Activating the sleeping mass:
the ability of the government and NGOs to influence
the salience of consumers
Master Thesis
MSC International Business and Management
Jesse Rutten S2351641
University of Groningen
Faculty of Economics and Business
Supervisor:
Dr. Halaszovich
2 ABSTRACT
Abstract Purpose – MNEs prioritize their stakeholders according to the salience they hold. The salience consists of three individual attributes; power, legitimacy and urgency. Although consumers by nature possess the attribute of power as is perceived by firms, consumers do not always make use of this power even while there are some ethical issues at play that do not align with their beliefs. The purpose of this paper is to examine whether other stakeholders, in this case the government and NGOs, can ‘activate’ consumers’ legitimacy and urgency by increasing the CSR awareness of consumers.
Methodology – A survey questionnaire was electronically distributed with a total of 169 respondents. A one-way ANOVA followed by a Tukey-test were conducted to analyze the conceptual framework.
Findings – The findings indicate that it is possible for other stakeholders to increase the salience of consumers through CSR awareness. The effect for the government was positive and significant and in line with their current role in the field of International Business. Despite the growing importance of NGOs in the current environment, the results indicate that NGOs are not yet able to influence the sali-ence of consumers to a significant amount.
Managerial implications – Managers of governmental entities, NGOs and MNEs should reassess and further investigate their stakeholder environment and especially their relationship with consumers. Whereas the influence of the government is significant and provides a potential new direction of oating, NGO’s managers could obtain value if they further investigate how they can influence the per-ception of consumers. For MNE’s managers, they should re-evaluate their current environment and should investigate if there are more dynamic relationships within their current stakeholder environment.
Research limitations – The generalizability of the findings is limited due to the exclusion of managerial perceptions into the full equation. Furthermore, only a limited amount of stakeholders are investigated where the focus was on stakeholders from the Western world.
Originality— Whereas most research regarding stakeholder salience approached to topic from a firm perspective, this paper utilized an individual approach to dive into the perception of consumers from a stakeholders’ perspective. Furthermore, previous literature treated the attributes in general as ready states, whereas this paper treats them as dynamic variables in order to make it more applicable for real business environments.
Keywords – CSR, stakeholder salience, power, legitimacy, urgency, CSR awareness
3 ACKNOWLEDGEMENTS
4 TABLEOFCONTENTS ABSTRACT ... 2 ACKNOWLEDGEMENTS ... 3 TABLE OF CONTENTS ... 4 LIST OF FIGURES ... 6 LIST OF TABLES ... 6 1. INTRODUCTION ... 7 2. LITERATURE REVIEW ... 10 2.1 Stakeholder salience ... 10 2.1.1 Power ... 12 2.1.2 Legitimacy ... 12 2.1.3 Urgency ... 14 2.2 Stakeholders journey ... 14 2.2.1. Consumers ... 15 2.2.2. Awareness ... 16
2.3 Impact of NGOs and Government ... 17
2.3.1. Government ... 17
2.3.2. NGO... 20
2.4 Multiplication... 22
2.5 Conceptual model... 23
3. METHODOLOGY ... 24
3.1 Context of the research ... 24
3.1.1 Belgium ... 24
3.2.2 Current case ... 25
3.2 Data collection ... 25
3.3 Sample target group... 26
5
3.8 Questionnaire design ... 33
4. RESULTS ... 35
4.1 Preliminary data analyses ... 35
4.1.1 Sample description... 35
4.1.2 Descriptive statistics ... 36
4.1.3 Reliability - Cronbach’s Alpha ... 37
4.2 Hypothesis testing ... 38
4.2.1 Checking suitability one-way ANOVA ... 38
4.2.2 One-way ANOVA ... 40
4.3 Sensitivity and Robustness check ... 42
4.3.1 Personal relevance ... 43 4.3.2 Age ... 44 4.3.3 Trust ... 45 5. DISCUSSION ... 47 5.1 Government ... 47 5.2 NGOs... 47 5.3 Multiplication... 48 5.4 Sensitivity check ... 49 6. CONCLUSION... 51 6.1 Theoretical contributions ... 51 6.2 Managerial contributions ... 51 6.3 Limitations ... 52 6.4 Future research ... 52 7. REFERENCES ... 54 8. APPENDICES ... 62 Appendix A: Survey... 62
Appendix B: Belgian government e-mail ... 72
Appendix C : Assumptions one-way ANOVA ... 74
Appendix D: one-way ANOVA ... 78
6 LISTOFFIGURES
FIGURE 2.1 TYPOLOGY OF STAKEHOLDERS ... 16
FIGURE 2.2 CONCEPTUAL MODEL ... 23
LISTOFTABLES TABLE 3. 1 MEASURE AND ITEM SPECIFICATION... 29
TABLE 4. 1 SAMPLE CHARACTERISTICS... 36
TABLE 4. 2 DESCRIPTIVES ... 37
TABLE 4. 3 CRONBACH’S ALPHA FOR ALL CONSTRUCTS ... 38
TABLE 4. 4 RESULTS OF ONE-WAY ANOVA ... 40
TABLE 4. 5 MULTIPLE COMPARISONS ... 41
TABLE 4. 6 GROUP STATISTICS PERSONAL RELEVANCE ... 43
TABLE 4. 7 NEW CATEGORIES AGE ... 45
TABLE 4. 8 TRUST GROUPS ... 46
7 1. INTRODUCTION
Over the last decades, Corporate Social Responsibility (hereafter: CSR) increased in im-portance for business worldwide, where CSR should be seen as a construct which is in a ‘state of emergence’ (Crane, McWilliams, Matten, Moon & Siegel, 2008: 7). Meaning that while the broad intention and goal of CSR is clear, in terms of theoretical and empirical underpinning there is work to do. This paper will add to this ongoing process by investigating the dynamics in the stakeholder environments that Multinational Enterprises (hereafter: MNE) have to cope with nowadays. The model of stakeholder salience (Mitchell, Agle & Wood, 1997) will be uti-lized and this model plays a fundamental role in relating the existing theory of stakeholders to CSR (Neill and Stovall, 2005). The model is used as a tool to analyze the stakeholder environ-ment of firms in order to investigate how important the different stakeholders from their envi-ronment are to the firm and how to prioritize these stakeholders based on values for three at-tributes, namely legitimacy, urgency, and power. As various scholars indicated, showing so-cially responsible behaviour should bring prosperity to the firm (Costa and Menichini, 2013; Peterson, 2004). However, the ability to derive positive gains from CSR is dependent upon how their stakeholders perceive this socially responsible behaviour (Costa and Menichini, 2013; Pe-terson, 2004), thereby illustrating the importance of having a significant and positive relation-ship with the different stakeholders of a firm when conducting business and CSR. Intuitive, it already makes sense that not all stakeholders are perceived as equally important by the firm and even further, the firm’s resources to respond to all stakeholders are limited. This is the point where the model of stakeholder salience found its origins and although it has been used for many purposes, its main use lies in the prioritization of stakeholders (Parent and Deephouse, 2007; Boesso and Kumar, 2009; Harvey and Schaefer, 2001).
attrib-8 utes of this potentially high salient stakeholder and add to the literature by arguing that stake-holders’ salience is not only determined by the firm but by factors outside the firm as well. Overall, only possessing the attribute of power as a stakeholder is not sufficient to influence firms (Mitchell et al.,1997). There has to be a need for exercising this power based on a legiti-mate reason to gain authority. The third attribute, urgency, determines whether the authority will be used.
As the factor of ‘activation’, this paper will dive into the literature regarding CSR-awareness, e.g. the level of knowledge someone possesses about certain CSR-activities. A con-sequence of a lack of awareness of an MNE’s CSR-activities is described as the reason for the difference between the attitude of consumers regarding CSR and their actual behaviour (Mohr, Webb & Harris, 2001; Bhattacharya and Sen, 2004). Furthermore, the paper will investigate two other stakeholders, the government and non-governmental organizations (hereafter, NGO) as ‘bringers’ of CSR awareness and will investigate their impact on consumers. Whereas the government occupies a more central role in consumers’ life and the stakeholder environment of many firms (Neville and Mungec, 2006), NGOs are a relatively new player in the international arena and have a different position towards consumers and firms when compared to the gov-ernment (Poret, 2014). However, where it can be said that the role and position of NGOs in international CSR is on the rise, the position of the government is subject to heavy pressures, because business is crossing more and more borders these days whereas the government is somewhat limited in its ability to cross borders. It will be interesting to see how the development and current state of both stakeholders will affect the individual perceptions of consumers.
9 investigated how stakeholders view each other as motivators for CSR. Furthermore, the article by Henriques and Sharma (2005) lists numerous influence-strategies used by stakeholders to influence the firm directly or indirectly through other stakeholders. This indirectly type of in-fluencing, which is also recognized by Neville, Bell and Whitwell (2011) and is being referred to as stakeholder multiplicity (Neville and Menguc,2006), illustrates how less salient stakehold-ers can potentially still achieve their goal by exerting pressure or influencing other stakeholdstakehold-ers who are (potentially) more salient. As Mitchell et al. (1997) already mentioned, the model of stakeholder salience should include potential relationships and possession of the attributes as well.All these theoretical aspects combined provides the following research question:
“Are the government and NGOs able to increase the salience of consumers through
CSR-awareness?”
10 2. LITERATUREREVIEW
2.1STAKEHOLDER SALIENCE
One of the first definitions of CSR was provided by Jones (1980), who described CSR as "the notion that corporations have an obligation to constituent groups in society other than stock-holders and beyond that prescribed by law or union contract, indicating that a stake may go beyond mere ownership" (Jones, 1980: 59). The question that arises is of what or whom those groups consist and how important these groups and their interests are to the firm. These groups are defined in literature as stakeholders, which is the definition I will utilize in this paper. Stake-holders have become a widely recognized phenomenon within the literature. Intentionally meant to provide full theoretical status for the stakeholder theory, the theory of stakeholder identification and salience was created by Mitchell et al. (1997). The goal of the model was to provide guiding on how managers could first identify stakeholders in their environment, which is seen as a crucial determinant of their survival (Sardinha, Craveiro, & Milheiras, 2013), and hereafter, prioritize them according to their importance.
The identification of stakeholders is based on existing literature in the area of stake-holder theory; stakestake-holders have been described by using more precise definitions such as "groups to whom the corporation is responsible" or as entities “in relationship with an organi-zation” (Mitchell et al., 1997: 856). A more widely accepted and used definition, however, is the broader one used by Freeman: “A stakeholder in an organization is (by definition) any group or individual who can affect or is affected by the achievement of the organization's objectives" (Freeman, 1984: 46). According to this definition, it can be reasoned that any group, in theory, can be classified as a stakeholder of a firm. The ranking of how important different stakeholders are to the firm, and therefore, the amount of attention managers should allocate to these stake-holders is being referred to as the salience of stakestake-holders. To assess how salient stakestake-holders are, Mitchell et al. (1997) provided three so-called ‘stakeholder attributes’ that are used to de-scribe and analyze the relationships between a firm’s managers and their stakeholders.
11 determined how it could be that salience could diminish or increase over time from the perspec-tive of a firm’s managers:
“1. Stakeholder attributes are variable, not steady state.
2. Stakeholder attributes are socially constructed, not objective, reality.
3. Consciousness and wilful exercise may or may not be present.” (Mitchell et al., 1997: 868)
To elaborate, the first characteristic could lead to serious damage for a firm if managers would only perceive the attributes as present or absent. Improper acknowledgment could lead to improper replies by the firm meaning that one could overvalue or undervalue individual stakeholders. Moreover, it is widely accepted through the literature that context plays a vital role in assessing the salience of stakeholders. To illustrate, Buysse and Verbeke (2003) con-cluded that MNEs who were more active in terms of environmental procedures perceived more stakeholders as being relevant to their firm. Jawahar and McLaughlin (2001) plead that the salience of specific stakeholders could fluctuate based on the stage (start-up vs. maturity) that the firm was currently in. Whereas initially it was declared that the attributes were similar like a summation, meaning that stakeholders were perceived increasingly salient as they gained more attributes, later it was found through empirical research that this does not necessarily hold. The findings of Parent and Deephouse (2007) illustrate that there is a ranking among the indi-vidual attributes. Based on the discussions provided in the literature, Neville, Bell and Whitwell (2011) constructed an updated definition of stakeholder salience:
“Stakeholder salience is the prioritization of stakeholder claims by managers based on their perception of the degree of power of the stakeholder and the degree of moral legitimacy and urgency of the claim.” (: 369)
12 2.1.1POWER
The first attribute that stakeholders could possess is power. Mitchell et al. (1997) agreed with previous literature and stated that "power may be tricky to define, but it is not that difficult to recognize: '[it is] the ability of those who possess power to bring about the outcomes they desire' (Mitchell et al.,1997: 865) and is based on the resource-dependency theory. However, this def-inition does not provide any arguments on how power can be applied and where it originates from. Etzioni (1964) provided categories of power based on the nature of the resource adopted; Coercive, utilitarian and normative power are mentioned as the three pillars. Coercive power consists of physical resources, whereas utilitarian power comprises of material and financial resources. Normative power is based on symbolic resources. Following this line of thought, an actor in a relationship has power when he possesses access to one of the three sources. Follow-ing research on this attribute by Wasserman & Galaskjewicz (1994) stated that power should be linked to social network theory by explaining that the position of a stakeholder within a network and the density of this network could determine the influence one possesses, which can be related to a network of stakeholders.
A significant finding by Parent and Deephouse (2007) relating to the power stakeholders could possess, entails that through their research it was found that the attribute of power is, compared to legitimacy and urgency, the most influential when determining salience. In other words, managers tend to give more priority to stakeholders that possess power compared to stakeholders who do not possess power. Furthermore, of the three before mentioned sources of power (coercive, utilitarian & normative) it was found that utilitarian power had, by a great extent, the most influential on the salience perception of stakeholders by managers. This finding departs from the original view of Mitchell et al. (1997), by showing that salience is not merely an aggregate of the individual attributes, but is subject to varying interplays between the attrib-utes.
2.1.2LEGITIMACY
13 that comprises of three levels of analysis which are the individual, organizational, and societal level (Wood, 1991).
Based on various demands from the literature stating this attribute is need of reassess-ment due to its troublesome nature and vagueness (Driscoll and Starik, 2004; Phillips, 2003), Neville, Bell and Whitwell (2011) provided an extensive redesign of the concept of legitimacy. Two problems were highlighted with the view adopted by Mitchell et al. (1997), the first prob-lem exposes that there is no clear separation between the legitimacy of the stakeholder itself and that of the claim (e.g. issue) of the stakeholder. As argued and found by Easley and Lenox (2006), both ‘types’ of legitimacy will have a different effect on salience of the stakeholders and therefore, both should be taken into consideration. Contrasting to this statement, Neville, Bell and Whitwell (2011) advance on this reasoning and argue that the legitimacy of the claim should be prioritized and that the legitimacy of the stakeholder itself is of far lesser importance. These authors argue that the legitimacy of the stakeholder does not provide enough information about the actual issue that managers must respond to. Although a stakeholder can be illegiti-mate, if it’s claim is morally legitimate it should enjoy consideration of the managers because “it is the right thing to do and because other stakeholders are likely to take up the cause and pressure the organization if it is not addressed” (Neville, Bell & Whitwell; 2011: 363). The second problem concerns the compound definition used by Mitchell et al. (1997) because it entails not just one but numerous dimensions. Neville, Bell and Whitwell (2011) claim that only the facet of moral legitimacy should be utilized within the model of stakeholder salience. They proposed to remove the pragmatic aspect from the concept of legitimacy due to partial overlap with the attribute of power. Moreover, cognitive legitimacy is due to its nature not relevant for legitimacy within the stakeholder salience model since it does not inform managers about the process of the evaluation of issues.
Their revised concept of legitimacy will go as follows: “The moral legitimacy of a
stake-holder’s claim is an assessment by managers of the degree to is which a claim exceeds a thresh-old of desirability or appropriateness within some personally, organizationally, and socially constructed system of ethical norms, values, beliefs, and definitions.”(: 369)
14 2.1.3URGENCY
The third attribute, urgency, adds a more dynamic focus to the model and is being referred to as the call for instantaneous attention of managers by stakeholders’ claim (Mitchell et al., 1997). Furthermore, the authors propose that urgency consists of two separate aspects which both should be accounted for in order for a matter to be seen as urgent: 1) How time-sensitive the claim is, so that to what extent delaying attention to the matter is seen as improper by the stake-holders 2) criticality; how critical the claim is from the stakestake-holders’ perspective. Until this paper, the concept of urgency was mostly left out of the literature regarding stakeholders, alt-hough some minor research was done by Wartick & Mahon (1994) who introduced ‘time’ to the theory of managing stakeholders and who argued in what timeframe issues or claims could become significant to firms.
Moreover, the concept of urgency was revisited after several demands emerging from the literature. First, although adding a dynamic aspect to the model, urgency on itself is not relevant for stakeholder identification but only for determining the salience of a stakeholder (Neville, Bell & Whitwell, 2011), which was underlined by the findings of Parent and Deephouse (2007) who found that groups which only possessed this attribute were not seen as salient. Moreover, it was argued that urgency would be influenced to a certain degree by the feasibility of the stakeholder’s claim that it will take place (Driscoll and Starik, 2004). Further-more, similar to the argument made at legitimacy, again only the claim should be taken into consideration and not the urgency of the stakeholder itself. Again, here it is argued that the urgency of the stakeholder falls into the domain of the power attribute since this ‘type’ of ur-gency is described as the readiness of a stakeholder to exert its power (Easley & Lenox, 2006).
2.2STAKEHOLDERS JOURNEY
15 Sharma (2005). This influencing of other stakeholders is being referred to as ‘stakeholder mul-tiplicity’ (Neville and Menguc, 2006) and describes that the interactions between the stakehold-ers can lead to their claims being complementary which results in a higher overall salience by adding an extra stakeholder and its legitimacy to the claim.
2.2.1. Consumers
16 Figure 2.1 typology of stakeholders
2.2.2.AWARENESS
When going more in-depth into how consumers perceive CSR-related activities, it is known that on average, the general awareness among consumers about the CSR-practices of MNEs is low due to the lack of information they have about these MNEs, where CSR awareness entails the degree of insight and information that consumers possess about CSR related to a specific situation or entity (Bhattacharya and Sen 2004; Carrigan and Attalla 2001). Wigley (2008)not only acknowledges these varying levels of CSR awareness among consumers, she also found that being exposed to information about a company’s CSR-activities increased the knowledge consumers possess and therefore their awareness increases as well.
17 (Vlachos, Tsamakos, Vrechopoulos, & Avramidis, 2009), and on the other hand do consumers also tend to have supportive attitudes if there is awareness of the CSR practices of a firm (Bhattacharya and Sen 2004). However, the current findings did not examine yet how aware-ness affects consumers in their role as a stakeholder and what factors are useful in creating awareness and influencing the salience of consumers as a stakeholder. This paper will argue that two other types of stakeholders, the government and NGOs, can influence consumers in such a way that they can gain possession or increased intensity of both attributes by increasing their awareness. This ‘journey’ that must be taken, will go through the attribute of legitimacy and hereafter continues to acquire urgency.
2.3IMPACT OF NGOS AND GOVERNMENT
The government, as well as NGOs, can both be classified as secondary stakeholders in general since they are not related to the core of the business of firms (Maon, Lindgreen & Swaen, 2009). Although exemptions exist, such as firms who sell products only to governmental entities or when its main client is an NGO and thereby making it a primary stakeholder, this paper will focus on the role of government and NGO’s where they operate as secondary stakeholders. 2.3.1.GOVERNMENT
18 policies and regulations (Fox, Ward & Howard, 2002; Zappalá, 2003). The motive for this rea-soning stems from the globalized context of current business. Governments’ position of domi-nance in terms of power decreased and shifted towards more power for firms because economic relationships now go across boundaries whereas the power of a government does not (Albareda, Lozano, Tencati, Midttun, & Perrini, 2008).
What even further impedes this change, is the fact that an issue can be perceived as violating in one country, but not in a different country (Votaw, 1972) thereby limiting the power of governments in that particular country. To illustrate, a small revelation of the empirical set-ting will be provided. The CSR-issue under investigation is about so-called ‘lootboxes’ in online games such as the popular soccer-game FIFA 19. In Belgium, the government found it constitutes gambling and therefore declared the feature illegal in order to protect the users and especially children among these users. Later on, due to the international attention for the issue, this issue was also investigated in the United Kingdom and the United States of America to see whether it matches the criteria of gambling. In both countries, it was decided that although there was a close resemblance, it did not fully constituted gambling and therefore was declared as legal. Multiple members of the UK’s parliament did state that the nature of the practice was questionable and promised to keep intensively monitoring the situation (www.digital-trends.com, 2017). Since no laws were broken and although a large share of the consumers was against this practice due to the unfairness and the ‘addictive money’, the government was lim-ited in their options to counteract. What I clarified here is that the power of a government is not always significant to a great extent even though there is a consensus among the population that a particular practice is not in line with their social values and beliefs. Therefore, it can be stated that in general, the government does possess the attribute of power, but to a smaller extent compared to consumers, for instance. This is in line with the Mitchell et al. (1997), who stated that attributes are a variable, and not stable.
In terms of legitimacy, it is known that actions are defined as legitimate if they match “desirability or appropriateness within some personally, organizationally, and socially
con-structed system of ethical norms, values, beliefs and definitions” (Neville, Bell and Whitwell,
19 legitimacy. As argued before, the legitimacy of the claim (instead of the stakeholder) is of higher importance, but this too holds in the case of government. When dealing with CSR-issues, governments will investigate whether the issue is against the laws in place and if it is, govern-ments will proceed to resolve the issue meaning that the claim is most likely to be legitimate.
The attribute of urgency might be somewhat different, there it is seen as bringing dy-namics to the model and moreover, the attribute is dependent on especially legitimacy in the eyes of managers. For governments, the criticality of an issue higher when the issue is directly against a particular law or regulation and lower when no rules are broken. Moreover, relating to the voluntary nature of CSR (Jeremy and Moon,2008) and that parts of CSR are to a large extent more perception-dependent, making it harder to react upon, the urgency might not always be significant for governments. This can also be ascribed to the varying levels of how CSR is perceived in different countries, as exemplified by Dobers and Halme (2009) who state that the results of CSR research in western countries may be inapplicable in other cultural domains. To conclude, this paper argues that government can be salient to a high degree on the three attrib-utes, however, that it depends in no small extent on the context of the situation in order to state how salient the government is and therefore it should be assessed per situation and not in a general manner.
20 is expected that their authoritarian role should have an urgent perception effect on consumers (Neville and Mungec, 2006). Furthermore, it is expected the central and authoritarian role the government plays in the daily lives of people will strengthen these perceptions held by consum-ers. All arguments combined, including the theoretical and empirical argument mentioned be-fore that stakeholders are able to influence other stakeholders, this paper expects that the gov-ernment will be an effective ‘bringer’ of CSR awareness. Therefore, the following hypotheses are proposed:
Hypothesis 1a: The government can have a positive and significant influence on the legitimacy
of consumers through CSR awareness.
Hypothesis 1b: The government can have a positive and significant influence on the urgency of
consumers through CSR awareness.
2.3.2.NGO
21 strong allies to aid them such as the consumers and governments (Henriques and Sharma, 2005) In essence, it comes down to causing consumer pressure as well as societal pressure to influence firms (Spar and Mure, 2003). To conclude, this paper states that their power is not as significant as compared to the government and consumers.
For legitimacy, the situation is straightforward. The source comes from the fact that NGOs stand for “commitment to work for the public interest, performance, accountability to different constituencies and compliance with legal norms are among them” (Arenas, Lozano & Albareda,2009: 179). The core of their activities and goals lies at resolving social issues such as protecting the natural environment or fighting poverty. These types of activities are respected throughout most cultures and societies, Poret (2004) illustrates this by citing that “GlobeScan “Trust in Institutions” surveys” revealed that among all countries that participated NGOs were seen as most trustworthy of all entities within society. Therefore, it can be concluded that NGOs possess the attribute of legitimacy. Again, the focus should be on the legitimacy of the issue and less on the NGO itself.
For urgency, the attribute that brings dynamics, it is again dependent on the situation. The criticality of the issue is most significant for NGOs for the reason they act based on their core goal of solving social issues. Depending per issue, the degree of criticality differs logically. One might assume that NGOs are not able to respond to all issues that are at hand, for the fact that in some cases, there might be too many issues. They will select the issues to act on based on the potential damage and direct threat (Arenas, Lozano & Albareda, 2009), meaning that they will select cases that are highly critical to them. In terms of time-sensitivity, again it differs per situation and issue and is heavily dependent upon multiple factors such as nature of the issue, immediate threat, how many and what kind of victims, etc.
22 information provided by the NGO as accurate and justified. Whereas for urgency, it is expected that significant influences will be found based on NGOs relatedness with consumers pressure. Furthermore, NGOs in general focus on the most critical issues that are happening in their do-main and it can be expected that this translates in an urgent perception effect to consumers since consumers are mostly exposed to the largest problems in society. All arguments combined, including the theoretical and empirical argument mentioned before that stakeholders are able to influence other stakeholders, this paper expects that NGOs will be an effective ‘bringer’ of CSR awareness.
Therefore, the following hypotheses will be proposed:
Hypothesis 2a: NGOs can have a positive and significant influence on the legitimacy of
con-sumers through CSR awareness
Hypothesis 2b: NGOs can have a positive and significant influence on the urgency of
consum-ers through CSR awareness
2.4MULTIPLICATION
As argued by Rowley (1997), MNEs do not necessarily react to every stakeholder separately but to the interplay of their different stakeholders within their environment. Furthermore, the article argues that managers will notice the complementarity of stakeholders, which could po-tentially enhance the salience that stakeholders possess. Neville and Mungec (2006) theorized that through strategic interaction and a match in internal coherence, a synergistic relationship could be created between stakeholders. This relationship could hold that stakeholders can in-fluence the strength of each other’s claim. For this case, it is expected that, since it is hypothe-sized that both entities will have a positive effect on consumers, the government and NGO will complement and reinforce their claims, resulting in a stronger positive effect. Furthermore, it can be reasoned that being exposed to both stakeholders leads to increased awareness of con-sumers, which can result in being more susceptible to CSR-practices and issues, thereby poten-tially strengthening the effect on consumers’ salience.
Therefore, the following hypotheses will be proposed:
Hypothesis 3a: The multiplication effects of government and NGOs will have a more significant
positive effect on the legitimacy of consumers than a stand-alone effect
Hypothesis 3b: The multiplication effects of government and NGOs will have a more significant
23
2.5CONCEPTUAL MODEL
As a result of the proposed hypotheses, the following conceptual framework is developed.
24 3. METHODOLOGY
3.1CONTEXT OF THE RESEARCH
The context of the research will be based on a recent case in Belgium about videogames where multiple game-publishers and the Belgian government were involved.
3.1.1BELGIUM
I will now explain the case of FIFA, the most popular soccer-game worldwide, by examining recent developments in Belgium. To gain more in-depth insight into this case, I send a direct mail to the gambling commission which can be found in Appendix B. In recent years game-publishers featured a new addition in their games: micro transactions. This feature holds that players can buy in-game rewards with real money instead of the traditional system of unlocking rewards by playing. For game-publishers, the reason is clear: earning additional money on top of the earnings they receive from selling the game itself. To increase the sales from these micro-transactions, they offer these rewards in the form of boxes that have randomized content in it without any indication of how likely it is that you will receive anything useful, e.g. lootboxes. While FIFA already added the lootboxes into their games in 2012, they did not encounter any significant backlash besides from minor criticism from a small part of the gaming community due to the unfair nature of these practices. However, due to the media being informed of those minor complaints, the whole issue became infamous in just a relatively short amount of time since questions were raised about whether or not these lootboxes constituted illegal gambling. Resulting from this event, the Belgian gambling commission was ordered by the Belgian gov-ernment to lead an investigation and publish a report about the practices that were at play among different popular games.
25 Britain and US, similar cases were made, but here, lootboxes did not meet the criteria to be officially seen as gambling. Nonetheless, the gaming community and other people still are heav-ily complaining about it and are increasingly demanding that action needs to be undertaken. Multiple publishers already decided to skip (at least some of) these practices due to the inter-national pressure put on these firms. These successes can be interpreted that managers of these firms started listening to their consumers because their salience increased to certain levels mak-ing it almost impossible not to comply with their demands.
3.2.2CURRENT CASE
This paper will focus on creating a similar case like the one Belgium aimed at capturing the ‘activation’ caused by the government. Moreover, since NGOs are playing a growing role in today’s environment (Poret, 2014), this paper will also see whether NGOs possess the ability to increase consumers salience. Through the case in Belgium, it can be reasoned that after the publications of the report that the legitimacy of consumers ‘increased’ since people now be-lieved that something illegitimate was happening. Hereafter, it can be reasoned that the urgency increased as well. Pressure by communities, parents, and others increased after the topic became more widely known and people demanded change, which in some cases already lead to results. The setting of this paper will have four different situations. The aim is to have four independent samples to assess the influences of the government and NGOs on the perception of the consum-ers.
1. Perception of the situation without the involvement of Government or NGO (Control-group)
2. Perception of the situation with the involvement of Government (Government-group) 3. Perception of the situation with the involvement of NGO (NGO-group)
4. Perception of the situation with the involvement of both the government and NGO(Both-group)
3.2DATA COLLECTION
26 in English. Moreover, due to the dominance of the English language in the international arena, it is appropriate to use (Noak and Gamio, 2015). After pre-testing the survey, and since it was expected that the largest share of the respondents would be Dutch, a Dutch version was also created through careful translation. Furthermore, the research attempts to capture whether there is a significant increase in one or both attributes of urgency and legitimacy, and not a precise influence. Capturing an accurate number is not possible in this context as explained in the lit-erature review due to the perception dependent nature of the attributes. Therefore, in line with the reasoning of Blumberg, Cooper and Schindler (2013), a non-probability sample will be uti-lized in this study. Due to time constraints and the small number of characteristics for the sam-ple group, a convenience samsam-ple will be used that provides the researcher the possibilities to include all persons who are available (Blumberg, Cooper, and Schindler, 2013). Where the ad-vantages of this type of sampling lie in the ease and low costs, disadad-vantages arise around the reliability and accuracy of the research (Blumberg, Cooper, and Schindler, 2013).
3.3SAMPLE TARGET GROUP
27
3.4DEPENDENT VARIABLES
Although the concept of stakeholder salience received multiple theoretical contributions when first developed by Mitchell et al. in 1997, empirical consistency in measurement methods is not there yet. The authors argue that the strengths of the model to IB lay in developing an even more “accurate and fine-grained theoretical understanding” on how MNEs and their managers should analyze stakeholders. The statement above is rather accurately explained by Neville et al. (2011) who state that “some may suggest that the development of a scientifically precise tool for the measurement of stakeholder salience is neither possible nor desirable” (Neville et al.,2011: 14), explaining the empirical inconsistency. Most of the studies conducted surveys on managers to measure the salience of stakeholder’s perception (Agle et al., 1999; Harvey and Schaefer, 2001), while others investigated salience by looking at the probability that a firm will actually respond to a particular issue of a stakeholder (Easley and Lenox, 2006). Furthermore, due to the perception-dependent nature of the attributes, it is near impossible to determine the exact value of an attribute. It is only possible to state the attribute is significant to a high or low degree and that a certain stakeholder is more salient on some or all attributes than others. The most common methods used are surveys (Agle et al., 1999) and interviews (Parent and Deephouse, 2007).
3.4.1LEGITIMACY
While initially legitimacy is determined from the perspective of the managers, this study will follow a different approach. There is a theoretical consideration for the attribute of legitimacy as why this new approach will add to the research, instead of hindering the research. This paper follows the logic from Easley and Lenox (2006), stating that this attribute is best assessed through the perceptions of society in general. They argue that the general public determines if something is in line with the norms and values of society (legitimate), and not the managers of a firm. Furthermore, I argued that the context of the situation plays a significant role in deter-mining the salience. The issue under investigation is widely known among multiple countries. It is one of the most popular games ever made, and the case of Belgium received massive at-tention in countries such as the UK and the USA where it too created a nation-wide discussion. Therefore, I claim that in this specific situation, the public opinion is suitable to use as a deter-minant of legitimacy. In line with the literature review, the legitimacy aspect will focus on the legitimacy of the issue.
28 questions were constructed in order to capture the legitimacy of the claim which can be found in table 3.1. After recoding the third item to ensure its alignment with the other items, all four items were aggregated into the variable of legitimacy. A higher score on the legitimacy meas-urement here means that the situation is more legitimate in the eyes of society, whereby a lower score means that the situation is not in line with what is seen as legitimate in society.
3.4.2URGENCY
As emphasized before, urgency is the attribute that brings a dynamic element to the model. It can be measured through two aspects: 1) Criticality 2) Time-sensitiveness. Like legitimacy, there is no consistent tool available in the literature to measure this attribute. However, this attribute is rather straightforward and therefore, more suitable to capture using survey questions. For criticality, mostly it is necessary to measure how important the claim is to the consumer as it is argued that the higher the importance, the more pressure a stakeholder will exercise (Mitch-ell et al., 1997). For time-sensitiveness, it is required to measure how inappropriate stakeholders think it is if managers delay their attention to the particular issue. It is found that time pressure makes the managers of an MNE more susceptible to act upon a claim (Dror, Busemeyer & Basola, 1999). Based on the definition provided and on surveys of researchers such as Agle et al. (1999) and on both the criticality and time-sensitiveness of the claim, multiple questions were constructed to capture all aspects which can be found in table 3.1. After recoding the fifth item to ensure its alignment with the other items, all four items were aggregated into the variable of urgency. A higher score means that a situation is perceived as being more urgent to the con-sumer, and a lower score means it is less urgent.
29 Table 3. 1 Measure and item specification
Constructs Items Shown to
Legitimacy All groups
- is appropriate and in line with my personal norms and values - is appropriate and in line with norms and values of society - has a negative (in)direct impact on the users of FIFA - is perfectly fine and in line with what I believe to be ethical
Urgency All groups
- I personally find it important that something should be done about this situation - For society in general, it is important that this issue and similar other issues are solved - It should be a priority of EA to solve this issue
- Managers of EA should direct their attention immediately to this issue - I believe that any delay in solving this issue is inappropriate
30
Government-perception trust The government of my country…
Government-group Both-group
- represents the norms and values of society in general - is trustworthy
- is here to help and protect its citizens
- is responsible to ensure that business is done in an ethical manner
NGO-perception of trust NGOs in general.. NGO-group
Both-group
- play an important role in addressing and solving social issues that would otherwise receive no attention and support
- make the world a better place to live in
31
3.5INDEPENDENT VARIABLES
3.5.1CSR AWARENESS
32
3.6CONTROL VARIABLES
3.6.1PERSONAL RELEVANCE
It is expected that people who are directly affected by certain issues have different reactions towards it compared to people who are not directly affected by a certain issue. Since the issue is only ‘visible’ for two years, I will ask if people have played FIFA Ultimate Team in recent years. Especially for the criticality-aspect of urgency, this control variable is relevant. One the one hand, it could be argued that people who are familiar with the games are more biased to-wards the product and do not see the negative consequences. On the other hand, it could also be that consumers who are familiar with the product have more information to base their per-ception on and thereby increasing the accuracy of these perper-ceptions. Either way, in the results section, the influence of personal relevance will be assessed. The questions can be found in Appendix A.
3.6.2TRUST
Furthermore, this paper argues that there is a possibility that the influence of the government and NGOs on a consumer’s perception is dependent on how this consumer perceives the gov-ernment and NGO in general. In other words, the level of trust consumers have in both entities. During the results section, it will be assessed whether the level of trust influences the results significantly. To capture the level of trust, multiple questions about the perception by individ-uals about what they believe is the role of these entities in CSR were created, which can be found in table 3.1.
3.6.3DEMOGRAPHICS
Previous research regarding consumer behavior proposed that individual characteristics (age, education, and gender) could influence the results (Kim and Kim, 2004).
Age. This variable age is an ordinal variable. This variable will equal 1 when the participant is between 18 and 24 years old and equal 2 when the participant is between 25 and 30 years old. This variable will equal 3 when the participant is between 31 and 40 years old and equal 4 when the participant is above 64 years old.
33 Education. For education, similar to age, an ordinal variable was created. This variable will equal 1 when the participant has completed secondary vocational education, equal to 2 when the participant has completed higher professional education and equal 3 when the participant has completed academic higher education. This variable equals 4 when the participant has com-pleted some other form of education.
3.7SOCIAL BIAS
Research that investigates consumer intentions is susceptible to social desirability bias (Green-wald, Carnot, Beach and Young, 1987). Social desirability is explained as ‘the need for social approval and acceptance and the belief that it can be attained by means of culturally acceptable and appropriate behaviors’ (Crowne and Marlowe, 1964: 109). The bias indicates that partici-pants are prone to have a more positive tendency in their answering that results in a more con-venient self-image, disregarding their real perceptions toward a certain situation. Typically, in research related to CSR and ethics, the social desirability bias occurs rather often due to con-sumers wanting to illustrate that they care about social issues such as the environment (Morrell and Jayawardhena (2010). In order to overcome this bias, it will be stressed at the beginning of the survey that all responses will be anonymous.
3.8QUESTIONNAIRE DESIGN
34 questions capturing legitimacy and urgency using a 7-Likert scale followed by the question regarding how they perceive the entity involved. As mentioned at the control variables, people were then asked whether they have played the game in recent years. If the answer was no, they were automatically sent to the end of the survey and were thanked for their participation. If the answer was yes, follow-up questions regarding the intensity and whether they spend money on the game were asked. Hereafter, for these participants, the survey ended as well.
Six people were asked to participate in a pre-test of the survey. Since all questions were constructed and adapted, a pre-test was necessary in order to ensure reliability and validity and to guarantee that the questions operated as intended (Hilton, 2017). Especially in formulating the questions, the feedback was helpful to find out whether people interpreted the question in the same and correct manner. Multiple adjustments were made in formulating the questions, both for the English and Dutch survey. Moreover, it was carefully tested whether the description of the situation was clear and how participants would perceive it. Again, adjustments were made to ensure that there was a more neutral description so people could form their own perceptions. The statements by the government and the NGO were, after minor adjustments, perceived as clear and sufficient.
35 4. RESULTS
4.1PRELIMINARY DATA ANALYSES
The research design of this study was outlined in the previous section; this paper will now advance by analysing the collected data and testing the hypotheses. The online survey was available for one week before it was closed. This resulted in a data set of N=233 respondents. Before analysing the data, a screening was needed to ensure that missing data, incomplete re-sponses, and other ‘odd’ items were removed. N=174 respondents completed the survey, from the incomplete responses N= 4 respondents completed the survey by 99%, meaning that they did not click continue on the final page. Logically, these were included as well, but the other incomplete responses were removed from the dataset. Furthermore, N=7 respondents were re-moved from the dataset due to completing the survey in less than 100 seconds. N=2 respondents checked all the same boxes for each answer, which therefore can be seen as a sign of not paying attention to the questions and the information at hand. This leaves us with a database of N=169 respondents. Furthermore, the third item of legitimacy and the sixth item were recoded in the opposite direction to ensure that their scores aligned with the other items.
4.1.1SAMPLE DESCRIPTION
36 Table 4. 1 Sample characteristics
Characteristics N=169 % Age 18-24 years 74 43,8 24-30 years 42 24,9 31-40 years 14 8,3 41+ years 39 23,1 Gender Male 111 66 Female 58 34 Nationality Dutch 154 91 Other 15 9 Highest educational level Secondary vocational 9 6 Higher professional 56 33 Academic higher 96 57 Other 7 4 Missing Familiarity with the
situation
Yes 45 27
No 124 73
4.1.2DESCRIPTIVE STATISTICS
37 Table 4. 2 Descriptives
N Mean Std. Deviation Skewness Kurtosis
Legitimacy control 47 3,6170 1,27673 ,640 -,056 government 44 2,8920 1,10811 ,344 -,973 both 37 3,1081 1,22125 ,227 -,771 ngo 41 3,2317 1,19162 ,598 ,240 Total 169 3,2234 1,22202 ,440 -,355 Urgency control 47 4,0993 1,41833 -,160 -1,134 government 44 5,0530 1,26510 -,902 ,593 both 37 4,9054 1,11079 -,590 ,007 ngo 41 4,5244 1,21710 ,124 -,400 Total 169 4,6272 1,31249 ,417 ,539
4.1.3RELIABILITY -CRONBACH’S ALPHA
38 Table 4. 3 Cronbach’s alpha for all constructs
Construct Number of items Cronbach’s Alpha
Control Legitimacy 4 .864 Control Urgency 6 .895 Government Legitimacy 4 .711 Government Urgency 6 .890 Both Legitimacy 4 .776 Both Urgency 6 .820 NGO legitimacy 4 .756 NGO Urgency 6 .868 Government perception 4 .770
Both- government perception 4 .814
Both-NGO perception 3 .817
NGO_perception 3 .880
4.2HYPOTHESIS TESTING
4.2.1CHECKING SUITABILITY ONE-WAY ANOVA
39 In order to check whether the gathered data is appropriate for this test, six assumptions should be met (www.statistics.laerd.com).
Assumption 1: The dependent variable should be measured in the interval or ratio level.
Legit-imacy and urgency are the dependent variables in this case, and are measured through a 7-point Likert scale and therefore pass this first assumption.
Assumption 2: The independent variable should consist of at least two categorical groups. In
this case, CSR awareness consist of four categorical groups (1. Control 2. Government 3. Both 4. NGO).
Assumption 3: Independence of observations; meaning that people cannot participate in more
than one group and that there is no relationship between the groups. All participants were ran-domly assigned to one of the four situations and were not aware of the different situations.
Assumption 4: Test for outliers. No significant outliers were detected as can be seen in
Appen-dix C.1.
Assumption 5: The data should be normally distributed for each category of the independent
variable. As can be seen in Appendix C.2, not all data is normally distributed for each category. Unfortunately, this will be a limitation of this study, although the magnitude of this ‘error’ will not be that significant. The test requires approximately normal distribution, and therefore, small violations are accepted. Even further, it is tested that when homogeneity of variances is met, departures from a normal distribution do not interfere with the results (Blanca, Alarcon, Arnau, Bono & Bendayan, 2017). However, this assumption mainly holds when the values for kurtosis and skewness range between -1 and +1. In table 4.2, it can be found that the skewness and kurtosis for all categories all match this criterion.
Assumption 6: Homogeneity of variances. As can be found in Appendix C.3, using a Levene’s
test, it is shown that there is homogeneity of variances.
40 4.2.2ONE-WAY ANOVA
First, the control group (1), Government (2), Both (3) and NGO (4) were all coded into catego-ries in order to fit into the factor-variable in the analysis. As the dependent variables, urgency and legitimacy were listed. Below are the results of the test.
Table 4. 4 results of one-way ANOVA
ANOVA
Sum of Squares df Mean Square F Sig.
Legitimacy Between Groups 12,608 3 4,203 2,910 ,036
Within Groups 238,272 165 1,444
Total 250,880 168
Urgency Between Groups 24,374 3 8,125 5,058 ,002
Within Groups 265,030 165 1,606
Total 289,404 168
For legitimacy, there is a statistically significant difference between groups as determined by one-way ANOVA (F(3,165)= 2.910, p = .036). As for urgency, here too is a statistically signif-icant difference between groups as determined by one-way ANOVA (F(3,165) = 5.058, p = .002). It is now known that for both attributes, there is a statistically significant difference be-tween at least two groups.
41 Table 4. 5 Multiple comparisons
Tukey HSD Dependent
Variable (I) category (J) category
Mean Difference (I-J) Sig.
Legitimacy control government ,72498* ,023
both ,50891 ,221 ngo ,38531 ,440 government control -,72498* ,023 both -,21606 ,852 ngo -,33966 ,563 both control -,50891 ,221 government ,21606 ,852 ngo -,12360 ,969 ngo control -,38531 ,440 government ,33966 ,563 both ,12360 ,969
Urgency control government -,95374* ,002
both -,80611* ,022 ngo -,42510 ,399 government control ,95374* ,002 both ,14762 ,954 ngo ,52864 ,223 both control ,80611* ,022 government -,14762 ,954 ngo ,38102 ,548 ngo control ,42510 ,399 government -,52864 ,223 both -,38102 ,548
*. The mean difference is significant at the 0.05 level.
Government
42 can be found in table 4.2 and Appendix D, the score for legitimacy is statistically significantly lower (2.89 ± 1.11, p = 0.023) for the government when compared with the control group (3.62 ± 1.28), thereby supporting Hypothesis 1a. For urgency, a higher score means that a situation is perceived as being more urgent to the consumer, and a lower score means it is less urgent. As can be found in table/Appendix, the score for urgency is statistically significantly higher (5.05 ± 1.17, p = .002) when compared to the control-group (4.01 ± 1.42). This result provides support for Hypothesis 1b.
NGO
For NGOs, similar to the government, the hypotheses predicted that NGOs can influence both the attributes of legitimacy and urgency of consumers through CSR awareness. As can be seen in Appendix D, for legitimacy, there is no statistically significant difference between the NGO-group (3.23 ± 1.29, p = .440) and the control-NGO-group (3.62 ± 1.28). For urgency, no statistically significant difference is found between NGO-group (4.52 ± 1.22, p = .399) and the control-group (4.01 ± 1.42). Therefore, no support is found for hypothesis 2a and hypothesis 2b.
Multiplication effect
For hypotheses 3a and 3b, it must be tested whether the effect of the government and NGO combined is stronger than the stand-alone effects of either the government or NGO. Therefore, the both-group must be compared with the government-group and the NGO-group. As can be found in Appendix D, values for legitimacy are provided (Mean= 3.11 and SD= 1.22). There are no statistically significant differences found for the both-group when compared to the gov-ernment-group (2.89 ± 1.11, p= .221) and compared to the NGO-group (3.23 ± 1.29, p= .969). For urgency, the values are Mean=4.91 and SD= 1.11. Here too, no statistically significant differences are found for the both-group when compared to the government-group (5.05 ± 1.17, p= .954) and the NGO-group (4.52 ± 1.22, p= .223). Therefore, no support is found for hypoth-eses 3a and 3b.
4.3SENSITIVITY AND ROBUSTNESS CHECK
43 4.3.1PERSONAL RELEVANCE
The first variable that will be examined is whether or not people have experience with the situ-ation, e.g., personal relevance. From table 4.6, it can be found that of N=169 respondents, N=45 respondents were familiar with the game, and N=124 people were not familiar. For the ‘yes’-group, the scores for legitimacy (M = 3.64 SD = 1.02) and urgency (M = 4.28 SD = 1.29) seem to differ from the scores of the ‘no’-group for legitimacy (M = 3.07 SD = 1.30) and urgency (M = 4.75 SD = 1.30). To examine whether there is a statistically significant difference between both groups, an independent samples T-test was conducted. The results are shown in Appendix E. Firstly; it must be determined whether there the assumption of equal variances holds. Levene’s test for equality of variances determined that for both legitimacy (F(2.68), p= .103) and urgency (F(0.47) , p= .828) the assumption of equal variances holds. Following on this finding, it can be seen that for legitimacy there is a statistically significant difference between the ‘yes’-group and the ‘no’-group (t(167)= 2.751, p= .007). For urgency as well, there is a statistically significant difference between both groups (t(167)= -2.083, p= .039).
Table 4. 6 Group Statistics personal relevance Personal relevance N Mean Legitimacy yes 45 3,6444 no 124 3,0746 Urgency yes 45 4,2815 no 124 4,7527
44 there is a statistically significant difference between at least two groups of people with no per-sonal relevance.
As can be found in Appendix E.1, exactly the same hypotheses are confirmed as compared to the first analysis. For legitimacy, there is only a statistically significant difference between the government (2.71 ± 1.06, p= 0.021) when compared with the control-group (3.62 ± 1.43). For urgency, there are statistically significant differences for the government-group (5.15 ± 1.19, p= .003) and the both-group (5.25 ± .89, p= .003) compared to the control-group ( 4.07 ± 1.38).
4.3.2AGE
Furthermore, the influence of age on the results was analysed as well. Initially, five categories were created for age. However, ‘below 18’ was not taken into account for this research due to ethical reasons and only functioned as a ‘safe ‘button. However, not one participant was below 18 of age. To see whether the results differ per age-group, I will first combine the four remain-ing categories into two larger ones, namely ‘young’ and ‘older.’ Categories ’18-24’and ’25-30’ will be forming the ‘young’- group (N=116) based on their legitimacy and urgency scores, which can be found in Table 4.7. Furthermore, in Appendix E.2 can be seen that they are heavy correlated on legitimacy(p=1.00) and quite heavy on urgency (p=.605). The categories ’31-41’ and ’41 or older’ will form the new category ‘older’(N=53) and will also be based on their legitimacy and urgency scores found in table 2132. Again, the correlations of legitimacy (p=.986) and urgency(p=.958) for these categories are established as well.
45 Table 4. 7 new categories age
Legitimacy Urgency
Age Mean SD Mean SD
18-24 3.43 1.19 4.35 1.33 25-30 3.42 1.20 4.66 1.26 31-41 2.70 1.14 5.13 1.07 41 or older 2.83 1.36 4.93 1.33 New categories Young 3.42 1.19 4.47 1.3 Older 2.79 1.29 4.98 1.27 4.3.3TRUST Government
At three of the four groups, the survey asked how they perceived the government, NGO or both in order to see whether the trust people have in these entities has an influence on their outcomes. I will create for both government and NGOs two categories; people who have a lot of trust and people who have less trust in the entity. Scores can range from 1-7, where 3.5 is the exact middle. Ideally, I would create categories based on below and above the average of 3,5. How-ever, as can be seen in Appendix E.3 for the government, this would leave us with too few observations(N=11) to make an actual meaningful analysis. Therefore, the mean(M=5.00) of the group will divide the respondents into the categories ‘low’ and ‘high’. A total of N=81(Low: N=36 and High: N=45) respondents were asked to indicate their beliefs toward the government, which can be found in the Appendix. An independent samples T-test was conducted to see whether there are statistically significant differences between the groups. Descriptive results can be found in table 4.8.
46 Table 4. 8 Trust groups
Trust Legitimacy Urgency
Partici-pants Mean SD Mean SD N= Government 81 Low 3.16 1.11 4.80 1.20 36 High 2.78 1.18 5.21 1.17 45 NGO 78 Low 3.48 1.16 4.42 1.31 31 High 2.97 1.19 4.89 1.05 47 NGOs
47 5. DISCUSSION
The objective of this research was to investigate if the government and NGOs could ‘activate’ consumers’ through CSR awareness. As argued in the literature review, it was expected that the attributes of legitimacy and urgency were the factors that could inhibit this activation. The main finding holds that it is possible for less salient stakeholders to activate influential stakeholders through CSR awareness.
5.1GOVERNMENT
As stated before, the changing international context that causes firms to operate increasingly across borders has resulted in increased pressure on the traditional dominant position of the government (Albareda, Lozano, Tencati, 2008). Although the position of the government, in general, may have changed in recent years, hypothesis 1a and 1b confirms that the government is still a force to be reckoned with in the international arena and moreover, is still widely re-spected and recognized by consumers throughout society which is in line with the view of Ne-ville and Menguc (2006). For their influence on legitimacy and urgency, as expected, consum-ers perceive the government as legitimate and thoughtful, and the result shows that the individ-ual perceptions of the consumers are significantly influenced by the government. Simply put, due to their authoritarian and central role it is reasonable to assume that ‘’when the government says something is bad, it probably is bad’ which makes the government a useful ‘tool’ for CSR-awareness where they increase the salience of consumers. Furthermore, as CSR in its full status is still growing and becoming more embedded in today’s global environment (Crane et al., 2008), the government is able to play a pivotal and meaningful role in this period that is char-acterized by the development of CSR.
5.2NGOS
48 more effort to achieve similar results as for instance, compared to the government. To illustrate, this study showed participants statements by either or both the government and NGOs and aimed to capture the individual perceptions hereafter. As can be found in table 4.2, the mean scores of legitimacy and urgency differ from the control group, which could imply that there is an effect, but simply not strong enough. This could imply that merely being present or providing a statement is not sufficient for NGOs, but that only more ‘heavier’ actions (Spar and Mure, 2003; Henriques and Sharma, 2005) can result in convincing consumers to respond to CSR. Whereas the government plays a central role in everyday life of people in general, NGOs do not enjoy this position in society. Therefore, NGOs could be perceived as less important and powerful, simply because not all consumers have an adequate and consistent experience with NGOs which could theoretically limit their capacity to alter the perception of consumers. Es-pecially the fact that for legitimacy no significant difference was found is disturbing since NGOs are seen as the ‘embodiment’ of fighting social issues. What could be a cause in this outcome is the fact that NGOs are not as central as governments are for instance. There are countless NGOs in every country, but there is only one government. It could very well be that NGOs do have significant influence, but only when certain consumers have affection towards the goal of the NGO. For example, someone who places a great deal of emphasis on the well-being of animals could be influenced more by the WWF than someone who has little affection with animals.
5.3MULTIPLICATION
49 and critical when selecting an ally to cooperate with, as a wrong match could decrease their effect and influence on the salience of the stakeholder they intend to reach.
Table 5. 1 Multiplication effect
Government Both NGO
Mean Mean Mean
Legitimacy 2.89 < 3.11 < 3.23
Urgency 5.05 < 4.90 < 4.52
5.4SENSITIVITY CHECK
50 the variables were found, meaning that although the absolute values are influenced by personal relevance, the relative scores remained unchanged. This could imply that the direct impact of the government and NGO is not really decreased by personal relevance of consumers, but only their ‘starting point’ on both attributes is determined by whether they have personal experience with the situation or not.
51 6. CONCLUSION
6.1THEORETICAL CONTRIBUTIONS
The aim of the present research was to explore the dynamics in the model of stakeholder sali-ence created by Mitchell et al. (1997) by investigating a recent event in the gaming-industry. The findings add to the literature by proving that there is value to approach the model from the stakeholder perspective, instead of the traditional, firm-oriented approach. To illustrate, the em-pirical findings indicated that there exists such a thing as ‘activation’ of the consumers, by showing that their salience increased through CSR awareness after the intervention of the gov-ernment. Previous research treated the attributes as one-dimensional as if stakeholders already by nature had the possession of certain attributes. This paper indicated that in practice, it is much more complicated and moreover, that the government as a stakeholder can increase the salience of consumers through their perception regarding a particular issue. Thereby, at the same time, this study illustrated the importance of defining apparent differences between holders (government vs. NGO) due to the different perceptions people hold about these stake-holders. Taken together, these results suggest that addition of perception and CSR awareness to the literature of stakeholder salience is worth the effort and provides useful insights into how this mostly theory-based model can be better applied to real business settings worldwide.
Furthermore, a new light has been shed on the position of the government and NGOs in the current international arena by adding relevant findings and insights to their global posi-tion. Another contribution is a fine-grained analysis and an updated review of the separate at-tributes. There was theoretical inconsistency in operationalizing the attributes, but hopefully, this review brings more uniformity to future studies.
6.2MANAGERIAL CONTRIBUTIONS
