21(2), 119–131.
Kenski, K., & Stroud, N. J. (2006). Connections between internet use and
political efficacy, knowledge, and participation. Journal of
Broadcasting & Electronic Media, 50(2), 173–192.
Kernighan, M.D., Church, K.W., & Gale, W.A. (1990). A spelling cor-
rection program based on a noisy channel model. In proceedings of
the 13th conference on computational linguistics-volume 2 (pp. 205-
210). Association for computational linguistics.
Kobayashi, T., Ikeda, K. I., & Miyata, K. (2006). Social capital online:
Collectiveuse of the internet and reciprocity as lubricants of democ-
racy. Information, Community & Society, 9(5), 582–611.
Kotlarsky, J., & Oshri, I. (2005). Social ties, knowledge sharing and
successful collaboration in globally distributed system development
projects. European Journal of Information Systems, 14(1), 37–48.
Krause, D. R., Handfield, R. B., & Tyler, B. B. (2007). The relationships
between supplier development, commitment, social capital accumu-
lation and performance improvement. Journal of Operations
Management, 25(2), 528–545.
Lam, H. K., Yeung, A. C., & Cheng, T. E. (2016). The impact of firms’
social media initiatives on operational efficiency and innovative-
ness. Journal of Operations Management, 47,28–43.
Laumann, E. O. (1966). Prestige and association in an urban community:
An analysis of an urban stratification system. Bobbs-Merrill
Company.
Lawson, B., Tyler, B. B., & Cousins, P. D. (2008). Antecedents and
consequences of social capital on buyer performance improvement.
Journal of Operations Management, 26(3), 446–460.
Lee, D. H., Im, S., & Taylor, C. R. (2008). Voluntary self-disclosure of
information on the internet: A multimethod study of the motivations
and consequences of disclosing information on blogs. Psychology &
Marketing, 25(7), 692–710.
Lee, K., Park, J., Kim, I., & Choi, Y. (2016). Predicting movie success
with machine learning techniques: Ways to improve accuracy.
Information Systems Frontiers,1–12. doi:10.1007/s10796-016-
9689-z
Lejeune, M. A., & Yakova, N. (2005). On characterizing the 4 C's in
supply chain management. Journal of Operations Management,
23(1), 81–100.
Leonardi, P. M. (2015). Ambient awareness and knowledge acquisition:
Using social media to learn Bwho knows what^and Bwho knows
whom^.MIS Quarterly, 39(4), 747–762.
Levi, D. S., Kaminsky, P., & Levi, E. S. (2003). Designing and managing
the supply chain: Concepts, strategies, and case studies.McGraw-
Hill.
Li, H., Li, A., & Zhao, S. (2009). Internet advertising strategy of multi-
nationals in China: A cross-cultural analysis. International Journal
of Advertising, 28(1), 125–145.
Li, Y., Lin, Z., Xu, L., & Swain, A. (2015). Bdo the electronic books
reinforce the dynamics of book supply chain market?^–a theoretical
analysis. European Journal of Operational Research, 245(2), 591–
601.
Lin, C., Kuei, C. H., Madu, C.N., & Winch, J. (2010). Identifyingcritical
success factors for supply chain excellence. IJSDS, 1(3), 49–70.
Liu, B. (2012). Sentiment analysis and opinion mining. Synthesis lectures
on human language technologies, 5(1), 1–167.
Liu, Y. (2006) Word of Mouth for Movies: Its Dynamics and Impact on
Box Office Revenue. Journal of Marketing 70(3), 74–89.
Luo, X., Zhang, J., & Duan, W. (2013). Social media and firm equity
value. Information Systems Research, 24(1), 146–163.
Ma, W. W., & Chan, A. (2014). Knowledge sharing and social media:
Altruism, perceived online attachment motivation, and perceived
online relationship commitment. Computers in Human Behavior,
39,51–58.
Mabert, V. A., & Venkataramanan, M. A. (1998). Special research focus
on supply chain linkages: Challenges for design and management in
the 21st century. Decision Sciences, 29(3), 537–552.
Madison, I. (2012). Why your social media metrics are a waste of time.
Harvard Business Review Blog Network.
Mangold, W. G., & Faulds, D. J. (2009). Social media: The new hybrid
element of the promotion mix. Business Horizons, 52(4), 357–365.
Marcus, A. A., & Goodman, R. S. (1991). Victims and shareholders: The
dilemmas of presenting corporate policy during a crisis. Academy of
Management Journal, 34(2), 281–305.
Inf Syst Front
Author's personal copy
Markova, S., & Petkovska-Mircevska, T. (2013). Social media and sup-
ply chain. Amfiteatru Economic, 15(33), 89.
Mason, P. R., & Davis, B. H. (2007). More than the words: Using stance-
shift analysis to identify crucial opinions and attitudes in online
focus groups. Journal of Advertising Research, 47(4), 496–506.
Meyer, J. P., Allen, N. J., & Sulsky, L. M. (1999). Commitment in the
workplace: Theory, research & application. Canadian Psychology,
40(4), 383.
Miller, A. R., & Tucker, C. (2013). Active social media management: The
case of health care. Information Systems Research, 24(1), 52–70.
Min, S., Kim, S. K., & Chen, H. (2008). Developing social identity and
social capital for supply chain management. Journal of Business
Logistics, 29(1), 283–304.
Morgan, R. M., & Hunt, S. D. (1994). The commitment-trust theory of
relationship marketing. The Journal of Marketing, 58(3), 20–38.
Nahapiet, J., & Ghoshal, S. (1998). Social capital, intellectual capital, and
the organizational advantage. Academy of Management Review,
23(2), 242–266.
O'leary, D. E. (2011). Theuse of social media in the supply chain: Survey
and extensions. Intelligent Systems in Accounting, Finance and
Management, 18(2–3), 121–144.
Pang, B., & Lee, L. (2005). Seeing stars: Exploiting class relationships for
sentiment categorization with respect to rating scales. In proceedings
of the 43rd annual meeting on Association for Computational
Linguistics (pp. 115-124). Associationfor computational linguistics.
Pedersen, T. (2000). A simple approach to building ensembles of Naive
Bayesian classifiers for word sense disambiguation. In proceedings
of the 1st north American chapter of the Association for
Computational Linguistics conference (pp. 63-69). Association for
computational linguistics.
Peng, M. W., & Luo, Y. (2000). Managerial ties and firm performance in a
transition economy: The nature of a micro-macro link. Academy of
Management Journal, 43(3), 486–501.
Popovič, A., Hackney, R., Tassabehji, R., & Castelli, M. (2016). The
impact of big data analytics on firms’high value business perfor-
mance. Information Systems Frontiers,1–14. doi:10.1007/s10796-
016-9720-4
Poria, S., Gelbukh, A., Cambria, E., Hussain, A., & Huang, G. B. (2014).
EmoSenticSpace: A novel framework for affective common-sense
reasoning. Knowledge-Based Systems, 69,108–123.
Portes, A. (1998). Social capital: Its origins and applications in modern
sociology. Annual review of sociology, 24(1), 1–24.
Putnam, R. D. (2001). Bowling alone: The collapse and revival of
American community. Simon and Schuster.
Quesada, G., Rachamadugu, R., Gonzalez, M., & Luis Martinez, J.
(2008). Linking orderwinning and external supply chain integration
strategies. Supply Chain Management: An International Journal,
13(4), 296–303.
Quesenberry, K.A. (2016). Fix your social media strategy by taking it
back to basics. Harvard Business Review, July 25. https://hbr.org/
2016/07/fix-your-social-media-strategy-by-taking-it-back-to-basics.
Accessed 25 June 2017
Ramos, L., & Young, G.O. (2009). The social technographics of business
buyers. Forrester research, Cambridge, MA, February, 20.
Ransbotham, S., Kiron, D., & Prentice, P.K. (2016). Beyond the hype:
The hard work behind analytics success. MIT Sloan Management
Review, March 8.
Reingen, P. H., Foster, B. L., Brown, J. J., & Seidman, S. B. (1984).
Brand congruence in interpersonal relations: A social network anal-
ysis. JournalofConsumerResearch,11(3), 771–783.
Ridings, C. M., Gefen, D., & Arinze, B. (2002). Some antecedents and
effects of trust in virtual communities. Journal of Strategic
Information Systems, 11(3 & 4), 271–295.
Rodriguez, M., Peterson, R. M., & Krishnan, V. (2012). Social media’s
influence on business-to-business sales performance. Journal of
Personal Selling & Sales Management, 32(3), 365–378.
Roth, A. V., Tsay, A. A., Pullman, M. E., & Gray, J. V. (2008). Unraveling
the food supply chain: Strategic insights from China and the 2007
recalls. Journal of Supply Chain Management, 44(1), 22–39.
Sahin, F., & Robinson, E. P. (2002). Flow coordination and information
sharing in supply chains: Review, implications, and directions for
future research. Decision Sciences, 33(4), 505–536.
Schönsleben, P. (2003). Integral logistics management: planningand con-
trol of comprehensive supply chains. CRC Press.
Shah, V., Nojin Kwak, R., & Lance Holbert, D. (2001). "Connecting"
and" disconnecting" with civic life: Patterns of Internet use and the
production of social capital. Political communication, 18(2), 141–
162.
Sharma, G., Qiang, Y., Wenjun, S., & Qi, L. (2013). Communication in
virtual world: Second life and business opportunities. Information
Systems Frontiers, 15(4), 677–694.
Shipilov, A. (2012). Use social media to develop emotional capital with
your employees. INSEAD BLOG, http://blog.Insead.Edu/2012/09/
use-social-media-to-develop-emotional-capital-with-your-
employees/ accessed on April 5th, 2012.
Simatupang, T.M., & Sridharan, R. (2001, November). A characterization
of information sharing in supply chains. In ORSNZ Conference
University of Canterbury, New Zealand.
Simchi-Levi, D., Kaminsky, P., & Simchi-Levi, E. (2003). Designing and
management the supply chain. Editorial: McGraw-Hill, Irwin, New
York , N Y.
Snub, A. N., & Stonebraker, P. W. (2009). The human impact on supply
chains: Evaluating the importance of ‘soft’areas on investigation
and performance. Supply Chain Management International Journal,
14(1), 31–40.
Sofia, R., Mendes, P., Damásio, M. J., Henriques, S., Giglietto, F.,
Giambitto, E., & Bogliolo, A. (2012). Moving towards a socially-
driven internet architectural design. ACM SIGCOMM Computer
Communication Review, 42(3), 39–46.
Standing, C., Holzweber, M., &Mattsson, J. (2016). Exploring emotional
expressions in e-word-of-mouth from online communities.
Information Processing & Management.
Suchman, L. (1987). Plans and situated actions: The problem of human
machine communication. Cambridge: Cambridge University Press.
Swain, A. K. (2016a). Big data analytics: An expert interview with Bipin
Chadha, data scientist for united services automobile association
(USAA). Journal of Information Technology Case and
Application Research, 18(3), 181–185. doi:10.1080/15228053.
2016.1223497.
Swain, A. K. (2016b). Mining big data to support decision making in
healthcare. Journal of Information Technology Case and
Application Research, 18(3), 141–154. doi:10.1080/15228053.
2016.1245522.
Swain, A., & Cao, Q. (2013). Exploring the impact of social media on
supply chain performance: A sentiment analysis. In The proceedings
of 44th Annual Meeting, Decision Sciences Institute, November (pp.
16–19). Maryland: Baltimore.
Swain, A.K., & Cao, Q. (2014a). Impact of online firm generated content
(FGC) on supply chain performance: An empirical analysis, in 2014
47th Hawaii International conference on system sciences (pp. 561-
573). IEEE.
Swain, A., & Cao, Q. (2014b). Strategic alignment and sustainability
performance: An operations management perspective. In the pro-
ceedings of 45th Annual Meeting, Decision Sciences Institute,
November 22-25, Tampa, Florida,USA.1695-1715.
Sweeney, J. C., Soutar, G. N., & Mazzarol, T. (2012). Word of mouth:
Measuring the power of individual messages. European Journal of
Marketing, 46(1/2), 237–257.
Tan, K. H., Wong, W. P., & Chung, L. (2016). Information and knowl-
edge leakage in supply chain. Information Systems Frontiers, 18(3),
621–638.
Inf Syst Front
Author's personal copy
Trusov, M., Bodapati, A., & Bucklin, R. (2010). Determining influential
users in internet social networks. Journal of Marketing Research,
47(4), 643–658.
Uysal, A. K., & Gunal, S. (2014). The impact of preprocessing
on text classification. Information Processing & Management,
50(1), 104–112.
Valenzuela, S., Park, N., & Kee, K. F. (2009). Is there social capital in a
social network site?: Facebook use and college students’life satis-
faction, trust, and participation1. Journal of Computer-Mediated
Communication, 14(4), 875–901.
Villanueva, J., Yoo, S., & Hanssens, D. M. (2008). The impact of marketing-
induced versus word-of-mouth customer acquisition on customer equi-
ty growth. Journal of Marketing Research, 45(1), 48–59.
Villena, V. H., Revilla, E., & Choi, T. Y. (2011). The dark side of buyer–
supplier relationships: A social capital perspective. Journal of
Operations Management, 29(6), 561–576.
Wang, X., & Disney, S. M. (2016). The bullwhip effect: Progress, trends
and directions. European Journal of Operational Research, 250(3),
691–701.
Warkentin, M. E., Sayeed, L., & Hightower, R. (1997). Virtual teams
versus face-to-face teams: An exploratory study of a web-based
conference system. Decision Sciences, 28(4), 975–996.
Wei, H.L., & Ju, P.H. (2010). Dimensions of social capital and supply
chain knowledge creation: The mediating role of learning informa-
tion exchange. In computers and Industrial Engineering (CIE), 2010
40th International conference on (pp. 1-5). IEEE.
Wu, W. Y., Chiag, C. Y., Wu, Y. J., & Tu, H. J. (2004). The influencing
factors of commitment and business integration on supply chain
management. Industrial Management and Data Systems, 104(4),
322–333.
Xia, R., Xu, F., Yu, J., Qi, Y., & Cambria, E. (2016). Polarity shift detec-
tion, elimination and ensemble: A three-stage model for document-
level sentiment analysis. Information Processing & Management,
52(1), 36–45.
Yim, B., & Leem, B. (2013). The effect of the supplychain social capital.
Industrial Management & Data Systems, 113(3), 324–349.
Zhang, J., & Piramuthu, S. (2016). Product recommendation with latent
review topics. Information Systems Frontiers, 1–9. doi:10.1007/
s10796-016-9697-z
Zhang, D., Si, L., & Rego, V. J. (2012). Sentiment detection with auxil-
iary data. Information Retrieval, 15(3–4), 373–390.
Dr. Ajaya Kumar Swain is an Assistant Professor of Quantitative
Management in the Finance and Quantitative Management Department
in the Greehey School of Business at St. Mary’s University. He received
his master’s degree in Industrial Management Systems Engineering from
University of Nebraska-Lincoln, and an MBA in Business Statistics and
PhD in Operations Management from Texas Tech University. His re-
search interests include predictive and social media analytics, operations
and supply chain performance, and corporate sustainability. His publica-
tions have appeared in European Journal of Operational Research,
Journal of Manufacturing Processes, IEEE Computer Society Journal,
and Journal of Information Technology Case and Application Research
among others.
Dr. Ray Qing Cao is a Professor of Supply Chain Management in the
Davies College of Business at University of Houston –Downtown. Dr.
Cao has published more than 56 research papers in journals such as
Journal of Operations Management, Decision Sciences, Journal of
Association for Information Systems, Communications of ACM,
International Journal of Production Research, European Journal of
Operational Research, Annals of Operations Research, International
Journal of Production Economics among others. Dr. Cao was the recip-
ient of the University of Missouri–Kansas City Trustee’s Faculty
Research Award (2005) and the Chancellor's Distinguished Research
Award at Texas Tech University (2012).
Inf Syst Front
Author's personal copy
A preview of this full-text is provided by Springer Nature.
Content available from Information Systems Frontiers
This content is subject to copyright. Terms and conditions apply.
