THE CSR COMMUNICATION STRATEGY OF IBERDROLA ON FACEBOOK AND TWITTER: A CORPUS-BASED LINGUISTIC AND CONTENT ANALYSIS


Universiteit Antwerpen, Bélgica
Universidad de Navarra, España

Abstract

In our increasingly digitalized civilization where sharing information and interacting about it is ever more made available to the general public, research into corporate communication practices of Corporate Social Responsibility (CSR) content on social media is desperately needed. This study makes a contribution to fill this research gap by performing a corpus-based content and multimodal linguistic analysis of the way in which an IBEX 35 energy company, Iberdrola, communicates about its CSR policy on social media. The corpus consists of 438 posts on Twitter and 126 posts on Facebook.

The results allow to draw the following conclusions: 1) Iberdrola’s information strategy varies from social media channel, where on Twitter the focus is through the environment, sustainability, social investment, stakeholders and arts, and on Facebook the focus is directed at stakeholders, thus motivating them to engage with CSR-related information; 2) there is certainly an interactive strategy, with many elements that characterize digital discourse, such as structures that elicit (positive) emotions, informal language use (like emojis and use of 1st and 2nd person) and other multimodal elements like hashtags and tags, which are all efforts at increasing brand trust, engagement, and admiration; 3) there are some underlying patterns within the informative and the interactive strategy, being that the overall way in which Iberdrola wants to communicate its CSR information is through informal and (positive) emotional language, with an exception of negative emotional language when discussing environmental themes, most probably to avoid accusations of greenwashing.

LA ESTRATEGIA DE COMUNICACIÓN DE RSC DE IBERDROLA EN FACEBOOK Y TWITTER: UN ANÁLISIS LINGÜÍSTICO Y DE CONTENIDO BASADO EN CORPUS   

Resumen

En nuestra civilización digitalizada, en la que el compartir información e interactuar sobre ella está cada vez más a disposición de todos, se necesitan estudios sobre las prácticas de comunicación empresarial de la Responsabilidad Social Corporativa (RSC) en los medios sociales. El presente estudio de corpus contribuye a llenar este vacío mediante un análisis de contenido y lingüístico multimodal de la forma en que una empresa energética del IBEX 35, Iberdrola, comunica sobre su política de RSC en los medios sociales. El corpus consta de 438 posts en Twitter y 126 posts en Facebook.

Los resultados permiten sacar las siguientes conclusiones: 1) la estrategia informativa de Iberdrola varía según el canal: en Twitter el enfoque está en medio ambiente, sostenibilidad, inversión social, grupos de interés y artes, mientras que en Facebook se dirige principalmente a los grupos de interés, motivándolos a comprometerse con la RSC; 2) existe indudablemente una estrategia interactiva que caracteriza el discurso digital, como las estructuras que suscitan emociones (positivas), el uso de lenguaje informal (los emoticones y el uso de la primera y segunda persona) y otros elementos multimodales como los hashtags y tags, que todos sirven para aumentar la confianza y el compromiso por la marca; 3) hay patrones subyacentes en la estrategia informativa e interactiva, en el sentido de que Iberdrola quiere comunicar su política de RSC a través de un lenguaje informal y emocional (positivo), con la excepción del lenguaje emocional negativo cuando se discuten temas ambientales, muy probablemente para evitar acusaciones de ecoblanqueo.

CONTEÚDO BASEADO EM CORPUS

Resumo

Na nossa civilização digital, na que compartilhar informação e interagir sobre ela está cada vez mais à disposição de todos, é preciso estudar sobre as práticas de comunicação empresarial da Responsabilidade Social Corporativa (RSC) nos meios sociais. O presente estudo de corpus contribui a preencher este vazio através de uma análise de conteúdo linguístico multimodal da forma na que uma empresa de energia do IBEX 35, Iberdrola, comunica sobre sua política de RSC nos meios sociais. O corpus contém 438 posts no Twitter e 126 posts no Facebook.

Os resultados permitem tirar as seguintes conclusões: 1) a estratégia informativa de Iberdrola varia de acordo com o canal: no Twitter o foco está no meio ambiente, sustentabilidade, investimento social, grupos de interesse e artes, enquanto que no Facebook aborda principalmente os grupos de interesse, motivando-os a se comprometer com a RSC; 2) sem dúvida existe uma estratégia interativa que caracteriza o discurso digital, tais como as estruturas que despertam emoções (positivas), o uso de linguagem informal (os emoticons e o uso da primeira e segunda pessoa) e outros elementos multimodais como os hashtags e tags, que servem para aumentar a confiança e o comprometimento pela marca; 3) existem padrões subjacentes na estratégia informativa e interativa, no sentido de que Iberdrola quer comunicar sua política de RSC a través de uma linguagem informal e emocional (positiva), com a exceção de linguagem emocional negativa quando se discutem temas ambientais, muito provavelmente para evitar acusações de eco branqueamento.

Keywords

Corporate Social Responsibility (CSR), Digital corporate communication, Social media, Energy sector, Spain, Iberdrola, Content-based analysis, Multimodal linguistic analysis, Corpus analysis.

INTRODUCTION

Both CSR communication and communication on social media have become important pillars in an increasingly complex and conscious society. The demand by society and shareholders for good CSR communication is growing by the day, and the realisation that we have a commitment towards our planet is simply undeniable (Bonsón & Ratkai, 2013). However, recent research byPostma (2015) shows that there is often disagreement and uncertainty about the suitability of social media in communicating CSR and that there is a need for concrete advice and a clear framework.

In the case of CSR in Spain, there are certain guidelines that set out a CSR strategy, such as the Law 11/2018 1 , but they do not specify anything in the area of social media. For that purpose, this paper examines the Facebook and Twitter posts of Iberdrola, a Spanish multinational operating in the electricity sector. This multinational was awarded the Best Corporate Governance Spain 2019 Award in 2019, and analysing the firm’s communication strategies can lead to relevant insights that will allow companies to build a better framework for CSR on social media in the future. On top of that, Iberdrola occupies a prominent position in the management of its social networks according to a study carried out by Epsilon. Both Facebook and Twitter posts are analysed because these channels have a high, and similar, number of followers and CSR content is published on both channels.

To sum up, CSR nowadays is a high-profile theme for companies and from that perspective, this paper wishes to provide an answer to the question of how a Spanish-speaking company fulfils its CSR policy via social media, mainly in terms of content and interactive potential. More precisely, the main research question (RO) that will be addressed is:

RQ “How does a Spanish company fulfil its communicative strategy around CSR on social media?”

The sub-questions that will be addressed are:

RQ1 “Which CSR-related themes are (most often) used?”;

RQ2 “Is there a possible discursive pattern in the way the messages are formulated, more precisely regarding the use of emotional language, informal language and multimodal elements?”

RQ3 “Are certain themes combined with certain textual patterns?”

In what follows, we first comment more thoroughly on the current situation of CSR communication in Spain. Then, the theoretical framework aims to approach the relevant literature on CSR communication on social media. Section 3 discusses the corpus and research methodology, and Section 4 comments on the results. In the final section, the results are discussed and compared to the literature and certain conclusions are drawn.

BACKGROUND OF THE STUDY

CSR communication in Spain

Over the past two decades, there is a positive trend among Spanish companies when it comes to CSR reporting. Seven Spanish firms used the Global Reporting Initiative (GRI) system for the elaboration of annual CSR reports in 2002. This number rose to 180 by 2012. It is also important to note that Spain is the country with the highest number of GRI Guidelines being implemented (Ministerio de Empleo y Seguridad Social, 2014). In terms of public and legal initiatives, in 2018, a new commercial law (Law 11/2018 on non-financial information and diversity) was ratified, which specifies the issues on which companies must provide information, specifically in the areas of environment, employees, human rights, corruption, society, suppliers, consumers and tax information. It is thus safe to assume that good CSR communication in Spain is now required by law, although there are still some questions about companies fulfilling these requirements, more specifically with respect to reporting on social media, as no framework or guidelines on this matter are provided. This probably explains why Spanish companies in general don’t score high on CSR information (Observatorio de Responsabilidad Social Corporativa, 2018).

As for the quality of the Spanish CSR communication, the latest report by the CSR Observatory (Observatorio de Responsabilidad Social Corporativa) (2018), which was created in 2004 to study and promote CSR, concludes that the level of CSR information in Spain is still quite low, with 33.62 points on average on a scale of 100. For reasons of representativeness and relevance, they used as a sample the thirty-five companies that on 31 December 2018 made up the IBEX 35 index of the Spanish stock market, among which Iberdrola. Their aim was to evaluate the quality of the information on aspects of Corporate Social Responsibility (CSR) provided in the documentation and public and accessible reports of these companies, and thus they studied annual reports, sustainability reports or similar reports. They established that although all of the companies analysed make explicit mention in their sustainability reports and non-financial information statements of the problems and challenges related to each of the areas of study -fundamentally in relation to the environment, society and corruption-, the quality of the information they present with respect to each of these aspects is still deficient. In addition, it does not reach an adequate level in any of the areas analysed despite the obligations imposed by the already mentioned Law 11/2018, which was used as an indicator for the analysis. It should be mentioned though, that when compared with the score obtained in 2017, we see that the score rises by all areas, which means that the entry into force of Law 11/2018 has had an effect of improving the information presented in those aspects on which it deals. Furthermore, by sector of activity, the one that obtains the best score is that of Oil and Energy. Among these is Iberdrola, that, although it scores under the 50 mark -with 49.66 points on a scale of 100-, ranks as third out of the 35 companies. The reasons that the CSR Observatory (Observatorio de Responsabilidad Social Corporativa, 2018) gives as to why the Oil and Energy sector is better positioned, may be related to greater social pressure and therefore a greater degree of vigilance on the part of civil society and the media as companies carry out activities with high risks of environmental and human rights impact. On the other hand, this is a sector of activity that is subject to greater regulatory requirements in the environmental and tax fields, among others.

CSR communication on social media

Traditionally, a company communicates its CSR efforts through announcement of reports, advertisements or web sites (Du, Bhattacharya, & Sen, 2010). Now, thanks to Web 2.0, interactive online conversations are possible via a wide range of alternatives like blogs and microblogs, content communities, social networking sites and social bookmarking or aggregators (Coombs & Holladay, 2012). This highly successful method of marketing leads companies to rethink their methods for communicating CSR systems as well and incorporating social media into their current communication policy has become crucial for companies (Brown & Nwagbara, 2018).

Postma’s study (2015) offers qualitative evidence for businesses’ approaches to the use of social media when engaging on CSR. However, the organizations in her study do not agree on the suitability of social media to communicate about CSR related topics. This division is consistent with conflicting results in literature (Fieseler et al., 2010; ) (Eberle, Berens, & Li, 2013; Haigh, Brubaker, & Whiteside, 2013). Most companies advocate for a media mix, due to the instability of social media and the lack of participation of stakeholders -which can instead be enhanced through stakeholder dialogue-. Postma demonstrates the difficulty of applying theoretical expertise in action. The discrepancy between CSR theory and reality indicates that organizations need practical guidance about how theory should really be put into practice. A framework on the design, applications and communication of a CSR policy could provide a solution to bridge this gap (Postma, 2015).

Brown and Nwagbara (2018) also have described some of the benefits and challenges of CSR communication on social media. Firstly, they note that social media helps businesses listen to target audiences and track which particular issues related to CSR are relevant to different stakeholders. Secondly, businesses can increase their legitimacy of CSR behaviour, although there is an increasing belief in scepticism regarding CSR activities reported through media (Du et al., 2010). In addition to many other advantages, such as cheap investment, performance, real-time communication, public relations and online archive choice, social media also has several holdbacks for CSR communication. Unfortunately, failing to obey etiquette rules, such as imposing unrequested information (Coombs & Holladay, 2012) can easily infringe such a valuable outlet. Another issue is the fact that businesses have little to no influence over the CSR information that flows through social media, which can easily become negative and is impossible to predict (Friedman, 2006). Furthermore, the language of social media also needs to be carefully designed to reflect proper corporate identity. In this respect, the literature focuses on emotional language, informal language and multimodal elements.

Emotional language

In Web 2.0, emotional marketing finds an ideal platform for growth. The conventional corporate contact, which is characterized by unidirectionality and verticality, has been enriched, and a modern way of communicating has emerged, implicating a multidirectional communication in which all the stakeholders of the organization participate. The usage of networks as well, in particular Facebook, Twitter, Instagram and LinkedIn, has undergone enormous growth in the organizational climate (Han & Chang-Hoan, 2013).

So far, research that combines emotions and feelings on Twitter has focused primarily on 'marketing' and sociological issues, but not on linguistic issues. In recent years, some efforts have been made to find analytical mechanisms capable of examining what kinds of emotions are present in both tweets and other forms of network publications and how these emotions affect 'virality’ (Stieglitz, 2013). However, studies which examine the emotions on Twitter in the Spanish corporate context from a linguistic point of view are scarce, with exception of F Fernández-Vallejo (2018). She carried out a study in which the expression of emotions in Spanish corporate Twitter accounts (including Iberdrola) is researched to observe which linguistic structures characterize or elicit emotions, and which lexical systems conceptualize and value the semantic field of emotions. Her study confirmed the validity of corporate messages as a relevant corpus for the study of the expression of emotions in the organizational environment, given the high percentage of tweets that describe or provoke emotions. The research indicated that those that dominate in Acciona and Iberdrola tweets were the emotions of positive valence like joy, confidence and trust, which generates respect and loyalty in the shareholder or consumer. Clearly, the elicited emotions based on the company’s purpose, vision and values help to reinforce the identity of the organization.

Informal language

The Internet has undoubtedly revitalized the age-old debate on the characteristics of the oral and the characteristics of the written, and mentioned as well the distinctive graphology that it has, which is based on the singularity and lexical-graphic creativity of the messages on the Internet (Crystal & Tena, 2002). While social media have become major communication channels for customers and brands alike, marketers are struggling on these channels to establish sustainable consumer-brand ties. Recent research shows that attempts by advertisers to cultivate social media ties with their customers are far from being successful (Fournier & Avery, 2011), where consumers often reject brand ads and criticize brands in their social spaces ( ; Gretry, Horváth, Belei & van Riel, 2017) (Noort & Willemsen, 2011).

In these cases, nonverbal signals, such as contact style, play a critical role in building trust and, subsequently, determining whether the relationship will advance beyond the initial encounters (Keeling, Mcgoldrick, & Beatty, 2010). To research this, Gretry et al. (2017) carried out a study in which they examined how using an informal (vs. formal) communication style affects brand trust in a context of social media. The experiments show that the effects of an informal style on brand confidence depend on whether consumers are familiar with the brand, so that the use of an informal style enhances (decreases) trust in brands that consumers are familiar (unfamiliar) with. Notably, brands appear to use a predominantly informal style in their communications on social media (Beukeboom, Kerkhof, & Vries, 2015), as they believe it fosters closeness. An informal style of communication is defined as "common, unofficial, familiar, casual, and often colloquial, and contrasts with formal in those senses" (Mcarthur, 1992). For example, it is common for marketers to refer by their first name to their customers, and to use emoticons and/or abbreviated expressions. Freedom in the use of capital and small letters, the revitalization of special punctuation signs, or the sweetening of the academic rules of punctuation is observed as well (Crystal & Tena, 2002; ) (Fernández-Vallejo, 2020).

Multimodal elements

Multimodal elements are also an essential aspect of social media discourse, as they influence searchability. The hashtag, a type of social tagging that enables micro bloggers to inject metadata into social media posts, is a key semiotic tool supporting this feature. While popularly thought of as topic-markers, hashtags in social media texts are able to construe a range of complex meanings. Annotating social media posts with these tags is a fast-paced but evolving activity and a process of linguistic innovation. As laid out by Zappavigna (2015), hashtags can serve for multiple things. They can serve as subject markers -their most commonly recognized position-, but they can also construe interpersonal significances or enact metacommentary at semantic discourse level. In addition, they can be seen as a form of social tagging, because the annotation of text serves as a concept of “searching” the social network.

In addition to hashtags, tags or appellative hyperlinks can also be considered as a special type of strategy inherent to social media discourse, which consists in referencing the relevant audience profile to increase changes of a commentary (Delfa, 2017). Although there is not a lot of literature to be found on theories around the use of this for social media discourse, Choudhury, Sundaram, Lin, John, and Seligmann (2009), concluded that the use of appellative hyperlinks improves recommendation performance significantly against image content alone.

METHODOLOGY

Corpus selection

To answer the research questions set out in the introduction, the content and textual elements of a corpus of 438 Tweets and 126 Facebook posts publicised by the Spanish firm Iberdrola were examined (see table 1). Iberdrola was chosen as a firm because it is part of the IBEX 35, the benchmark stock market index of the Bolsa de Madrid, Spain's principal stock exchange. In this particular case, Iberdrola is one of the leaders of the Oil and Energy sector. In addition, it has a strong commitment to sustainability and CSR at an environmental and cultural level, which has led them to develop sponsorships of social initiatives as well. In fact, according to the 15th edition of the 'Global 100 Most Sustainable Corporations in the World', published annually by Corporate Knights, Iberdrola was named as one of the world's 100 most sustainable companies.

Another important element as to why Iberdrola was chosen, is the fact that it occupies a prominent position in the management of its social networks, standing out for fans, interactions, virality and engagement (according to a study carried out by Epsilon in 2016). In addition to that, they have a good follower amount, which is relevant as to the reasons why they are maintaining such loyalty from stakeholders on social media. The current number of followers on Twitter is 65.652 and on Facebook is 76.242. This is quite interesting, because when FFernández-Vallejo (2018) carried a study about Iberdrola’s online presence, the number of followers on Twitter was only 48.055.

The corpus covers a three-month period from October 1st, 2019 to December 31st, 2019. This period of time was marked by some important holidays, like Christmas and New Year’s Eve, and a really big event in Madrid, namely the COP25. The 2019 United Nations Climate Change Conference was the 25th United Nations Climate Change conference and it was held in Madrid from the 2nd of December until the 13th of December. There is no doubt that this has had a big impact on the social media strategy of Iberdrola. Other important observations were sponsorship of various events, implementation of infrastructure works related to sustainability and CSR initiatives, but these happen all year round.

To obtain the corpus, the programme Twlets [http://twdocs.com] was used for Twitter, which directly downloaded the tweets into an Excel sheet. For Facebook, the corpus was obtained manually and put into a Word document. A pretest was applied to 30 Tweets and 30 Facebook posts to check if the selected themes (which were selected through a Word Frequency Query) were applicable to the sample.

In short, the corpus addresses the small-scale empirical study, convinced with Thornbury (YEAR) that a small and homogeneous corpus can also be valid. The approach to the study of language is adopted from two complementary perspectives, the content and linguistic analysis.

Table 1: Number of Tweets and Facebook posts analysed.

Corpus

Number of posts

Words analysed

Average of posts per day

Twitter

438

12.759

4,76

Facebook

126

5.313

1,36

Total

564

18.072

Research techniques

The aim of this study is to examine the CSR communication of a Spanish-speaking company, in this case Iberdrola, on Facebook and Twitter. Twitter and Facebook are seen as two separate parts of the digital communication strategy, that is examined in terms of content and interactive potential.

The content analysis was structured as follows. Firstly, the deductive part took place, meaning the first part of the study went into researching previous literature about CSR and social media. It quickly became clear that although there are many frameworks for CSR communication in reports (like the GRI guidelines), there were not many frameworks around CSR communication on social media, nor was there clarity around the appropriateness for it. There is a lot of discrepancy between CSR theory and reality, and, as indicated in the previous section, organizations are unsure about the suitability of social media for CSR, often advocating for a media mix (Postma, 2015). There is also confusion about the impact of CSR communication on social media, with conflicting opinions (Cho, Furey, & Mohr, 2016; Etter, 2013). This is where the urge for the development of CSR communication strategies to encourage publics to engage with corporations comes in place, which demonstrates various aspects of CSR by including both internal and external publics involved in the programs (Cho et al., 2016).

Secondly, a Word Frequency Query was carried out in Nvivo, 2 which is intended to identify possible themes, particularly in the early stages of a project. The output of this inductive approach was combined with the deductive approach and then used to code the content of all posts. In order to check the reliability of the method, a second coder was also asked to evaluate 10% of the Tweets and Facebook posts (44 Tweets and 13 Facebook posts). The agreement percentages between the two coders who performed the task were over 89% for all content-specific themes, and even higher for the textual themes. In order to take into account the likelihood of the agreement between users occurring by chance, the Kappa coefficient was measured as well. The Kappa coefficient was always between 0,55 (k > 0) and 1,00 (k = 1) for Twitter, which is fair to very good agreement, and between 0,33 (k > 0) and 1,00 (k = 1) for Facebook.

As for the linguistic part of the analysis, all posts were coded based on the presence of emotional, informal and multimodal elements, as exposed in section 2. The next section presents the results of the content and the linguistic analysis.

RESULTS

Content analysis

The content of both the Tweets and the Facebook posts were analysed according to 8 themes, which can be found in the table below (see table 2. Because of the fact that posts often treated more than one theme, and were thus coded multiple times, a lot of themes have a high number of references coded. The results of the CSR content will be discussed according to frequency and/or occurrence of the themes.

Table 2: Results of content analysis.

Themes

Twitter

Facebook

Frequency of references coded

Coverage percentage

Frequency of references coded

Coverage percentage

Environmental theme

120

13.70%

22

13.89%

Sustainability

129

14.73%

14

9.16%

Interests and current affairs

148

16.88%

69

47.03%

Stakeholders

148

16.88%

44

31.49%

Projects

92

10.50%

29

19.25%

Cultural references

39

4.45%

15

12.82%

Innovation

21

2.40%

1

0.60%

Contests

53

6.05%

48

42.55%

For both Twitter and Facebook, “Interests and current affairs” have the highest percentage of coverage. This theme was coded when the Tweets or Facebook posts were about sports, specific days/holidays, events, and in a smaller quantity, animals. See the posts below for an example of holidays (1) and animals (2).

1. Primer ítem “#Halloween �� da mucho MIEDO por un día, pero destruir nuestro �� nos debe alarmar SIEMPRE �� ¡No convirtamos el planeta en un escenario de TERROR! #halloween2019 #LordPlastik” 3

2. “Cuidar a los animales en la naturaleza y en casa nos define como seres humanos. ¡Protejámoslos! #DíaMundialdelosAnimales” 4

What’s distinct is that for Twitter, the theme “Stakeholders” occupies the 1st place as well (148 references coded for both “Stakeholders” and “Interests and current affairs”). “Stakeholders” was coded according to subthemes customers, employees, women and families. An example of customers (3) and women (4) can be found below.

3. “En #Navidad �� y SIEMPRE, nos comprometemos a NO CORTAR la luz a nuestros clientes en situación vulnerable que no puedan pagar sus facturas, gracias a acuerdos con instituciones y ONG” 5

4. “¿A qué retos se enfrentan las mujeres en el nuevo entorno digital? Expertas del sector piden más igualdad de oportunidades en el I Encuentro de mujeres digitales #EmakumeDigitalak en #Bilbao ��‍������‍♀️” 6

For the second most concurrent theme, there is a noticeable difference between the communication on Twitter and Facebook. On Twitter, the second most common theme is “Sustainability”, with three subthemes; renewable energy, emission reduction, and products/services (because of the fact that Iberdrola’s business falls under the sector of sustainable energy). An example of renewable energy (5) is the following:

5. “Terminó la #COP25, pero el compromiso y ACCIÓN por el clima CONTINÚA ➡ Impulsamos las #renovables y la sostenibilidad por un �� mejor para TODOS #TiempoDeActuar” 7

This implies a big difference with the communication strategy on Facebook, where “Contests” (even though they are for motivating stakeholders and CSR-related) are the second main theme. Whereas contests occupy the second place on Facebook, they only occupy the 5th place on Twitter. The same goes for “Sustainability”, which only occupies the 6th place on Facebook. This means that both media channels attach different value towards these themes. An example of a Facebook posts with a contest (6) theme is the following:

6. “Frenar el cambio climático es imprescindible para dejar un �� habitable a los que vienen... ¡y a nosotros! Reivindícalo en Instagram y participa en nuestro concurso #PorelClima junto a @IgersSpain, con motivo de la #COP25 en Madrid.” 8

The third most frequent theme for Twitter is the “Environmental theme”, with whereas for Facebook it’s “Stakeholders”, and after that “Projects” (social, cultural, environmental, etc.). These are interchangeable for the next theme, with “Projects” occupying this space on Twitter and “Environmental themes” occupying this space on Facebook. This means that both the environmental theme as the social theme carry approximately the same importance on both social media outlets, with almost always more than 13% coverage percentage. The environmental theme was coded in line with three subthemes; environmental issues, respect/appreciation towards environment, environmental tips. An example of projects (7) and environmental issues (8) can be found below.

7. “Pippi Calzaslargas tenía a Pequeño Tío, su “gran”ayudante y amigo��... y en el mundo real, niños con discapacidad psíquica tienen a los �� de la Fundación @Caballo_Amigo, con la que colaboramos en ����‍��, para ayudarles a desarrollar sus habilidades e integración ��” 9

8. “NO, no es algo NATURAL. Ni obra de un agente secreto��‍. No lo trajeron los extraterrestres��. No llegó ahí por casualidad��‍♂‍. La culpa de la INVASIÓN del plástico es nuestra. Hagamos caso a Cousteau, el gran defensor del ��, y paremos YA a #LordPlastik ����‍��” 10

Themes that are present to a lesser extent, although they should be mentioned, are “Cultural references” (9) and “Innovation”. More determinant, “Innovation” cannot be considered as a part of the CSR communication for Facebook, whereas the theme was in fact present on Twitter. This could be because of the fact that Twitter communication is more CSR in its pure form, which is informing, and on Facebook it is almost always a combination of informing and interacting.

9. “El pavo, el marisco, el jamón, el pescado, la ensalada, el cava, el turrón, el ��...En #Navidad y SIEMPRE, necesitamos a la gente de campo y mar. Sin nuestros pueblos, no tendríamos NADA. ¡Porque no solo lucen las ciudades! Feliz #Navidad������ #PueblosQueBrillan ��������” 11

Linguistic analysis

The results of the linguistic discourse of both the Tweets and the Facebook posts will be discussed according to frequency / occurrence of the textual elements, as presented in the table below (see table 3).

Table 3: Results of linguistic analysis.

Themes

Twitter

Facebook

Frequency of references coded

Coverage percentage

Frequency of references coded

Coverage percentage

Emotions Sustainability

Negative valence

69

7.88%

18

12.06%

Positive valence

312

35.60%

100

72.11%

Informal textual elements

Abuse of punctuation

155

17.69%

80

57.27%

Emojis

388

44.28%

118

86.61%

Use of capital letters at liberty

248

28.30%

90

62.20%

Sound mimicry

1

0.11%

/

/

Use of 1st and 2nd person

308

35.15%

94

73.08%

Multimodal elements

Hashtags

329

37.56%

108

81.01%

Tags

114

13.01%

20

16.22%

Emotions

Both Twitter and Facebook have more references coded for positive valence (10), although Facebook has more percentual coverage, meaning the overall positive valence on Facebook is bigger -or more intense- than on Twitter. Positive valence was coded when posts implied a feeling of surprise, happiness, trust or pride. The same goes for negative valence (11), which is -although it is used less- more intense on Facebook. Negative valence was coded when the feeling of anger, sadness or fear was implied. This could mean that the overall communication strategy on Facebook focuses on more strong emotions, whether they are positive or negative. On Twitter, a small amount of posts is merely neutral.

10. “¿Y si sonreímos? Usemos bien el mejor recurso: ¡TU energía positiva, la de TODOS! Construir, amar, disfrutar... �� #DíaMundialdelasonrisa” 12

11. “��️��️��️��️��️ Cinco islas ENORMES que no figuran en los mapas... Grandes masas de plástico flotante que amenazan a nuestros océanos. ¡Frenemos YA el plástico de un solo uso! #TiempoDeActuar #COP25” 13

Informal textual elements

Emojis make up a big part of the (informal) communication for both Twitter and Facebook, with the highest percentual coverage for both social media outlets. The second most common informal element is the use of 1st and 2nd person, whether it’s the use of verbs, adjectives or personal pronouns. This counts for Twitter, as well as Facebook.

The use of capital letters at liberty occupies the third place for both Twitter and Facebook. This means that a lot of words are written in full capital letters when the emphasis is on that word. This is very unusual and informal coming from a big corporation.

Abuse of punctuation is present in a lesser extent, though it is not less important than the other informal elements. Even though less references have been coded, the coverage percentage is quite high, meaning they carry great importance when it comes to the social media communication. Sound mimicry however has little to no influence, with only one reference coded.

Multimodal elements

In both communication media, Twitter and Facebook, hashtags play an important role. With a high to very high coverage percentage, 37.56% for Twitter and 81.01% for Facebook, it is safe to assume that they are very recurrent. It is relevant to mention however, that the percentage for Facebook is much higher. As stated earlier, this can be due to the fact that the communicative strategy on Facebook is more intense, meaning a higher stakeholder interactivity is expected. Twitter, on the other hand, can be purely informative (for CSR) as well, and is not only focused on interactivity.

In terms of tags of other people or corporations, Facebook and Twitter have approximately the same amount of coverage, that is to say that there is no big difference in terms of CSR communication. The coverage percentage is high as well, which lies between 13% and 16% percent.

Integration of content and linguistic analysis

Below, the correlations between findings in the content analysis and findings in the linguistic analysis will be discussed, to reveal any patterns that might be present. The purpose of this section is to answer a range of questions that might come forward when patterns in data are shown.

For both Twitter and Facebook, the subject with the most positive valence is the theme “Stakeholders”, meaning positive valence is mostly used when posts are directed at stakeholders. For Twitter, negative valence is mostly used when informing about environmental themes and issues. For Facebook howeer, the theme with the highest percentage of negative valence is “Interests and current affairs”. Nonetheless, it is a very close call with “Environmental issues”, with only one reference less coded. Thus, it can be assumed that there is no big difference in the type of valence between Facebook and Twitter.

Another striking observation are the different types of informal textual elements that are linked to types of content, which also vary depending on the social media channel. On Twitter, the theme “Stakeholders”, for example, is mostly combined with “Emojis”, although it should be noted that “Emojis” have an overall high coverage percentage. “Interests and current affairs” are very often combined with “Abuse of punctuation”, “Free use of capital letters” and “Use of 1st and 2nd person”. This is probably due to the fact that these themes are to implicate stakeholders more than other themes and are thus very interacting in terms of strategy.

Regarding informal textual elements for Facebook however, all textual elements are predominantly used when communicating around “Interests and current affairs”. In addition, the second theme that all the textual elements are mostly combined with, is “Contests”. This is quite different from Twitter, where “Sustainability” and “Environmental themes” were frequently used as well. As mentioned previously, this could be because more stakeholder interactivity on Facebook is expected, and the focus is therefore on other themes.

In terms of multimodal elements, for both Twitter and Facebook hashtags are most often used when communicating about “Interests and current affairs”, probably due to the fact that Iberdrola often uses daily events as an advantage, such as festivities and sports events. It is important to note however, that these daily events are almost always linked to a CSR activity or subject. Tags were most often used when communicating about “Contests”, to motivate stakeholders and increase interactivity even more, but also to inform them about projects and associations they are working with.

DISCUSSION

The results in the previous section show that Iberdrola combines an informative content strategy and interactive linguistic strategy in its CSR communication on social media, resulting in interesting observations, which contribute to a clearer framework for CSR communication on social media in general.

Iberdrola’s information strategy on social media

If we look at the results of the frequency of CSR themes, there are certain conclusions we can draw for Iberdrola’s information strategy. Overall, posts that involved sports, specific holidays and/or thematic days, events and animals were present most often, both on Twitter and on Facebook. However, these themes were almost never present alone, meaning they were always accompanied by a more serious CSR subject, like an environmental theme, a sustainability theme, etc. This means that Iberdrola’s information strategy consists in combining serious CSR topics with matters like general interests and current affairs. Most possibly this serves to increase (less aware) stakeholders’ interest and engagement, in order to communicate the firm’s points on CSR in a more approachable way.

The same counts for the theme “Stakeholders”, which occupies a pretty high position on both social media channels. This theme was again often coded with other serious CSR themes, but in some cases its own as well. This could be explained by the fact that it is again a strategy that serves to increase engagement from stakeholders, whether they are existing customers or not, towards CSR-related themes. However, this theme can be considered as a ‘purer’ CSR theme as well, because it not only targets families and existing, or future customers, but also women and employees. It thus has a dual effect, because it is regarding speaking up about integrity and human rights, as well as it is about engaging and influencing potential customers.

Regarding the next main theme, there is a noticeable difference between the communication on Twitter and the communication on Facebook. On Twitter, the second most common CSR theme is about sustainability, whereas on Facebook, contests (even though they are for motivating stakeholders and CSR-related) are coded most often. We can thus assume that Iberdrola’s communication varies between Twitter and Facebook, probably because they want to target different types of stakeholders. While Twitter serves for ‘purer’ CSR content, Facebook is about targeting stakeholders and increasing engagement, whether it is for marketing reasons or for CSR reasons. This can be explained by the fact that Twitter content is more difficult to ‘spread’ than Facebook content, and thus directed at people who are already interested in Iberdrola’s content and already follow the account. Facebook on the other hand works as a way of motivating stakeholders to increase their interest in CSR information.

Thereafter, the environmental themes and projects were coded most often, carrying approximately the same importance on both social media outlets. These are purely CSR-related themes, and both have high coverage percentages, almost always higher than 13%. It can thus be concluded that although ‘engaging’ subjects have a considerable importance, the purer CSR-related themes certainly carry a lot of weight as well. It is important to note that although Iberdrola is an energy company and is thus expected to communicate more about emission reduction and such, social themes are in no way inferior to environmental and renewable themes. In other words, whilst it is expected from Iberdrola to prefer certain themes, the firm certainly does not. In a way, this might decrease stakeholders’ possible prejudices of ‘greenwashing’, simply because Iberdrola communicates about so much more.

Themes that are present to a lesser extent, although they should be mentioned, are cultural references, meaning local art and heritage, and innovation. These are both present on Twitter to a greater extent, thus we can assume that the communication strategy on Twitter is more about CSR in its pure form, which is informing, whereas Facebook combines the informing and engaging much more.

As mentioned in the theoretical framework, the CSR Observatory (Observatorio de RSC, 2018) concluded that the quality of CSR information in Spain in 2018 was still quite low, and that the information presented by most companies was far removed from the objective of accountability pursued by standards such as Law 11/2018 since it was more oriented towards a presentation of business achievements and good practices. However, the sector that achieved the best score was that of Oil and Energy, among which is Iberdrola, as third out of the 35 companies. If we look at Iberdrola’s CSR communication on social media, we can conclude that it is generally in line with its CSR reports, with many posts involving human rights and transparency around employees. However, there was not a single post involving corruption, something that is possibly a taboo in Spain. It is unsure, of course, if mentioning this would improve, or rather worsen stakeholder engagement.

Furthermore, if we compare the present study with previous findings about the energy sector (De Rosario et al., 2011), we can conclude that Iberdrola does communicate about environmental performance and social development, but not about economic value, which indicates that shareholders are certainly not the first target group in social media strategies of Oil and Energy firms. Also, in Iberdrola’s communication about environmental concerns, the company is not afraid to communicate about CO2 emissions, which is a negative impact of Oil and Energy companies. This seems to be the opposite of Van Zandvoort’s study (2016), although it is important to mention that Iberdrola already does well in terms of sustainable energy, so they have more positive aspects to communicate about.

Iberdrola’s interactive strategy on social media

The emotions that prevail in this study are those of positive valence, which are mostly coded with the theme “Stakeholders”. This confirms the observation of Fernández-Vallejo (2018, 2020) that the communicative strategy is on social media is about generating pleasant content for stakeholders specifically. The same counts for Facebook, with an even more coverage percentage for posts with positive valence.

Given the fact that the use of positive emotions is intended to generate admiration and reliability, and to reinforce the identity of the organization, it can be considered as part of the interactive strategy of social media. As mentioned before, Iberdrola has high ratings in terms of engagement and the management of its networks, which means that eliciting positive emotions does help. The only few structures that did elicit negative emotions were accompanied by environmental content, which means that a subtle combination of both can be favourable, to avoid hints of greenwashing.

As for informal textual elements, they were coded extremely often, meaning almost the entire social media strategy of Iberdrola is based on informal language. Although Gretry et al.’s study (2017) claims that an informal style can be perceived as inappropriate, Iberdrola uses this as an advantage, and it can certainly be considered as part of the interactive strategy. It should be noted however that Gretry et al.’s study was based on marketing content, and not on CSR content.

In a way, this means that Iberdrola’s main goal is to ensure customer loyalty, brand trust and to further cultivate ties, and they do this by using informal language. Because CSR content in reports can sometimes be difficult to read or comprehend by the public (Smeuninx, Clerck, & Aerts, 2018), Iberdrola combines serious CSR communication with informal language. We can thus conclude that informal language when communicating about CSR does foster closeness between stakeholders and brands, exactly because the follower and engagement rate for Iberdrola is so high.

The most prominent multimodal aspects in the corpus are hashtags and tags (or ‘appellative hyperlinks’), which were also coded when analysing the interactivity of Iberdrola’s social media posts. According to C Catamutto and Delfa (2016), hashtags are a sort of mechanism that serve to counteract the ephemeral tendency of social media messages. In other words, they help to the recovery (and to generate a certain textual unity) of posts, which are otherwise quite temporal.

On the other hand, appellative hyperlinks, or tags, are a special type of strategy consisting in referencing the relevant audience profile to increase changes of a commentary.Delfa (2017) states that this is also an essential element in digital discourse, because this hyper textuality leads to highly coherent discourse.

If we look at Iberdrola’s way of implementing these elements, we can conclude that the firm uses both strategies to enhance digital discourse and are thus using these characteristics of social media to improve searchability and engagement (Zappavigna, 2015). The coverage percentage for hashtags is high, with 37.56% on Twitter and 81.01% on Facebook. Tags have a slightly lower amount of coverage percentage, but still quite high, with 13.01% on Twitter and 16.22% on Facebook. This means that Iberdrola considers both of these elements as essential.

CONCLUSION

As exposed in section 2, there is a considerable discrepancy between CSR theory and reality, which is why organizations are unsure about the suitability of social media for CSR, often advocating for a media mix (Postma, 2015). In addition, there is confusion about the impact of CSR communication on social media, because some claim that there’s a tendency to engage more with non-CSR messages than with CSR messages, which might be because of the fact that the majority of CSR-related posts utilized an informing strategy rather than an interacting one, and organizations do not utilize the benefits social media offer (Cho et al., 2016). Others claim that CSR communication has a more positive impact than products/services, although it is less used on social media (Etter, 2013). Furthermore, findings state that European companies’ CSR reports tend to be credible, but mediocre, and that they must address different topics to address an audience of multiple stakeholders in order to prevent scepticism and raise awareness (Lock & Seele, 2016). This is where the urge for the development of CSR communication strategies rises to encourage publics to engage with corporations and various aspects of their CSR, by including both internal and external publics involved in the programs (Cho et al., 2016). There is thus a serious need for a research framework that could provide the basis for practical guidance.

In this study, we contribute to this goal by examining the CSR communication on social media of Iberdrola, a Spanish IBEX 35 company that scores well on both CSR information as social media management, we try to provide a clear example for other companies to follow. The results of RQ1, “which CSR-related themes are (most frequently) used?”, suggest that the CSR focus of Iberdrola’s information strategy varies from social media channel. On the one hand, the focus on Twitter is through the environment, sustainability, social investment, stakeholders and arts. On the other hand, the information strategy on Facebook is directed at stakeholders, and thus directed at motivating them in terms of engaging with CSR-related information. On both social media channels, the CSR themes are certainly important, with environment, sustainability and social investment being predominantly present, but they are informed of in a different way. n the other hand, appellative hyperlinks, or tags, are a special type of strategy consisting in referencing the relevant audience profile to increase changes of a commentary.Delfa (2017) states that this is also an essential element in digital discourse, because this hyper textuality leads to highly coherent discourse.

The results of RQ2, “is there a possible discursive pattern in the way the messages are formulated?”, suggest that that there is certainly an interactive strategy, with many elements that characterize digital discourse, and that certainly are made use of. There’s a clear use of structures that elicit (positive) emotions, informal language use and highly discursive elements, such as hashtags and tags. Positive structures, emojis, use of 1st and 2nd person and hashtags are amongst the most used elements. All these elements are efforts at increasing brand trust, engagement, and admiration.

The results of RQ3, “are certain themes combined with certain textual elements?”, suggest that there are some small underlying patterns between the information strategy and the interacting strategy, but that the overall way in which Iberdrola wants to communicate its CSR information is through informal and (positive) emotional language. The only observation that can be made, is that the scarce use of negative emotional language is always combined with an environmental theme, most probably to avoid ‘greenwashing’ accusations.

RECOMMENDATIONS

Naturally, the study has its limitations (e.g. sample size, time period and so on). Although posts from two social media outlets were studied, and the sample size was representative enough, only one IBEX 35 energy company was studied in detail. In addition, the time period of the social media posts was during an eventful three-month period, with important holidays and climate conferences happening. A longitudinal analysis could reveal whether communication changes over time, which should be subject to further research. A study that includes retweets, favourites and likes could also provide more insight in engagement with the different types of content and textual elements, which is now merely based on assumption. All of these elements would be of great relevance in response to the current need of a research-based framework on CSR communication on social media.

REFERENCES