Damned If You Do, Damned If You Don’t
Weapons Manufacturers, Tobacco Companies, and CSR Communication on Twitter
Chinmay Brahme Student ID: 11696648
Master’s thesis
Graduate School of Communication Master’s programme Communication Science
Supervisor Dr. Piet Verhoeven
Abstract
Corporate Social Responsibility (CSR) and its communication are integral components in enhancing organizational reputation. CSR and its communication has shown to be beneficial for a host of organizations across the globe. But, is this true for organizations that polarize opinion in the extreme fashion that weapons manufacturers and tobacco companies do? A substantial amount of literature exists on analyzing and critiquing, CSR and its communication, of a large majority of global organizations. However, scant attention has been paid to CSR efforts of weapons manufacturers and tobacco companies. Both sectors generate revenue in excess of billions of dollars and also carry out CSR activities. But the way in which they communicate CSR, the engagement it receives, and the multitude of reactions that it elicits, is lacking. This study aims to look at CSR communication of weapons manufacturers and tobacco companies on Twitter. It examines which communication strategies they prefer: broadcasting, reactive, proactive, and analyzes engagement received. The results of this study do not provide significant support to a particular communication strategy, and fail to clearly distinguish one particular strategy as a clear leader. While use of CSR-specific frames did not display significant differences in engagement, CSR topics and their communication did. Some topics came in for increased flak as compared to others for both industries. However, a clear light is shone on how CSR communication, predominantly for weapons manufacturers than tobacco companies, is a rocky road strewn with abuse, vilification, and a constant shadow of skepticism.
Introduction
“You are war criminals”, “You have the blood of so many innocents on your hand. Shame on you!”, “Stop making cancer sticks”, “Merchants of death”, “You don’t mean this! Making millions off blackened lungs”. These reactions are not spur-of-the-moment utterances in the aftermath of a tragedy but parts of Tweets that were posted in response to corporate social responsibility (CSR) communication about scholarships being awarded to underprivileged students and volunteering efforts to help communities reeling from natural disasters. Reactions like these, varying in degree, severity, and coarseness of language, are a frequent occurrence on CSR Tweets of organizations who operate in the weapons manufacturing and tobacco sector.
Corporate Social Responsibility (CSR) is a term that holds a growing amount of currency for businesses and corporate organizations globally (Frederick, 2008). Businesses and corporate organizations are increasingly looking to communicate activities that are responsible, beneficial, and helpful to their environment (Grover, 2014). CSR and its effective communication are tools that has been proven to boost organizational reputation positively (Sethi, Martell & Demir, 2016). Weapons manufacturers and tobacco companies are no stranger to CSR activities and their communication either. Both industries are highly profitable, and both operate comprehensive CSR programmes (Halpern & Snider, 2011; Palazzo & Richter, 2005).
A preferred medium for the communication of CSR activities for a majority of corporate organizations globally, including weapons manufacturers and tobacco companies is Twitter (Shin, Pang & Kim, 2015). Both operate active Twitter accounts amongst other social media platforms, using these to disseminate information about their working, new products, and
CSR activities (Shin et al. 2015). The use of Twitter as a medium for official communication by organizations has seen to have risen over the last decade (Rybalko & Seltzer, 2010). Today, Twitter enjoys a user-base of more than 330 million accounts (Statista, 2018) and has an almost global footprint. Greenberg (2010) argues that it is beneficial for an organization to operate a Twitter account, using it to disseminate communication and also to establish links with stakeholders and publics.
Weapons manufacturers and tobacco companies have been a part of the Twitter bandwagon. However, their CSR programmes communicated on Twitter have been seen to come in for scathing criticism no matter the size, reach, and commitment of these offerings (Halpern & Snider, 2011; Palazzo & Richter, 2005). This can partly be attributed to the lens that is used to view the products that these two industries manufacture which are said to ‘facilitate harm’ (Byrne, 2007). A prominent view is that both industries (weapons and tobacco) are incapable of being ‘responsible’ since they violate long-held ethical standards (Baker, 2005) through the production of their core offerings (weapons, cigarettes).
The weapons industry also commonly known as the defence industry (industries and corporations that manufacture weapon systems for governmental and private clients) (hereafter referred exclusively as weapons manufacturers) is amongst some of the biggest revenue-generating industries in the world. According to the Stockholm Peace Research Institute (SIPRI, 2012), 1.8 trillion US Dollars were spent on defence purchases globally in 2012. The tobacco industry is a similar behemoth, having produced 7.44 million metric tons just in 2013 (Statista, 2018). Tobacco like arms is purchased, consumed, and utilized all over the world. One of tobacco’s biggest uses is as an ingredient in producing cigarettes. In 2016, approximately 5.5 billion cigarettes were consumed (Statista, 2018), making tobacco and
cigarette producing companies, some of the most profitable in the world. Just as an example, the Imperial Tobacco Group, a British company which ranks only at number four in the world according to market share, generated a revenue of USD 39.1 billion in 2016 (Imperial Tobacco PLC, 2017).
Interestingly, a look at some of the numbers that are available on the actual allocation to CSR programmes by arms and tobacco manufacturers makes for interesting reading. In 2011, Phillip Morris International claimed to have spent upwards of 29.5 million US Dollars on CSR programmes on a global level (Phillip Morris International, 2017), while Imperial Tobacco’s 2017 Sustainability Report (Imperial Tobacco PLC, 2017), while declining to mention exact figures allocated for CSR, claims that three million Pounds was allocated for ‘community partnership investment’. Similarly, Lockheed Martin, the world’s biggest weapons manufacturer (Deloitte, 2017) claims to have spent 45.5 million US Dollars in 2017 on its entire CSR programme (Lockheed Martin, 2017). BAE Systems, another top-5 weapons manufacturer (Deloitte, 2017), claims to have spent 11 million Pounds annually on CSR programmes in 2015 (BAE Systems, 2015).
Both industries execute comprehensive CSR projects but these efforts are generally met with reactions that range from ‘not doing enough’ to ‘oh, they are just being forced to do it, they don’t really care!’ (Shim, Chung & Kim, 2017). Shim et al. (2017) came to this conclusion after running a between-subjects experimental design that considered individual ethical orientation and the media framing of CSR topics. Nowhere are these reactions strikingly clear than the arena of social media (Saxton, Gomez, Ngoh, Lin & Dietrich, 2017). Mention of CSR efforts by arms manufacturers are often met with derision, negativity and exhibit a
classic ‘catch-22’ situation where the industry can be pilloried for not doing enough CSR or cop flak for ‘not being serious enough about it’ (Bennett, 2017).
It is therefore interesting to study how these two controversial industries approach the communication of their CSR activities. More importantly, we would like to study differences, if any, in the communication of these CSR programmes specifically on Twitter. Does a difference exist between the type of CSR communication that each of the industries engage in? Do they have a distinctive direction or tilt in terms of type of CSR initiative and more? What are the reactions that this communication spawns? As Hirschhorn (2004) says, tobacco companies train their sights on launching and promoting CSR initiatives that deal with health, fair-trade, and community outreach. Weapons manufacturers seem to focus more on diversity at the workplace, initiatives in education, and philanthropy amongst others (Halpern & Snider, 2011). A study looking specifically at these two industries has not been carried out so far and thus presents an interesting avenue for empirical exploration. According to Aqueveque, Rodrigo & Duran (2018), weapons manufacturers and tobacco industries can be identified as part of the ‘controversial industries’ (CIs) club but as their work suggests, it would not be fair to group weapons manufacturers together with other industries that are perceived as ‘harm-inducing’ like alcohol and gambling. Furthermore, existing empirical literature (Byrne, 2007; Halpern & Snider, 2011) has put forth the argument that weapons manufacturers while at times selling their products to private clients, are largely regarded as agents of the state. This leads to diminished coverage of their business activities due to overarching legislation and non-disclosure clauses to protect national security (Byrne, 2007; Halpern & Snider, 2011). Conversely, the access and availability of tobacco products across the world is much greater, even in highly-regulated societies (Paul et al. 2010), providing tobacco companies much greater visibility in the lives of private citizens. Thus, to put it
simply, while it is easy to gain access to a pack of cigarettes in almost every corner of the world, the same cannot be said about a fighter aircraft or an armoured fighting vehicle. Thus, it is expected that differences exist in the reception of CSR communication on Twitter between the industries. This study proposes to study these differences by focusing the lens on the official Twitter accounts of certain leading members of these two industries leading us to investigate:
RQ: What are the differences in the CSR communication efforts of the weapons industry as compared to the tobacco industry in communication on Twitter?
Theoretical Background
CSR Communication on Social Media
It is obvious that CSR and its effective communication has a positive impact on organizational reputation, encourages positives perceptions of the organization in society, and is an effective tool of stakeholder engagement (Devin & Lane, 2014; Johansen & Nielsen, 2011). In contemporary times, it is accepted that the internet, specifically social media, can allow organizations to disseminate CSR communication quickly, effectively, and also achieve the elusive goal of reaching a wider chunk of the public (Manetti, Belluci, & Bagnoli, 2017). The benefits of social media are apparent, its speed, reach, and easy accessibility giving rise to platforms where organizations find a channel to showcase their initiatives (CSR) while allowing for almost real-time feedback, engagement, and possible endorsement, through the various functions (reactions, shares) (Saffer, Sommerfeldt, & Taylor, 2012; Araujo & Kollat, 2018; Uzunoğlu, Türkel, & Yaman Akyar, 2017).
It is Etter’s (2014) work that shines a clear light on specific tactics/strategies adopted by organizations for disseminating CSR communication on Twitter. Etter (2014) puts forward three clear strategies, which can be deployed by organizations on Twitter. The first is the ‘broadcasting’ model which states that information be released without offer of dialogue. The second is ‘reactive’ wherein an organization identifies a query, and addresses it specifically. A ‘proactive’ approach is defined as one where an organization encourages dialogue, calling for stakeholders to present their views, agreements, disagreements, compliments, and criticisms, about a piece of communication that has been disseminated by the organization. The biggest difference between a ‘reactive’ and a ‘proactive’ strategy can be explained by the fact that an organization is actively encouraging dialogue in the latter while volunteering information, as opposed to only communicating information after being ‘prodded’ by stakeholder(s) on a specific topic (Araujo & Kollat, 2018).
Etter (2014) states that a broadcasting strategy of communication is adopted by a majority of organizations on Twitter. According to him, the benefits of such a strategy are obvious, chief being the fact that the narrative is always controlled by the organization. Grunig & Grunig (1992) also voice a fear that the use of ‘reactive’ and ‘proactive’ strategies, essentially the use of dialogue, can threaten organizational power, since they allow the opportunity for multiple narratives to emerge. This can also be said to be the reason why weapons manufacturers and tobacco companies, prefer to operate one-way communication strategies, partly due to the nature of products that they manufacture, but also due to the fact that dialogic communication has not really shown clear benefits across industries (Morsing & Schultz, 2006). It is also notable to consider that though there exists substantial academic literature on the benefits of implementing dialogic communication strategies between organizations and stakeholders (Colleoni, 2013; Etter, 2014; Kollat & Farache, 2017) practical considerations faced by
communications teams of organizations often render them untenable. Building on this, weapons manufacturers and tobacco companies can utilize all three strategies for effective engagement of stakeholders on Twitter for the CSR communication that they disseminate. However, as Byrne (2007) posits, both industries’ incapacity to be perceived as ‘responsible’ due to the nature of the products that they manufacture (Baker, 2005), might lead to these organizations preferring the adoption of a particular strategy. Thus, it is expected that to avoid opening up the organization to criticism other than the one it generally receives, a communication strategy, closely resembling a monologue (broadcast) is adopted instead of one that invites stakeholders to participate in a conversation (proactive).
Stakeholder engagement on Twitter
The presence of an engaged party of stakeholders is essential for most corporate organizations, more so on Twitter. While there exist a number of definitions for the actual term, here we would like to utilize one that Devin & Lane (2014) put forward that concerns an organization meeting, discussing, and responding, to the expectations held by its stakeholders. Engagement at its simplest can be explained as using a piece of communication that hooks stakeholders’ attention and manages to sustain it in a manner beneficial to the organization. Stakeholder engagement if done right, can be an effective method of two-way communication (Brown & Dillard, 2013; Brown, 2009) which can produce meaningful dialogue between organization and stakeholders. However, since only communication on Twitter of certain organizations is being focused on and not their entire communication offerings, it is necessary to state that there are certain impediments in gaining effective stakeholder engagement. Chiefly, disseminating information to a wide swathe of stakeholders (In this case, only followers of the organization’s Twitter handle are reached), crystallizing/distilling/addressing concerns of stakeholders as a whole, and finally, actually
providing a platform for stakeholders to communicate effectively with the organization (Unerman & Bennett, 2004; Manetti & Belluci, 2016).
Stakeholder engagement and its subsequent growth depends on a couple of factors. Organizations need to ensure that a steady stream of communication (In this case CSR communication) is maintained, leading the organization to be more transparent, as is expected with an increased information discharge (Etter & Fiesler, 2010). Lastly, communication in this regard will have to assume the form of a dialogue, where there is contribution from both parties (organization and stakeholders). Both parties have certain benefits through this arrangement. Organizations through a steady stream of information hope to generate complimentary sentiments, leading to behavior that positively impacts the organization and its reputation (Du, Bhattacharya & Sen, 2010). The stakeholders as Fassin (2009) states, have an interest in the actions of the organizations and are always receptive to information as it affects them and their interests. Such a finding is mirrored by the research undertaken by Desy Ratna Yuwita & Sulistyo Kalanjati (2017), albeit in an Indonesian setting where it was found that a steady stream of CSR communication had a positive effect on the organizations’ reputation.
Halpern & Snider (2011) state that arms manufacturers while often selling to private customers, are mostly regarded as agents of the state, and thus their connection to a common individual is scarce. Tobacco companies on the other hand according to Wakefield, Miller & Woodward (1999) are held in low regard, especially by a common individual on the street. Arms manufacturers as Thompson (2013) argues, benefit from a veil of secrecy which is brought on by the fact that the products are sold to governments and these sales are not widely publicized due to national security implications. Thus, the public does not have access
to a stream of information on the sector in the popular media (Thompson, 2013). Thus, as seen earlier, it is in the interest of weapons manufacturers and tobacco companies to attempt to engage with stakeholders, through dialogue on Twitter. An intention to engage in a two-way conversation can also according to Fooks, Gilmore, Collin, Holden & Lee (2012) aid organizations to use CSR communication to better reputation, guide public narrative (in this case through Twitter) towards a preferred direction, and also override government-induced regulation on publicity to present a picture more aligned towards the organization’s benefit to stakeholders. In line with this argumentation, it is hypothesized that;
H1: Use of the broadcasting strategy in CSR Communication results in higher levels of stakeholder engagement for arms manufacturers and tobacco companies as compared to the use of a reactive or a proactive strategy
H2: Use of the proactive strategy in CSR communication leads to a higher number of positive comments for arms manufacturers as compared to tobacco companies
CSR Fit
According to David, Kline, & Dai (2005), the gamut of CSR activities and the sectors in which to operate CSR initiatives is diverse and one with multiple possibilities. According to Carroll’s (1979) definition of CSR - CSR programmes can range from philanthropy (monetary or in kind) to promoting environmental sustainability and conservation. However, it is important that organizations pick or launch CSR initiatives in sectors which seem ‘credible’ and also exhibit a connection to the core workings of the organizations (Nan & Heo, 2007; Menon & Kahn, 2003). Du et al. (2010) have argued that the choice of CSR initiative or the sector does have an influence on the perception of the organization by
stakeholders. Nan et al. (2007) through their study demonstrated that a weak congruence between type of CSR initiative and the organization’s main area of operation can lead to suspicion and ineffective evaluation of CSR efforts and consequently the organization’s reputation.
A comprehensive look at trends about CSR communication on Twitter over the last decade has shown that there are certain areas/sectors that are favored by organizations to launch and publicize CSR initiatives (Forum, 2012; McPherson 2015; McPherson 2017). Sectors like healthy living, social issues, empowerment of disadvantaged classes, and gender parity among others seem to have generated engagement and garnered popularity on Twitter. Topics like environmental conservation/safeguarding, diversity at the workplace, and encouragement and support to young talent have seen to have grown over the last couple of years (McCarthy, Levin & Hunt, 2017). Aqueveque et al. (2018) are of the view that ‘controversial industries’ which include weapons manufacturers and tobacco companies, need to hit upon a ‘CSR fit’ that is ‘transactional’ and ‘legitimacy seeking’. This ‘CSR fit’ according to the authors bears a close relation to core business activities of weapons manufacturers and tobacco companies and that connection is easily visible.
Use of Frames in CSR communication
The concept of framing messages in a particular way has been one of the cornerstones of organizational communication and it is no different with CSR communication. According to Entman’s (1993) definition, framing can best be explained as selecting certain aspects of an issue at hand, training the lens squarely on them, and communicating these issues in a way that highlights particular problems, leads to positive/negative moral evaluation, and causal interpretation. Taking this point further and employing De Vreese’s (2005) study as a
foundational block, the use of frames can provide tools to ensure the interpretation of the subject in line with the intended motive. According to van den Heijkant & Vligenthart (2018), the use of frames in CSR communication is imperative since it allows the organization’s communication team to tweak the dissemination and the potential perception in line with the organization’s goals. Thus, it is not a stretch to imagine that even if an organization’s target behind implementing a CSR initiative is purely economic or to comply with legislation, a tactical employment of a frame might lead to the message being perceived in more gratuitous, benevolent light and can have the power to influence perceptions of the public at large (Waller & Conway, 2011; van den Heijkant & Vliegenthart, 2018).
Carroll’s (1979) work on the use of CSR frames is instructional in the way that it identifies three categories of frames that organizations are most likely to utilise whilst communicating CSR initiatives. Carroll (1979) lists ‘economic responsibility dimension’, ‘ethical responsibility’, and ‘philanthropic responsibility’. Since these three frames are rather broad and encapsulate numerous activities under their aegis, van den Heijkant & Vliegenthart, (2018) takes a further step, breaking down these three frames into a further five - ‘economic performance and growth’, ‘society’s welfare’, ‘environmental protection’, global business activities and responsibilities’, ‘fair and sustainable products’, and ‘educational and charitable program’. These categories of frames, while providing a wider net, also allow for CSR initiatives to be classified as accurately as possible.
Lastly, Fooks et al. (2012) argue that an ingenious use of frames to dress up CSR messages can hold value for an organization’s long-term business strategy and also allow it to gain access to previously barred areas. Fooks et al. (2012) while studying CSR policies and their eventual communication of British American Tobacco (BAT) in the United Kingdom, found
that usage of frames that revolved around ‘public health’, ‘harm reduction’, and ‘educating future generations’, served a larger benefit. The organization found a route into the arena of public debate and governmental departments. Fooks et al. (2012) find that the effective use of frames in CSR communication by BAT, allowed them the means to ultimately further the organization’s commercial interests.
H3: CSR communication of initiatives using the social welfare frame leads to a higher number of positive comments for tobacco companies in comparison with arms manufacturers.
CSR Scepticism
A bogey attached to CSR communication, irrespective of industry and sector, is the scepticism attached to it (Skarmeas & Leonidou, 2013). Scepticism can best be explained as ‘an individual’s tendency to question, doubt, and disbelieve’ (Foreh & Grier, 2003; Boush, Friestad & Rose, 1994). This scepticism in a CSR context according to Skarmeas & Leonidou (2013) can stem from a variety of reasons, chiefly the effect that use of products manufactured by organizations cause, the related negative social impact, and the perceived damage to the environment (Gowri, 2004). Also, as Foreh & Grier (2003) argue, the information available to the public about mismanaged CSR efforts, irrespective of organization, can serve to increase this scepticism. This effect can be amplified when it comes to weapons manufacturers and tobacco companies whose products seem to have a tag of ‘lethality’ attached to them (Byrne, 2007). Negative information about products and services is seen to command more prominence in a public’s conscience as opposed to complimentary information (Carson, 2003; Lange & Washburn, 2012). Considering the stance of Oh, Bae & Kim (2016), the authors argue that ‘sinful firms’ (tobacco and weapons
manufacturers are both classified under this tag by the authors) are likely to fall prey to the fact that scepticism about CSR communication would accelerate the creation of negative opinions about CSR and also the organization in general. In the context of weapons manufacturers, Byrne (2007) argues that in no way should weapons manufacturers be considered responsible, fuelling more scepticism. Weapons manufacturers according to Byrne (2007) fail on environmental and social equity indicators of responsibility, thus according to him any of their CSR communication should be not given any credence. Doward (2018) cites the examples of CSR efforts by arms manufacturers about educational initiatives for schoolchildren coming in for criticism, with an increasing number of commentators calling these initiatives “a push by weapons manufacturers to legitimize their business and ensure normalisation of their existence.”. Palazzo & Richter (2005) say that CSR efforts and communication while used as a ‘legitimising agent’ across other industries, cannot be applied the same way for the tobacco industry as their products run counter to any argument of legitimization.
H4: Arms manufacturers communication about CSR initiatives dealing with social responsibility and educational initiatives receive higher negative comments as compared to
tobacco companies.
Likes, Retweets, and Comments on Twitter
According to Bonson & Ratkai (2013), Twitter-specific functions like Retweets, Likes, and Comments can be an effective indicator of how engaged stakeholders are towards CSR communication of the organizations that are part of the study and can also be deployed as a means of understanding how stakeholders react to particular pieces of communication (Albu & Etter, 2014). The feature of allowing users to comment on communication put out by an
organization is in line with the dialogic model that Etter (2014) and Morsing & Schultz (2006) reiterate. Comments can be an insightful way of understanding the mood of stakeholders and the public towards a particular piece of CSR communication and that responding to comments can be seen to have a positive effect on perception of the communication (Rim & Song, 2013). Del Vicario, Zollo, Caldarelli, Scala & Quattrociocchi (2017) find, comments can optimise the reach of a particular social media post, often amplifying it to a larger audience and can also serve to create perceptions and polarisations through the tone of the comments. Also, as Lee (2014) mentions, empirical findings do exist that point towards social media users evaluating topic/topics favourably or unfavourably based on the comments of people, whom they would or would not necessarily know. Thus, keeping this in mind, we would expect:
H5: Tobacco companies’ CSR communication receives higher negative comments than that of arms manufacturers.
Method
In order to investigate the possible differences in communication strategies adopted by weapons manufacturers and tobacco companies on Twitter with regard to CSR communication, a manual quantitative content analysis was employed of selected content (Tweets) of the four organizations identified as the sample.
Sample
In order to satisfy the empirical requirements of this study, a convenience sample constituted of Tweets disseminated by the official organizational handles of the four organizations (two arms manufacturers and two tobacco companies) was identified. The four organizations were
identified by their ranking from the list of ‘The World’s Biggest Arms Companies in 2017’ published by Forbes (McCarthy, 2017) and ‘Top 10 Tobacco Companies of 2017’ published by Insider Mag (Vatu, 2017). The organizations are - Lockheed Martin, BAE Systems, Phillip Morris International, Imperial Brands PLC. While, Lockheed Martin is an American organization, BAE Systems is a British organization. Similarly, Phillip Morris International identifies as an American organization while Imperial Brands PLC identifies as a British organization. Thus, the sample maintains uniformity in the organizations whose CSR communication is analyzed. A proprietary data extraction tool to collect (scrape) data (Tweets) from the Twitter accounts of the four organizations. Tweets published on 1st January 2017 to 31st December 2017 were included in the sample. The rationale driving the timeframe was to provide this study as contemporary an outlook as possible and also allow efficacy in the coding process.
Table 1 Distribution of sample (Tweets) by organization
Measurement and procedures
Coding Process
Web-based survey creation software ‘Qualtrics’ was used to construct and mount the codebook which was employed for the manual quantitative content analysis. The final sample of 1105 Tweets, was coded by a single coder. An elaborate coding process was instituted and
Twitter Account Number of Tweets
@LockheedMartin 404
@BAESystemsPLC 254
@InsidePMI 396
honed over a test coding phase. The process began with the coder identifying the organization to which the unit (Tweet) belonged to after which it was to be indicated whether the Tweet had a CSR reference. For a Tweet to be identified as CSR, it had to fit the following description - A tweet will qualify as being a CSR-related Tweet if there is no overt marketing approach about the product that the organization produces within the Tweet. Tweets that talk about the products produced by the organization, stock prices, media coverage related to sales and purchases of the products, institutional/management appointments and departures do not qualify as CSR Tweets. While organizations will not explicitly mention the term ‘CSR’ in Tweets, a Tweet qualifies as ‘CSR-related’ when it talks about efforts taken by the organization for the constituents outside of the organization (Etter, 2014; Carroll, 1979). If the unit (Tweet) did not have a CSR reference as per the above definition, the coder was instructed to terminate the coding process and take up the next unit for analysis. Further minute details of the coding process can be found in the codebook (attached in appendix).
Variables and operationalization
The chief objective of this research study was to investigate the differences in CSR communication strategies and their perception, disseminated by four organizations belonging to two sectors - weapons manufacturers and tobacco companies, on Twitter. The independent variable thus employed for this study was ‘CSR strategies’ and to observe the effect it has on the dependent variable ‘perception of CSR communication strategies’. The dependent variable was further broken down into individual elements in order to gain a more focused picture. The independent variable ‘CSR strategies’ was operationalized through an existing research study carried out by Etter (2014) where he lays out three communication strategies that organizations have been observed to employ for communication on Twitter. The three communication strategies - ‘Broadcasting’, ‘Reactive’, ‘Proactive’ are mutually exclusive
from each other and have unique characteristics that set them apart. The ‘broadcasting’ strategy employs the use of communication which simply expresses the message in the form of a statement. Information that is disseminated is clearly ‘stated’ in this communication strategy and no effort at encouraging dialogue is made. The ‘reactive’ strategy as the name suggests is a strategy that is essentially a reaction to a query/concern raised by an individual/organization on Twitter. The use of the ‘reactive’ strategy is clearly identified when Tweets begin with the use of the ‘@’ symbol. Lastly, the ‘proactive’ strategy sees the organization deploy a distinct ‘call to action’ along the lines of ‘read more, click to find out, watch this video now, find us here’ etc. While the ‘proactive’ strategy might feature the use of the ‘@’ sign, the Tweet does not begin with the usage of the sign. The independent variable (present in codebook as Q11) is measured on a nominal level.
‘Perception of CSR communication strategies’ which is the latent dependent variable was operationalized by the use of the numeric variable ‘stakeholder engagement’, ‘positive comments’, and ‘negative comments’. In the absence of a validated scale for ‘stakeholder engagement’ in previous literature, the variable was computed by arriving at the sum of Q12 (number of Likes) and Q13 (number of Retweets). Since Likes and Retweets are the most effective indicators of engagement on Twitter (Haddadi, Dacres, & Perver, 2013) it was decided to utilize them for the computation.
Further variables used in the study are ‘topic of the tweet’ (Q8) which is measured nominally with the use of five different categories - social responsibility, health and safety, environmental sustainability, responsible trade practices, and educational initiatives. These topic-heads are based on the existing categories utilized by Corporate Responsibility Magazine to classify CSR initiatives across sectors. The variable ‘frame employed in the
Tweet’ (Q9) is also measured nominally with five categories - social welfare, environmental protection, global business activities/responsibilities fair trade, and educational charitable philanthropic processes. These frames were employed by making use of van den Heijkant & Vliegenthart’s (2018) study which also employed content analysis and based the above-mentioned typology on Carroll’s (1979) study which was the first to actually put forth the use of these frames in CSR communication. Variable ‘number of Likes’ (Q12) and ‘number of Retweets’ (Q13) were numeric variables along with ‘number of Comments’ (Q14), ‘Positive comments’ (Q15), ‘Negative comments’ (Q16). Positive comments were identified by adhering to the definition - ‘A comment that is laudatory/complimentary of the organization/initiative/Tweet’. Negative comments were identified if they reflected - ‘A comment that is critical of the organization/initiative/Tweet including abusive language.’ These definitions were constructed by referring to the study of Rim & Song (2016) which studied reactions/comments of individuals to CSR efforts on social media and the study of Bae & Lee (2012) which made a distinction between positive and negative comments on Twitter.
Reliability
In order to gauge reliability, a test-coding exercise saw another individual code 50 units (Tweets). This exercise displayed that sufficient reliability was found with all items showing a Krippendorff’s Alpha (KA) value above .7. The highest value found for certain items was 1. Consequently, it is seen that reliability measures were found to be satisfactory for a majority of items. Variables like ‘number of Retweets’, ‘use of associated media’, ‘number of likes’, ‘number of comments’, and ‘to which organization does the Tweet belong to’ displayed a KA value of 1 since the responses to these variables were extracted from a
common dataset, eliminating possible room for variance. All KA values have been reported in the Appendix.
Plan of analysis
In order to investigate the hypotheses, a set statistical tests permitting the effective investigation of each hypothesis was identified. In order to test H1, which sought to investigate whether there was a significant difference between communication strategy used and the engagement it generated, a one-way analysis of variance (ANOVA) was employed. An ANOVA would clearly demonstrate whether there was a case of statistically significant differences between the use of a particular communication strategy and the engagement it subsequently produced. In order to test H2, which posited that the use of the ‘proactive’ strategy led to significant differences in positive comments for weapons manufacturers than tobacco companies, an Independent Samples T-Test was deployed to investigate whether there was a significant difference in the means of the positive comments between weapons manufacturers and tobacco companies. H3 also saw the use of an Independent Samples T-Test, aiming to identify whether a significant difference existed in positive comments for tobacco companies as compared to weapons manufacturers in the case of a specific frame being employed. H4 and H5 were also to be investigated by using the Independent Samples T-Test. In H4’s case, the test sought to find whether a statistically significant difference existed between weapons manufacturers and positive comments when communicating about a particular topic as compared to the tobacco industry. H5 was concerned with probing whether a statistically significant difference existed between negative comments attached to tobacco companies’ CSR communication as compared to weapons manufacturers.
The manual quantitative content analysis of the Tweets posted by the four organizations which formed the sample displayed that approximately every fifth Tweet had a clear CSR-reference (n=346) (31.31%). Since every Tweet with a non-CSR CSR-reference was eliminated from analysis, a total number of 346 Tweets constituted the final sample.
Figure 1. Percentage distribution of CSR tweets by organization
CSR Tweets were sorted into categories with regard to the CSR topic that they ‘referred’ to. Social responsibility (n=133) figured at the top of the pile, followed by education initiatives (n=122), responsible trade practices (n=45), environmental initiatives (n=36), and health initiatives (n=10). The precise break-up of the Tweets by topic is illustrated below;
Figure 2. Distribution of CSR Tweets by topic 0,00% 5,00% 10,00% 15,00% 20,00% 25,00% 30,00% 35,00% 40,00% Lockheed Martin (146 CSR Tweets) BAE Systems (80 CSR Tweets) Philip Morris International (109 CSR Tweets) Imperial Tobacco (51)
Percentage of CSR Tweets per Organization
Engagement, the dependent variable, was calculated by summing the total number of Likes and Retweets that each Tweet generated. The values that were subsequently displayed ranged from 0.50 to 4970 (M=53.53, SD=277.04). As far as communication strategies employed by the organizations are concerned, broadcasting (n=232) (67.05%) ranked on top, proactive was placed at second place (n=99) (28.61 %), while the reactive strategy found a negligible number of takers (n=14) (4.04%). Interestingly, 333 Tweets (96.5%) used associated media - images, videos, GIFs, hyperlinks, while 221 (64.1%) also included a hashtag which is a feature emblematic of Twitter.
Figure 3. Percentage of communication strategy utilization in CSR Tweets 0 20 40 60 80 100 120 140 Social Initiatives (133)
Health and Safety (10) Environmental Initiatives (36) Responsible Trade (45) Educational Initiatives (122)
In order to test H1 which aimed to see whether the use of the ‘broadcasting’ strategy for CSR communication by weapons manufacturers and tobacco companies resulted in higher stakeholder engagement as compared to the use of the ‘reactive’ or ‘proactive’ strategy, a one-way Analysis of Variance (ANOVA) was executed. The results of the one-way within subjects ANOVA showed no significant effect between engagement received by the broadcasting strategy compared to the reactive and proactive strategy F (2, 343) = 2.53, p = .08.
In order to test, H2, which sought to investigate whether the use of the proactive strategy in CSR communication leads to a higher number of positive comments for arms manufacturers as compared to tobacco companies, an independent samples t-test was employed. There was a non-significant difference in the scores for weapons manufacturers (M=1.09, SD=5.12) and tobacco companies (M=.00, SD=.00); t (97) =-1.89, p=.06. Interestingly, tobacco companies received no positive comments on their CSR communication when using the proactive
0 10 20 30 40 50 60 70 80
Broadcasting (67.05%) Reactive (4.04%) Proactive (28.61%)
Communication Strategy utilization
H3 looked at whether CSR communication using the social welfare frame leads to a higher number of positive comments for tobacco companies in comparison to weapons manufacturers. An independent samples t-test displayed no significant difference between weapons manufacturers (M=1.01, SD=5.06) who used the ‘social welfare’ frame as compared to tobacco companies (M=.07, SD=.32) t(130) = 1.62, p=.11 using the same frame. Though weapons manufacturers do receive a higher number of positive comments when using the ‘social welfare’ frame as compared to tobacco companies, the difference is not significant.
H4 aimed to test whether weapons manufacturers CSR communication about initiatives dealing with ‘social responsibility’ and ‘educational initiatives’ received a higher number of negative comments as compared to tobacco companies. An independent samples t-test displayed that there was a statistically significant difference between weapons manufacturers receiving higher number of negative comments when posting CSR communication about the topics ‘social responsibility’ and ‘educational initiatives’ (M=1.05, SD=5.15) as compared to tobacco companies posting about the same topics (M=.09, SD=.28) t(253)=2.62, p=.01 leading us to accept the hypothesis.
H5 investigated whether tobacco companies’ CSR communication receives higher number of negative comments than that of weapons manufacturers. An independent samples t-test showed a statistically significant difference between weapons manufacturers (M=1.01, SD=4.84) receiving a higher number of negative comments than tobacco companies (M=.13, SD=.38) t(343)=1.99, p=.04 leading us to reject the hypothesis.
This study aimed to explore the differences in CSR communication on Twitter of weapons manufacturers and tobacco companies, and also cast a look at the perceptions that this communication generates. The communication strategies considered as part of the study were based on the ones outlined in existing literature by Etter (2014) and also Morsing & Schultz (2006). Etter’s (2014) claim that ‘broadcasting’ as a strategy is the preferred choice of organizations in communicating CSR-related aspects was not upheld by for weapons manufacturers and tobacco companies. While there was a clear preference for the use of the ‘broadcasting’ strategy, as proved by the data, a significant difference between its use and that of other strategies was not found. However, the p value was reasonably close to achieving significance and thus it could be possible that a significant difference between the ‘broadcasting’ strategy could indeed be found with a larger sample size. This aspect, while in line with previous literature, moves away from the point put forward by Grunig & Grunig (1992) that suggests the use of strategies closely resembling ‘proactive’ and ‘reactive’ are not in an organization’s best interests. Both, weapons manufacturers and tobacco companies, deployed the ‘proactive’ strategy, often calling on stakeholders to put forth ideas, suggestions, comments, and critiques. Although findings were insignificant, the use of the ‘proactive’ and ‘reactive strategy are not shunned by organizations, thus lending credence to arguments advanced by Colleoni (2013), Etter (2014), and Kollat & Farache (2017) that dialogic communication is beneficial to organizations in terms of promoting stakeholder engagement, especially on social media.
Use of the ‘proactive’ communication strategy spawning a higher number of positive comments for weapons manufacturers, the findings did not display statistical significance. Brown & Dillard’s (2013) contention that a dialogic method of communication, if effected can lead to positive outcomes, could not be upheld in this study. However once again, the p
value’s proximity to significance does point towards the fact that the ‘proactive’ strategy can indeed be beneficial and provide organizations complimentary outcomes in their communication efforts. Wakefield et al. (1999) argue that tobacco companies are not perceived positively and that finding can be said to be justified in this study since tobacco companies did not receive any positive comments when they deployed the proactive strategy in their CSR communication.
While the data was unable to display significant results in favor of the use of particular frames and resulting difference in positive and negative comments, it can be perceived that the existing literature on CSR fit and the importance of choosing the correct frame is essential for the desired perception of CSR efforts. But, with weapons manufacturers receiving a higher number of negative comments for their CSR efforts, the rationale of Oh et al. (2016) that organizations under their grouping of ‘sinful firms’ can attract greater negative perception for their CSR efforts, if they frequently engage in that type of communication seems to hold. H4’s acceptance can also be tied to findings put forth by Oh et al. (2016), Skarmeas & Leonidou (2016), and Doward (2018) which say that information about mismanaged CSR efforts, the perceived motivation behind these efforts, and negative publicity around products, can all contribute to increasing skepticism towards CSR and its communication.
Following Oh et al’s. (2016) argument, the rejection of H5 wherein it was expected that tobacco companies’ CSR communication would receive a higher number of negative comments, is surprising. Tobacco’s ease of availability across the globe even in societies where there is high regulation, along with significant government-induced cautionary legislation (Paul et al. 2010), would result in expectations of tobacco not being positively
evaluated. However, according to the data, that does not seem to be the case. Though tobacco companies do receive negative comments on their CSR communication, weapons manufacturers receive a significantly higher number. However, this can once again be put down to Oh et al’s. (2016) contention that frequent CSR communication can be counterproductive. Tobacco companies’ dissemination of CSR communication is staggered, and not as prolific as that of weapons manufacturers.
In conclusion, while this study was largely unable to obtain significant results or findings, these can partly be explained by several limitations that were encountered in the course of this study. A set of recommendations, which could perhaps lead to a more comprehensive study, are discussed in the following section.
Limitations and Recommendations for Future Research
One of the first limitations that this study faced seemed to be the sample size. The study analyzed 1104 Tweets which though comprehensive in number were whittled down to a final 345 units. These 345 units were Tweets with an actual CSR reference and were utilized for analysis. It is recommended that future research use a larger sample, spread over a larger timeframe in order to possibly come through with significant results. The reason for utilizing the 1104 Tweets in the sample were dictated by constraints of time, efficacy, and resources.
The activity levels of the organizational Twitter accounts that made up the sample also differed. While Lockheed Martin posted over 400 Tweets, Imperial Tobacco posted only 51 in the timeframe that was considered for the study. While there are arguments in previously published literature, specifically Etter (2013), for considering only CSR communication accounts as part of the sample, doing so in this study would have disturbed the uniformity of
the sampled organizations. It was observed that there was a very high number of neutral comments on Tweets posted by the organization. Comments were classified as neutral when they did not make any reference to the content posted as part of the original Tweet. This proliferation of neutral/unrelated comments led to the creation of some ambiguity about the actual engagement levels of the users who posted these comments. Were they indeed engaging with the content that was posted by the organization?
Implications and Conclusion
Though Etter (2014), Araujo & Kollat (2018) and Morsing and Schultz (2006) endorse the ‘proactive’ strategy as the one that is most profitable for enhancing stakeholder engagement. The study found that the ‘broadcasting’ strategy, whose ethos lies in one-way communication without possibility of dialogue, is the one that is preferred. The findings of this study do indeed throw up an array of interesting implications regarding the use of strategies, the motivations behind their use, and the response that the communication tailored along these strategies receives. While Etter (2014) argues that the ‘proactive’ strategy will have beneficial implications in stakeholder engagement for organizations, weapons manufacturers and tobacco companies do not necessarily share the same view, as evidenced by their propensity to go for the rather ‘hands-off’ ‘broadcasting strategy. This can be put down to Morsing & Schultz (2006) argument that a tendency to reveal too much information, indulge in increased levels of sense-making through the encouragement of dialogue, might be counterproductive to organizational goals (Grunig & Grunig 1992). This point is further amplified on the dynamic arena that is Twitter, where organizations do not exercise complete control on who views their CSR communication initiatives (Albu & Etter, 2014; Araujo & Kollat, 2018). The ‘reactive’ strategy though as the data shows, is shunned by both weapons manufacturers and tobacco companies. This intransigence can perhaps be attributed to the
fact that these organizations do not want to single out users/stakeholders in communicating CSR initiatives (Araujo & Kollat, 2018). The ‘reactive’ strategy at its core sees organizations identifying individual queries addressed to it and then provide a response. Instead, organizations seem to prefer communicating to their full share of stakeholders (in this case Twitter followers) and achieve greater visibility for their message. This sheds light on the fact that whilst the ‘reactive’ strategy is recommended for effective communication, its practical value is questionable, in line with Colleoni’s (2013) argument that there exists a chasm between academic models and communication realities. The operationalization of these three communication strategies is not clearly defined in past literature (Albu & Etter, 2016; Vernuccio, 2014). This study tried to establish as clear a division between the three as possible but social media’s constantly fluid nature can often result in amalgamations, a ‘borrowing of bits and bobs’ from other strategies which adds to the haze around the efficacy of communication strategies on Twitter.
Organizations being more receptive to the use of the ‘proactive’ strategy, can be explained by the attendant benefits that its attributes of transparency and posture of being perceived as ‘open to dialogue’ (Etter, 2014; Kollat & Farache, 2017). Trends in the data analyzed, show that using the ‘proactive’ strategy can in fact encourage conversation, not all of it negative. This is also closely tied to the aspect of weapons manufacturers and tobacco companies preferring the use of certain frames and topics to present their CSR communication (van den Heijkant & Vliegenthart, 2018; Nan & Heo, 2007). Both weapons manufacturers and tobacco companies are organizations with a global footprint and often a global workforce. The inclination to communicate about CSR initiatives using the frame ‘social welfare’ is high. This particular frame encompasses a wide variety of initiatives, for example support to veterans, celebrating and promoting diversity at the workplace, lauding volunteering efforts
that aim to benefit society, etc. The two weapons manufacturers, placed a substantial amount of focus on educational initiatives and the promotion of certain streams of education like STEM (Science, Technology, Engineering, Mathematics) along with a focus on veteran benefit programmes. Since both organizations are engaged in the manufacture of technologically superior products which are sold to national militaries, this propensity to fund initiatives which ultimately are closely related to its core business is in line with Du et al’s. (2012) findings about CSR fit and its eventual success being down to the relevancy and connection of the initiative to the organization’s core goals (Fooks et al. 2012). Similarly, it was observed the Phillip Morris International placed emphasis on fair trade initiatives which dealt with helping farmers gain better yields, sensitization to current trade trends, and cooperating on initiatives that cracked down on illicit trade. These initiatives seem to tie in with Phillip Morris International’s stated commercial goals, help in streamlining business procedures, and ultimately work towards enhancing its organizational reputation, which is the stated goal of CSR (Fooks et al. 2012; Hirschhorn, 2004).
A largely negative perception of CSR communication on Twitter, an albatross around the neck of organizations operating in these two sectors, is constant (Oh et al. 2016). This study’s findings elaborate that while both weapons manufacturers and tobacco companies face almost-constant backlash in the form of negative comments on and around their CSR communication, the weapons manufacturers seem to be under more intensive attack (Oh et al. 2016). Tobacco companies’ products are accessible to almost every individual across the planet whereas it is highly inconceivable for a private citizen to have the opportunity of purchasing a fighter aircraft or a frigate. But weapons manufacturers’ perceived collaboration with large-scale violence in terms of armed conflict seems to play a decisive role in the
number of negative comments directed at the organizations operating in this sector (Gowri, 2004).
While there is much more to CSR communication by weapons manufacturers and tobacco companies than meets the eye, it is perhaps useful for communication practitioners in both industries to pay closer attention to the ‘less is more’ adage. As Morsing & Schultz (2006) say, too much communication about CSR might turn out to be counterproductive. Thus, it is important to not go overboard with CSR communication, as this has the potential to strengthen skepticism towards this communication (Oh et al. 2016). While organizations face a constant challenge, straddling the line between skepticism and authenticity when it comes to CSR communication. This delicate balance becomes even finer when viewed through the lens of controversial industries like weapons manufacturing and tobacco (Oh et al. 2016), with the products they make ensuring that they continue to remain in the eye of the storm even while attempting to carry out responsible activities (Aqueveque et al. 2017).
Bibliography
Albu, O. B., & Etter, M. (2015). Hypertextuality and Social Media. Management Communication Quarterly, 30(1), 5-31. doi:10.1177/0893318915601161
Aqueveque, C., Rodrigo, P., & Duran, I. (2018). Be bad but (still) look good: Can controversial industries enhance corporate reputation through CSR initiatives? Business Ethics: A European Review, 27(3), 222-237. doi: 10.1111/beer.12183
Araujo, T., & Kollat, J. (2018). Communicating effectively about CSR on Twitter. Internet Research, 28(2), 419-431. doi: 10.1108/intr-04-2017-0172
BAE Systems. (2015). BAE Systems Corporate Responsibility Summary 2015.
Bae, Y., & Lee, H. (2012). Sentiment analysis of twitter audiences: Measuring the positive or negative influence of popular twitterers. Journal of the American Society for Information Science and Technology,63(12), 2521-2535.
Baker, M. (2005). Can companies that make products that kill be socially responsible? Retrieved from http://mallenbaker.net/article/clear-reflection/can-companies-that-make-products-that-kill-be-socially-responsible
Bennett, O. (2017). Corporate social responsibility: ‘Hey everyone, smoke our brand of cigarettes and together we’ll save the environment’. The Independent.
Bonsón, E., & Ratkai, M. (2013). A set of metrics to assess stakeholder engagement and social legitimacy on a corporate Facebook page. Online Information Review, 37(5), 787-803.
http://dx.doi.org/10.1108/oir-03-2012-0054
Brenkert, G. (2000). Social Products Liability: The Case of the Firearms Manufacturers. Business Ethics Quarterly, 10(1), 21. doi: 10.2307/3857691
Brown, J., & Dillard, J. (2013). Critical accounting and communicative action: on the limits of consensual deliberation, Critical Perspectives on Accounting, 24(3), pp. 176-190.
Byrne, E. (2007). Assessing Arms Makers’ Corporate Social Responsibility. Journal Of Business Ethics, 74(3), 201-217. http://dx.doi.org/10.1007/s10551-006-9228-9
Carroll, A. (1979). A Three-Dimensional Conceptual Model of Corporate Performance. The Academy Of Management Review, 4(4), 497. http://dx.doi.org/10.2307/257850
Carroll, A. (1991). The pyramid of corporate social responsibility: Toward the moral management of organizational stakeholders. Business Horizons, 34(4), 39-48. doi: 10.1016/0007-6813(91)90005-g
Dacres, S., Haddadi, H., & Purver, M. (2013). Topic and Sentiment Analysis on OSNs: A Case Study of Advertising Strategies on Twitter.
David, P., Kline, S., & Dai, Y. (2005). Corporate Social Responsibility Practices, Corporate Identity, and Purchase Intention: A Dual-Process Model. Journal Of Public Relations Research, 17(3), 291-313. doi: 10.1207/s1532754xjprr1703_4
Davis, J. (1995). The Effects of Message Framing on Response to Environmental Communications. Journalism & Mass Communication Quarterly, 72(2), 285-299. doi: 10.1177/107769909507200203
De Vreese, C.H. News framing: Theory and typology. (2005). Identifying Information And Tenor In Texts, 13(1), 51-62. doi: 10.1075/idjdd.13.1.06vre
Del Vicario, M., Zollo, F., Caldarelli, G., Scala, A., & Quattrociocchi, W. (2017). Mapping social dynamics on Facebook: The Brexit debate. Social Networks, 50, 6-16. doi:
10.1016/j.socnet.2017.02.002
Deloitte. (2017). 2017 Global aerospace and defense sector financial performance study. Deloitte Publication
Desy Ratna Yuwita, A., & Sulistyo Kalanjati, D. (2017). The association between corporate social responsibility disclosure of cigarette company and company’s financial
performance. SHS Web Of Conferences, 34, 04004. doi: 10.1051/shsconf/20173404004
Devin, B.L. & Lane, A. B. (2014). Communicating Engagement in Corporate Social Responsibility: A Meta-Level Construal of Engagement, Journal of Public Relations Research, 26(5), pp. 436-454.
Dorfman, L., Cheyne, A., Friedman, L., Wadud, A., & Gottlieb, M. (2012). Soda and Tobacco Industry Corporate Social Responsibility Campaigns: How Do They
Compare?. SSRN Electronic Journal. doi: 10.2139/ssrn.2503131
Doward, J. (2018). Arms industry spends millions to promote brands in schools. Retrieved from https://www.theguardian.com/world/2018/sep/01/arms-industry-spending-millions-normalise-weapons-in-schools
Du, S., & Vieira, E. (2012). Striving for Legitimacy Through Corporate Social
Responsibility: Insights from Oil Companies. Journal Of Business Ethics, 110(4), 413-427. doi: 10.1007/s10551-012-1490-4
Du, S., Bhattacharya, C., & Sen, S. (2010). Maximizing Business Returns to Corporate Social Responsibility (CSR): The Role of CSR Communication. International Journal Of
Management Reviews, 12(1), 8-19. doi: 10.1111/j.1468-2370.2009.00276.x
Entman, R. (1993). Framing: Toward Clarification of a Fractured Paradigm. Journal Of Communication, 43(4), 51-58. doi: 10.1111/j.1460-2466.1993.tb01304.x
Etter, M & Fieseler, C.(2010). On Relational Capital in Social Media. Journal of the Swiss Association of Communication and Media Research, Vol.10, pp.167-189.
Etter, M. (2013). Reasons for low levels of interactivity. (Non-) interactive CSR communication in twitter. Public Relations Review, 39(5), pp. 606–608.
Etter, M. (2014). Broadcasting, reacting, engaging – three strategies for CSR communication in Twitter. Journal of Communication Management, 18(4), pp. 322–342.
https://doi.org/10.1108/JCOM-01-2013-0007
Fieseler, C., & Meckel, M. (2010). Corporate Social Responsibility in the Blogosphere, pp. 599–614. https://doi.org/10.1007/s10551-009-0135-8
Fooks, G., Gilmore, A., Collin, J., Holden, C., & Lee, K. (2012). Erratum to: The Limits of Corporate Social Responsibility: Techniques of Neutralization, Stakeholder Management and Political CSR. Journal Of Business Ethics, 112(2), 367-367. doi: 10.1007/s10551-012-1357-8
Foreh, M., & Grier, S. (2003). When Is Honesty the Best Policy? The Effect of Stated Company Intent on Consumer Skepticism. Journal Of Consumer Psychology, 13(3), 349-356. doi: 10.1207/s15327663jcp1303_15
Forum, F. L. (2012, July 24). The Top 10 Trends in CSR for 2012. Retrieved May 22, 2018, from https://www.forbes.com/sites/forbesleadershipforum/2012/01/18/the-top-10-trends-in-csr-for-2012/#2ee2ba81701c
Frederick, W. (2008). Commentary: Corporate Social Responsibility: Deep Roots, Flourishing Growth, Promising Future. Frontiers In Psychology, 7. doi:
Gowri, A. (2004). When Responsibility Can't Do It. Journal Of Business Ethics, 54(1), 33-50. doi: 10.1023/b:busi.0000043497.23400.00
Grover, A. (2014). Importance of CSR in Inclusive Development. Procedia - Social And Behavioral Sciences, 157, 103-108. doi: 10.1016/j.sbspro.2014.11.013
Grunig, J. E., & Grunig, L. A. (1992). Models of public relations and communication. In J. E. Grunig (Ed.), Excellence in public relations and communication management (pp. 285-325). Hillsdale, NJ: LEA.
Halpern, B., & Snider, K. (2011). Products That Kill and Corporate Social Responsibility. Armed Forces & Society, 38(4), 604-624.
http://dx.doi.org/10.1177/0095327x11415490
Hirschhorn, N. (2004). Corporate social responsibility and the tobacco industry: hope or hype?. Tobacco Control, 13(4), 447-453. doi: 10.1136/tc.2003.006676
Imperial Brands PLC. (2017). Imperial Brands PMC Growing Our Business Responsibly, Sustainability Report 2017. Retrieved from
http://www.imperialbrandsplc.com/content/dam/imperialbrands/corporate/responsibility/app roach-and-performance/Sustainability_Report_2017.pdf
Kemp, G. (1994). The Continuing Debate Over U.S. Arms Sales: Strategic Needs and the Quest for Arms Limitations. The ANNALS Of The American Academy Of Political And Social Science, 535(1), 146-157. doi: 10.1177/0002716294535001011
Kollat, J., & Farache, F. (2017). Achieving consumer trust on Twitter via CSR communication. Journal of Consumer Marketing, 34(6), pp. 505–514.
https://doi.org/10.1108/JCM-03-2017-2127
Lee, J. (2014). Are Some People Less Influenced by Others' Opinions? The Role of Internal Political Self-Efficacy and Need for Cognition in Impression Formation on Social
Networking Sites. Cyberpsychology, Behavior, And Social Networking, 17(9), 571-577. doi: 10.1089/cyber.2013.0713
Lee, K., Oh, W., & Kim, N. (2013). Social Media for Socially Responsible Firms: Analysis of Fortune 500’s Twitter Profiles and their CSR/CSIR Ratings. Journal Of Business
Ethics, 118(4), 791-806. doi: 10.1007/s10551-013-1961-2
Lockheed Martin. (2017). Lockheed Martin 2017 Sustainability Report. Retrieved from https://www.lockheedmartin.com/content/dam/lockheedmartin/eo/documents/sustainability/L ockheed_Martin_Sustainability_Report_Full_2017.pdf
Manetti, G., Bellucci, M., & Bagnoli, L. (2017). Stakeholder engagement and public
information through social media: a study of Canadian and American Public Transportation Agencies, The American Review of Public Administration, 47(8), pp. 991-1009.
Mayer-Sommer, A., & Roshwalb, A. (1996). An examination of the relationship between ethical behavior, espoused ethical values and financial performance in the U.S. defense
industry: 1988?1992. Journal Of Business Ethics, 15(12), 1249-1274. doi: 10.1007/bf00411812
McCarthy, N. (2017). The World's Biggest Arms Companies [Infographic]. Retrieved from
https://www.forbes.com/sites/niallmccarthy/2017/12/11/the-worlds-biggest-arms-companies-infographic/#4acbe7b4475a
McPherson, S. (2015, January 04). Eight CSR Trends to Watch Out For in 2015. Retrieved May 22, 2018, from https://www.forbes.com/sites/susanmcpherson/2014/12/31/five-csr-trends-to-watch-out-for-in-2015/#3315619d59e0
Menon, S., & Kahn, B. (2003). Corporate Sponsorships of Philanthropic Activities: When Do They Impact Perception of Sponsor Brand?. Journal Of Consumer Psychology, 13(3), 316-327. doi: 10.1207/s15327663jcp1303_12
Morsing, M., & Schultz, M. (2006). Corporate social responsibility communication:
stakeholder information, response and involvement strategies. Business Ethics: A European Review, 15(4), pp. 323–338. https://doi.org/10.1111/j.1467-8608.2006.00460.x
Nan, X., & Heo, K. (2007). Consumer Responses to Corporate Social Responsibility (CSR) Initiatives: Examining the Role of Brand-Cause Fit in Cause-Related Marketing. Journal Of Advertising, 36(2), 63-74. doi: 10.2753/joa0091-3367360204
Oh, H., Bae, J., & Kim, S. (2016). Can Sinful Firms Benefit from Advertising Their CSR Efforts? Adverse Effect of Advertising Sinful Firms’ CSR Engagements on Firm
Palazzo, G., & Richter, U. (2005). CSR Business as Usual? The Case of the Tobacco Industry. Journal Of Business Ethics, 61(4), 387-401. doi: 10.1007/s10551-005-7444-3
Paul, C., Mee, K., Judd, T., Walsh, R., Tang, A., Penman, A., & Girgis, A. (2010).
Anywhere, anytime: Retail access to tobacco in New South Wales and its potential impact on consumption and quitting. Social Science & Medicine, 71(4), 799-806. doi:
10.1016/j.socscimed.2010.05.011
Phillip Morris International. (2017). Phillip Morris International Annual Sustainability Report 2017. Retrieved from https://www.pmi.com/docs/default-source/pmi-sustainability/pmi-sustainability-report-2017.pdf
Review, 19(3), 219-243. doi: 10.1057/s41299-016-0004-1
Rim, H., & Song, D. (2013). The Ability of Corporate Blog Communication to Enhance CSR Effectiveness: The Role of Prior Company Reputation and Blog
Responsiveness. International Journal Of Strategic Communication, 7(3), 165-185. doi: 10.1080/1553118x.2012.738743
Rim, H., & Song, D. (2016). “How Negative Becomes Less Negative”: Understanding the Effects of Comment Valence and Response Sidedness in Social Media. Journal Of
