1 The grass is always greener
A qualitative content analysis of CSR media coverage of four Dutch newspapers
Matteo Sabbatini 10679189
Universiteit van Amsterdam Lecturer: Jelle Boumans
2 Abstract
Corporate social responsibility (CSR) is a topic which has been studied extensively. However very little research has been done on media coverage of CSR. This research will attempt to fill this gap in the literature by studying the media coverage of the CSR practices of
supermarkets in four Dutch newspapers. In addition to contributing to the literature, this will provide companies with a practical tool to aid them in constructing effective CSR
communication campaigns. The following research question stands at the base of this study: ‘Which factors influence the media coverage of Dutch supermarket’s CSR activities?’. Two hundred articles of four different Dutch newspapers were studied using qualitative content analysis. The variables frame, tone, type of newspaper and type of supermarket were chosen for the analysis. Three one-way variance analysis were conducted to test the possible
relations between dependent variables frame and tone independent variables type of newspaper and type of supermarket respectively. Furthermore, descriptive statistics were used to conclude which actors were named most in newspaper coverage. The results of this research showed that soft discount stores are written about significantly more positive than hard discount stores. No significant results were found for the relation between frame and type of newspaper, and tone and type of newspaper. The actors quoted most in the articles were actors affiliated with supermarkets. The results of this research give way to future, similar, research by making a first step in documenting trends in the newspaper media
landscape on CSR reporting. Furthermore, this research can serve as an aid to companies who wish to know more about CSR media coverage to enhance the effectiveness of their own CSR communication campaigns.
3 Introduction
The supermarket industry in the Netherlands is a highly competitive market. This battle has intensified since 2003 when consumer distrust in the biggest supermarket in the Netherlands, Albert Heijn, increased (Luijten & Reijnders, 2009). It is also an industry where corporate social responsibility (CSR) plays an important role. External organisations have reshaped the perception of the consumer on the supermarket industry by for example starting campaigns focussed on the ‘plofkip’ (fattened chicken) (Jansen, 2014).
CSR is a topic which has been extensively studied. Furthermore, it is a topic of growing importance for the corporate world (Sen & Battacharya 2001; Carrol 2001). However, existing literature on CSR seems incoherent and sometimes even contradictory. Even the definition of CSR is an intensely debated topic (Carroll, 1999). Many authors define CSR by using the assumption that corporations are responsible to environment and society (Dolnicar & Pomering, 2009; Wagner, Lutz & Weitz, 2009). This paper will attempt to contribute to the search of a universal definition of CSR by studying how Dutch newspapers cover CSR
initiatives of Dutch supermarkets. This will, in turn, aid researchers who are interested in conducting future research about CSR by providing a point of view on the definition of CSR which has not been extensively researched. Little research has been conducted on the link between CSR reporting and media coverage. However, one study found that firms engaged in CSR enjoy more positive media attention (Cahan, Chen, & Nguyen, 2015). Further research on the link between CSR and media coverage is lacking; current studies are focussed on different varieties of CSR and CSR communication strategies. Research shows that, among many other things, companies should be open and clear about their CSR practices (Dawkins, 2004; Schmeltz, 2012), proactively communicate about CSR (Wagner, Lutz & Weitz, 2009), and interact with their consumers when communicating about CSR (Colleoni, 2013). This research will attempt to fill in the void in the literature by conducting a case study on Dutch supermarkets. This study will compare supermarkets with inherently different approaches to CSR by examining media coverage, thus adding to the research done by the authors named above and creating grounds for future research. Furthermore, the results of this study can serve as a resource for the CSR communication of companies. When more is known about the different factors which possibly play a role in media coverage of CSR, corporations will be better equipped to make successful CSR communication strategies.
4 The aim of this study is to find out which factors influence the tone of media coverage on Dutch supermarkets’ CSR initiatives. By doing so a first step will be made in bridging the academic gap on the relation between CSR and media coverage. Moreover, this study will look at the different frames used in media reporting about Dutch supermarkets’ CSR activities. Again, little research has been conducted about framing regarding CSR media reporting. Various scholars have studied how companies can frame their CSR practices in internal and external communication (Tengblad & Ohlsson, 2010; Schmeltz, 2012) and what effect framing has on the judgement of consumers on CSR communication of a company (Wang, 2007; Case, 2008). However, the way CSR communication is framed in the media is yet unexplored. This study will attempt to fill this gap in the current literature. Moreover, studying which frames are used in the media coverage will give companies a guideline for how they should frame their CSR communication towards newspapers.
To achieve the goals listed above this research will attempt to answer the following question: Which factors influence the media coverage of Dutch supermarkets’ CSR activities?
Since this topic is complex, a selection was made which variables to research in this paper. Current literature was used to operationalise and define these variable and concepts. Tone, Actor, type of supermarket, Frame and Type of newspaper are the most important variables in this paper. A more detailed description of the variables, operationalisation, and research questions will be given in the method section of this paper. This paper will be structured in the following way. First an overview of the concepts and variables used will be given. All the relevant variables will be operationalized and defined. This will be done using existing literature combined with additions from the author. Furthermore, examples will be given on how the variables appear in the sample. After this an analysis of the chosen method of analysis will be given. In the results section of this paper the results of the research will be explained and shown graphically. The possible implications of these results will be discussed in the conclusion of the paper, along with a short summary of the findings. Finally, in the discussion the limitations of this research will be addressed, and recommendations for future research will be made.
5 Conceptual Model
Problem statement
The following question stands at the base of this research: ‘Which factors influence the media coverage of Dutch supermarkets’ CSR activities?’ This research question was constructed to help resolve challenges on two different, but interconnected, levels: academic and practical. Firstly, academic research on CSR media coverage is lacking. Very little research has been done on this subject, and none in particular on the Dutch supermarket industry. This problem also shows the connectedness of the academic and the practical. Since no studies have been done on this topic, supermarkets would be unable to consult academic papers if they, for example, want to know which actors are quoted most in newspapers. Another problem of the academic literature on this subject is the fact that many authors disagree on a conceptual level about the definition of CSR. Again, this problem ties to both the academic and professional world. Future research on CSR and CSR communication would benefit from a universal definition of CSR since no time would have to be spent on discussing contradicting
definitions. On the other hand, a clear definition of this concept would help companies to get a clear image of what exactly is meant by CSR, and consequently adapt their policies
accordingly. Finally, research has shown that companies struggle to create CSR
communication strategies which fit in the media landscape (Schmeltz, 2012). This paper will attempt to solve these problems by posing the following research questions.
RQ1: Does the type of newspaper affect the tone of media coverage?
The variable tone is the dependent ratio variable, in the analysis which will be used and type of newspaper, the independent nominal variable. In this research the variable tone describes how the CSR initiative of a supermarket is written about in an article. Existing literature makes the distinction between quality newspapers and popular newspapers (Semetko & Valkenburg, 2000; Ruigrok & van Atteveldt, 2005). These authors argue that quality newspapers focus more on the factuality of an issue and in general attempt to avoid sensationalist content. On the other hand, popular newspapers are more sensationalist and attempt to elicit an emotional response from the reader. It is assumable that these newspapers write about a popular topic such as CSR in a different manner, however no research has been done to support this assumption. Therefore, this research question was constructed to test this possible relation.
6 RQ2: Does the tone of media coverage differ significantly between different types of
supermarkets?
In this research question, tone is the dependent ratio variable and type of supermarket is the independent nominal variable. The different types of supermarkets included in this variable are: hard discount stores, soft discount stores, and ecological supermarkets. Hard discount stores carry mostly own-brand products and focus on low prices and operating costs, while soft discount stores carry more premium brands and focus on service and quality more. Ecological supermarkets are supermarkets which only sell sustainable food. The possible relation in this research question is interesting to study since current research implies that there might be such a relation. Wagner, Lutz & Weitz (2009) state that companies which use proactive communication strategies when communicating their CSR generate less perceived hypocrisy toward the company. Since ecological supermarkets have sustainability as their core business model, assumedly their communication about CSR might seem more proactive than hard discount stores which are known for having the lowest possible prices as their purpose.
RQ3: Does the type of supermarket affect the type of frame used in media coverage of Dutch supermarkets?
Frame is the nominal dependent variable and type of supermarket the independent nominal variable. As in RQ2, this relation might be expected because the media might have an existing bias towards different types of supermarkets. Furthermore, different types of supermarkets may focus on a specific part of an issue in their CSR campaigns. These
differences may manifest themselves in the media coverage of said campaigns. Research has shown that companies which use discourse based on social responsibility and economic profit in their CSR communications have more positive ratings on their CSR initiatives by
consumers (Nielsen & Thomsen, 2007). Following these results, RQ3 will further study this possible relation.
RQ4: Which actors are quoted most in media coverage of Dutch supermarkets?
An actor is a party (human or non-human) which can have influence or be influenced by a certain issue (Bechmann & Lomborg, 2012). This is a descriptive research question which was constructed to create an overview of the trends on media coverage of this subject. No assumptions are made since no research was found on this topic. The results of this research
7 question will show which actors are named most, and which least, in the newspaper articles studied.
8 Method
To attempt to answer the research questions in this paper a quantitative content analysis was conducted. ‘Content analysis is a systematic, objective, and quantitative method for studying communication messages and developing inferences concerning the relationship between messages and their environment’ (Krippendorff, 1980). For this research, qualitative content analysis was the correct method, since newspaper articles are communication methods, and the aim of this study was to make statements about the possible relationships between the concepts found in these messages.
To gather the data necessary for the analysis the LexisNexis database was utilized. After attempting various permutations of the search terms, the following were used, using the ‘power search’ function:
(Supermarkt OR lidl OR Aldi OR Jumbo OR marqt OR ekoplaza) AND (maatschappelijk verantwoord ondernemen) OR (maatschappelijk OR verantwoord OR biologisch) AND NOT SUBJECT (recept!)
All national newspapers were included in the search. This resulted in 1930 newspaper articles. A stratified sample was taken to obtain the research units, which are the newspaper articles in this case. First, results were filtered on type of newspaper. After this, 60 articles were selected from each newspaper which is relevant to this paper: NRC handelsblad (n = 223), De Volkskrant (n = 334), De Telegraaf (n = 125) and Algemeen Dagblad (n = 140). The final sample was acquired by using a simple random sampling strategy. 60 articles from each newspaper were selected using a random number generator. A pre-test sample was acquired by pooling all the articles together and taking a simple random sample of n = 15. The coder read these articles to get a general idea of which variables were involved. After doing this, the only change made to the conceptual model was to add the ecological supermarkets to the variable type of supermarket.
Operationalisation & Definition Type of newspaper
In earlier research done on Dutch newspapers a common distinction between popular newspapers and quality newspapers can be found. These authors categorize NRC
9 and the Algemeen Dagblad as more popular, sensationalist newspapers (Semetko &
Valkenburg, 2000; Ruigrok & van Atteveldt, 2005). The criteria used in making these categorizations were readability, focus on popular issues, and target audience. Considering that the authors named above have also conducted a content analysis on Dutch newspapers, their categorization will be used as a basis for this research. The Dutch newspapers:
‘Volkskrant’ and ‘NRC Handelsblad’ were chosen as the quality newspapers in this study while ‘De Telegraaf’ and ‘Algemeen Dagblad’ were categorized as popular newspapers. Type of newspaper has been coded as a nominal variable with the two dimensions named above (quality and popular).
Type of supermarket
Once again, existing literature was used for the operationalisation of this variable. Per the literature, supermarkets are usually categorized as either a ‘hard discount store’ or a ‘soft discount store’. Hard discount stores usually carry a majority of ‘private labels’ (brands owned by the retailer), have aggressive pricing strategies, and low operating costs. On the other hand, soft discount stores have a larger variety of product, higher operating costs, higher prices, and lower percentage of private labels (Colla, 2003). German supermarket chains Aldi and Lidl are examples of hard discount chains (Denstadli, Lines, & Grønhaug, 2005) while Albert Heijn is seen as a soft discount store (Luijten & Reijnders, 2009). A similar categorization of supermarkets has been made in this paper: Albert Heijn and Jumbo were categorized as soft discount stores while Lidl and Aldi were categorized as hard discount stores. Furthermore, a third category of supermarkets will be researched. Namely, the supermarkets which focus on ecological/biological products. The supermarkets Marqt and Ekoplaza were included in this category.
Tone of coverage
Tone has been used as a variable in many studies. The most used distinction is: negative, neutral, or positive (Collins, Abelson, Pyman, & Lavis, 2006; Miller, Tyler & Mor, 2013). The authors of these papers coded the tone of an article by looking at the way the media described the issue. Since none of these studies are about CSR coverage of Dutch supermarkets, this research will use a tailored operationalisation.
The variable tone will be coded as a ratio/scale variable with the following three dimensions: negative (-1) , neutral (0), and positive (1).
10 If the article focusses on contradictions or unethical behaviour of the supermarket the article will be coded as negative. This is also the case if the author of the newspaper or other actors quoted in the article are cynical about the CSR initiative and/or is trying to convince the reader that the supermarket in question should be distrusted. If the purpose of the news article is only to inform the reader about the CSR initiatives of one or more supermarkets the article will be coded as neutral. No statements are made about the perceived social desirability of the initiative. Neither is the author or other actors quoted trying to persuade the reader to take on a certain viewpoint regarding the issue at hand. Various opinions might be given about the supermarket or it’s CSR initiative however no winner is chosen. If the positive aspects of the CSR initiative are clearly underlined the article will be coded as positive. The issue is talked about in a way that clearly states the benefits it has on society and environment. The author or other actors quoted are trying to convince the reader that the CSR initiative is socially and ethically desirable. The supermarket(s) named in the article are positively spoken of. Since tone is a latent concept which can be subjectively interpreted, the codebook of this research contains questions which will help the coder interpret the articles based on the
operationalization given above. Actors
An actor is a party (human or non-human) which can have influence or be influenced by a certain issue (Bechmann & Lomborg, 2012). The actors to be included in the variable were determined after the pre-test and the study of the literature. Actors affiliated with
supermarkets, non-governmental institutions, and commercial institutions were included. Moreover, scientists, journalists, consumers, suppliers of the food industry, and the
‘consumentenbond’ (a Dutch organisation which represents the interests of consumers) were included. These actors were coded as present when they were directly, or indirectly named or quoted in the article.
Frames
‘Framing essentially involves selection and salience. To frame is to select some aspects of a perceived reality and make them more salient in a communicating text, in such a way as to promote a particular problem definition, causal interpretation, moral evaluation, and/or treatmeant recommendation for the item described’ (Entman, 1993). In essence, framing is how a certain issue is presented in the media. This research will use five frames which have been often used in Communication research. These frames are: the human-interest frame, the
11 economic consequence frame, the conflict frame, and the morality frame (Semetko &
Valkenbug, 2000). More than one frame can be present in the article. The operationalization of the specific frames is as follows. The frame ‘human interest frame’ was excluded after conducting the pre-test. This frame proved difficult to code, and when it was present, it was always present as economic consequence, or better suited to another frame. This was the only frame were such difficulties were encountered, thus the choice was made to exclude this variable.
Responsibility frame
‘When an issue is framed according to the responsibility frame it is presented in such a way that the responsibility or blame for the cause or the solution is placed on political authorities, individuals or groups’ (Dirikx & Gelders, 2010). If the article in question contains a direct or indirect mention of a supermarket or other party which is responsible for a certain issue, the responsibility frame will be coded as present. Furthermore, this frame will also be coded as present when the resolution of a certain issue is the responsibility of a certain actor.
Economic consequence frame
‘The economic consequences frame emphasizes the manner in which an issue will
economically affect people’ (Dirikx & Gelders, 2010). The economic-consequence frames present a certain issue by underlining the financial or economic side. This frame is present in the article when mentions are made of a financial climate or situation. This can for example be the case when the article in question describes the gains or losses made after the
implementation of a new biological line of food of a supermarket. In a great majority of the cases this frame is present when numbers (of a currency) are mentioned.
Conflict frame
‘Articles that reflect the conflict frame emphasize conflicts between parties/individuals and stress the points of divergence’ (Dirikx & Gelders, 2010). As the name implies, this frame presents a conflict. In this context, it might be an incongruity between promises of a
supermarket and their behaviour. The conflict frame shows two or more parties with differing opinions and might choose a winner in these arguments. Furthermore, the conflict frame can be present in an article when past or possible problems with an issue are underlined.
12 ‘The morality frame presents situations from a religious/moral angle. As it is difficult for journalists to give a moral message while adhering to the journalistic norm of objectivity, this is often bypassed by quoting others’ (Semetko and Valkenburg, 2000; Dirikx & Gelders, 2010). The morality frame frames an issue by focussing on the moral side of an issue. An issue or opinion might explicitly have been labelled as good or bad, or this might be implied in the text. This frame is also present when the author of the article does not make an
implication or direct observation about the morality of an issue, but simply states the moral dimension of the issue so that the reader might decide for himself.
Factor analysis
To measure the latent variables of the frames, a factor analysis was used to, where necessary, construct a scale. This method was used since the presence of a certain frame was measured by using different questions in the coding process. When using a factor analysis it will become clear if the frames are multidimensional concepts. If more dimensions are found, they will be interpreted by looking at the questions of the variables which form the dimension.
For the variable responsibility frame two factors with an eigen values above 1 were found. These two factors accounted for 48.9% and 30.2% of total variance. The varimax rotated component matrix showed that items responsibility towards society and responsibility towards environment had a factor loading of 0.88 and 0.87 respectively with factor 1. Items problem caused by actor and problem solved by actor had a factor loading of 0.86 and 0.85 with factor 2, respectively. The ‘compute variable’ command was used to construct scale variables of the different factors. In this case, both scales were reliable. Factor 1 had a Chronbach’s Alpha score of α = 0.66 and factor 2 α = 0.73. It is probable that factor one describes the responsibility frame when the article talks specifically about an actor having a responsibility towards society or environment. Indeed, many of the articles coded talked about environmental issues and CSR being a responsibility of a company towards society or the environment.
For the variable conflict frame two factors with an eigen value above 1 were found as well. These two factors accounted for 49.6% and 30.2 % of total variance. The items
inconsistencies named’ and conflict with established norms and/or values had a factor loading of 0.87 and 0.90 respectively with factor 1. The items quotes from different actors and policies/standpoints from different actors had a factor loading of 0.85 and 0.88. These
13 high factor loadings are expected since the items which are formed in to a factor are very closely related. When an inconsistency of a CSR policy was named in an article, it was almost always implied that this inconsistency is not desirable. A scale variable was again constructed for both factors. The Chronbach’s alpha for factor 1 was 0.78 and for factor 2 it was 0.74. In both cases the values indicate a reliable scale, and it was not needed to delete items for a higher reliability.
One factor with and eigen value above 1 was found for the economic consequence frame. This factor accounted for 57.85% of total variance. The item is the economic climate named? had a factor load of 0.90 whilst the items is an economic benefit or disadvantage stated and is the economic position of an actor stated? had lower factor loads of 0.70 and 0.67
respectively. The scale variable which was constructed for this factor was barely reliable, with a Chronbach’s alpha score of 0.61.
For the morality frame one factor with an eigen value of above 1 was found. This factor accounted for 74.82% of total variance. The items on the scale, ‘are the words good or bad explicitly used?’, ‘are moral implications or complexities stated?’, and ‘guidelines and/or rules on expected behavior’ had factor loads of 0.90, 0.83, and 0.86. This scale was found to be reliable with a Chronbach’s alpha score of 0.83.
Intercoder reliability was calculated with Lotus. The value 0.86 signifies a reliable codebook, and there is no need to remove variables. The variable actor affiliated with supermarket had the highest Lotus value, 1.00, while the variable supplier(s) of the food industry had the lowest Lotus value, 0.68.
14 Results
Descriptives
The sample size consisted of N=200 newspaper articles. When looking at the variables ‘supermarket’ and ‘tone’ a clear distinction can be seen. The variables ‘hard discount store’ and ‘soft discount store’ are mentioned the most with respectively 33.2% and 28.1%. Most ofthe articles (n = 108), 53.7% were written with a neutral tone (16.9%, n = 34) were written in a negative tone, and 28.6% (n = 57) had a positive tone. These findings are further
confirmed by the characteristics of the variable ‘tone’ (M = 0.12, SD = 0.66). Regarding the actors named in the articles, the most prominent one named was ‘actor
affiliated with supermarket’. This variable was present in more than half the articles (n = 108, 53.7%). The variable which had the second highest frequency was ‘actor affiliated with non-governmental agency’ (n = 37,18.6%). The least named actors were ‘actors affiliated with trade unions’ (n = 7,3.5%), ‘consumentenbond’ (n = 7, 3.5%), and ‘scientists’ (n = 13, 6.5%). The frequencies and percentages of all the actors are shown in table 1. This answers RQ 4: Which actors are quoted most in media coverage of Dutch supermarkets? Actors affiliated with supermarkets are quoted the most. Thus, in the majority of the cases, supermarkets get the opportunity to talk about their CSR initiatives in the press.
The economic frame was used most (M = 0.52, SD=0.84) and the morality frame the least (M = 0.29, SD = 0.24). The frequencies and percentages of the quoted actors are named in table 2.
15 Table 1: Frequencies and Percentages of Actors Quoted.
Actor Frequency Percentage
Affiliated with supermarket 108 53.70% Affiliated with NGO 37 18.60% Consumer 23 11.20% Affiliated with governmental institution 18 8.70% Suppliers of food industry 17 8.30% Affiliated with commercial institution 15 7.30% Scientists 14 6.80% Affiliated with trade union 7 3.40% Consumentenbond' 7 3.40% Journalist 0 0.00%
Table 2: Means and Standard deviations of frames used.
Frame Mean Std. Deviation Economic-Consequence 0.52 0.36 Responsibility 0.51 0.34 Conflict 0.31 0.39 Morality 0.29 0.24
16 One-way variance analysis.
To test the possible relationship between tone and newspaper a one-way variance analysis was conducted. This was the appropriate analysis since the dependent variable (tone) is a ratio variable and the independent variable (newspaper) is a nominal variable. The results showed that the Algemeen Dagblad on average wrote the most positive (M = 0.24, SD = 0.69) and De Telegraaf the most negative (M = 0,041, SD = 0.67). Since the groups were of equal size we assumed equal variances. No significant results were found after conducting the variance analysis (F (3) =1.88, p = n.s). Thus, RQ1 cannot be confirmed. The tone does not differ significantly between the high quality and popular newspapers. The expected
assumption that these two different types of newspapers write about CSR in a different tone due to their inherent structure cannot be confirmed by the results of this research. Another one-way variance analysis was conducted to test the possible relationship between type of supermarket and tone. In this case tone was the independent variable and type of supermarket the independent variable. Since the groups were not of equal size, Levene’s test for equal variances was conducted. The result of this test was not significant (F(6,192)=0.57, p=0.76) so once again we assume equal variances. The results of the variance analysis were found to be significant (F (6,192) = 3.98 p=0.01). Since the variable ‘type of supermarket’ has seven different dimensions, a Bonferonni post-hoc test was conducted to see between which supermarkets the difference in tone was found. Soft discount stores (M = 0.30, SD = 0.66) were mentioned in articles with a significantly more positive tone that hard discount stores (M = -0.18, SD = 0.63) (Mdifference =- 0.48, SD=0.12, p=0,01). RQ 2 can be confirmed, type of supermarket does indeed affect tone. Soft discount stores are written about more positively than hard discount stores. Yet another one-way variance analysis was conducted to test RQ3, with type of supermarket as independent variable and all the constructed scale variables of the frames as dependent variables. No significant results were found for any of the frames.
17 Conclusion
Corporate social responsibility is a hot topic for all industries, and the supermarket industry is no exception. However, many companies struggle to find adequate strategies to handle their CSR communication. Little research has been done on how CSR is reported in newspapers and which factors influence this process. This paper attempts to solve these two problems by making a first step in filling the void in the literature and by doing so providing a practical overview which can potentially be used by supermarkets.
In this research, no significant results were found in the analysis used for the relation between type of newspaper and tone. Thus, the average tone does not differ significantly between quality and sensational newspapers. Moreover, most of the articles were written in a neutral tone, the facts were simply stated and no judgement was made on the initiative. This implies that newspapers are not biased against or for the CSR initiatives of supermarkets.
Supermarkets should therefore not need to cater to each newspaper differently when
communicating about their CSR. This means that no additional media training is required for employees of the company which might have been effective had there been a significant difference in tone between the various newspapers.
A significant result was however found, when analyzing tone of media coverage and type of supermarket. After conducting the one-way variance analysis, and consequently the
Bonferonni post-hoc test, it was found that articles which talk about ecological supermarkets are written in a significantly more positive tone than articles which talk about hard-discount stores. Since this particular topic has not been studied, making direct links to existing literature proves challenging. However, when the copious amounts of literature available on how companies should communicate their CSR are consulted, the results of this paper seem to be in line with what other authors have found. Research has shown that CSR practices are rated more positively when companies communicate in a clear, open, and credible manner (Schmeltz, 2012). This is in line with the results of this paper. It is probable that the CSR initiatives of hard-discount stores are seen as less credible than the those of ecological supermarkets, since the latter’s core business (selling organic and sustainable food) can be seen as CSR for hard-discount stores. This implication can be expanded when looking at the study by Wagner, Lutz & Weitz (2009). These authors conclude that companies which use a reactive CSR communication style are perceived as more hypocritical than companies which use a proactive communication style. Again, this is in line with the findings of this paper
18 since hard-discount stores have, for example, started their campaign on sustainable chicken and reducing the amounts of ‘plofkip’ (fattened chicken) in their stores only after a critical campaign of activist group ‘Wakker Dier’ (Jansen, 2014).
No significant results were found after conducting the analysis to find out if the type of newspaper influences the type of frame used. Once again, these results imply that newspapers are unbiased in the use of their frame, at least in the sense that they do not use specific frames more for specific supermarkets. However, the results do show that the responsibility frame and the economic consequence frame were most commonly used to describe the CSR initiatives. This implies that supermarkets should communicate their CSR practices with a focus on these two frames. This can be done by focusing on the responsibility of the
supermarket towards the environment and the economic benefit of the CSR initiative in press releases and other communication such as social media and the company website.
Furthermore, employees can be given media training and instructed to highlight these factors when answering questions of the press. The finding that the responsibility frame is used most agrees with current literature. Wagner, Lutz, & Weitz, (2009) define CSR as ‘a company’s obligation to exert a positive impact and minimize its negative impact on society’. However, other authors have more difficulty defining CSR. Nielsen & Thomsen (2007) even describe it as a ‘diffuse and almost non-operational concept’. The results of this research imply that in following studies, confusion about the definition of CSR might be avoided by using a definition which highlights the responsibility of the company towards the environment and society, since this is the way the media frames the issue and will thus be more in line with how the concept is used in ‘the real world’.
Out of all the actors, actors representing supermarkets/the food industry were mentioned most in the articles coded. Thus, the supermarkets themselves are an important source of
information for journalists when reporting about CSR initiatives. While this may seem obvious, these findings highlight the importance for supermarkets to have well informed and trained spokespersons which communicate with the media. In the articles which were
included in the sample of this research, supermarkets had many opportunities to explain their CSR initiatives. This was also the case when, for example, an activist group criticized a supermarket. In most cases actors related to supermarkets were also quoted or mentioned in these articles.
19 Limitations and recommendations
This paper has taken a step in the right direction to research newspaper reporting of CSR. However, some limitations were encountered. Firstly, the sample size is rather limited. Only 50 articles were coded per newspaper, with a total of 200 coded articles. A limited sample size means that on should be cautious when generalizing the results of this study. Future research on this topic should in any case contain a larger sample of research units.
Furthermore, the diversity of the sample could have been larger. Only four different Dutch newspapers were researched. Statements made about quality or sensationalist Dutch newspapers should be made with this fact in mind. Future research could include all Dutch national newspapers to the sample size. By doing so a complete picture of the newspaper media landscape in the Netherlands would be given. Moreover, it would be interesting for this future research to go more in-depth on the newspapers researched. It could be researched if certain newspapers report about CSR more than other newspapers in general. By doing so the literature about this topic will be further expanded. This kind of research will also give companies a broader view of the media landscape, which can be useful for them when implementing their CSR communication strategies. Secondly, the variable tone proved to be a complex variable, especially in combination with ‘actors mentioned or quoted’ and ‘type of supermarket’. In some cases, an article would talk positively about one supermarket and negatively about another and giving the clear preference to a certain supermarket. In other cases, a third-party actor might have made a statement on this conflict. In these cases, it was challenging to answer ‘positive’ ‘neutral’ or ‘negative’ to the question: ‘what is the tone of the article?’ even when looking at the question provided as assistance in the codebook. In fact, these questions made the matter more confusing at times, since the answers were
sometimes contradictory. This problem can be resolved in future research by operationalising this variable even more carefully. All possible cases should be considered and explained in the codebook so that the coder has a clear idea on how to code tone. Finally, only one coder was used to code the articles. When coding many articles in one extended period it is possible that a singular coder might show certain biases while coding. Furthermore, after several hours of coding, a method or routine of reading articles was present with the coder. While this enhanced the efficiency and speed of the coding, it is possible that certain nuances of articles might have been missed while coding articles for a prolonged period. This problem can be remediated in future research by using more than one coder to code the sample of articles.
20
While limitations were certainly present in this research, may well serve as a springboard for future research on this topic. The variables which were explored and researched are certainly a good foundation for future research. Since this topic has been researched so little, the possibilities for future research are numerous. A similar research could be done on an international scale, comparing newspapers from different variables. This would give
companies operating on an international information about the media trends regarding news coverage on CSR. With this information, the companies can adjust, or homogenize their communication across the different countries in which they operate. Consequently, this paper has served as a basic practical tool for supermarkets. While certainly not definitive, the findings of this research can serve as a starting point for supermarkets who want to know more about CSR coverage in Dutch newspapers and as a motivation for scholars to conduct research on this topic.
21 References
Bechmann, A., & Lomborg, S. (2013). Mapping actor roles in social media: Different perspectives on value creation in theories of user participation. New media & society, 15(5), 765-781.
Cahan, S., Chen, C., Chen, L., & Nguyen, N. (2015). Corporate social responsibility and media coverage. Journal Of Banking & Finance, 59, 409-422.
http://dx.doi.org/10.1016/j.jbankfin.2015.07.004
Carroll, A. B. (1991). The pyramid of corporate social responsibility: Toward the moral management of organizational stakeholders. Business horizons, 34(4), 39-48.
Carroll, A. (1999). Corporate Social Responsibility: Evolution of a Definitional Construct. Business & Society, 38(3), 268-295.
http://dx.doi.org/10.1177/000765039903800303
Case, B., Exchange, R. F. P., & Partnerships, C. (2008). Corporate social responsibility priming and valence of CSR framing on CSR judgments.
Colla, E. (2003). International expansion and strategies of discount grocery retailers: the winning models. International Journal Of Retail & Distribution Management, 31(1), 55-66. http://dx.doi.org/10.1108/09590550310457845
Colleoni, E. (2013). CSR communication strategies for organizational legitimacy in social media.Corporate Communications: An International Journal, 18(2), 228-248.
http://dx.doi.org/10.1108/13563281311319508
Collins, Patricia A., et al. "Are we expecting too much from print media? An analysis of newspaper coverage of the 2002 Canadian healthcare reform debate." Social Science & Medicine 63.1 (2006): 89-102.
Dawkins, J. (2005). Corporate responsibility: The communication challenge. JCOM, 9(2), 108-119. http://dx.doi.org/10.1108/13632540510621362
Denstadli, J. M., Lines, R., & Grønhaug, K. (2005). First mover advantages in the discount grocery industry. European Journal of Marketing, 39(7/8), 872-884.
22 Dirikx, A., & Gelders, D. (2010). To frame is to explain: A deductive frame-analysis of Dutch and French climate change coverage during the annual UN Conferences of the Parties. Public Understanding of Science, 19(6), 732-742.
Ellerup Nielsen, A., & Thomsen, C. (2007). Reporting CSR-what and how to say it?. Corporate Communications: An International Journal, 12(1), 25-40.
Entman, R. M. (1993). Framing: Toward clarification of a fractured paradigm. Journal of communication, 43(4), 51-58.
Hopmann, D., Vliegenthart, R., De Vreese, C., & Albæk, E. (2010). Effects of Election News Coverage: How Visibility and Tone Influence Party Choice. Political Communication, 27(4), 389-405. http://dx.doi.org/10.1080/10584609.2010.516798
Krippendorff, K. (2004). Content analysis: An introduction to its methodology. Sage.
Luijten, T. & Reijnders, W. (2009). The development of store brands and the store as a brand in supermarkets in the Netherlands. The International Review Of Retail, Distribution And Consumer Research, 19(1), 45-58. http://dx.doi.org/10.1080/09593960902781268
Miller, E. A., Tyler, D. A., & Mor, V. (2013). National newspaper portrayal of nursing homes: Tone of coverage and its correlates. Medical care, 51(1), 78.
Pomering, A. & Dolnicar, S. (2008). Assessing the Prerequisite of Successful CSR Implementation: Are Consumers Aware of CSR Initiatives?. Journal Of Business Ethics, 85(S2), 285-301. http://dx.doi.org/10.1007/s10551-008-9729-9
Ruigrok, N., & Van Atteveldt, W. (2007). Global angling with a local angle: How US, British, and Dutch newspapers frame global and local terrorist attacks. The Harvard International Journal of Press/Politics, 12(1), 68-90.
Semetko, Holli A., and Patti M. Valkenburg. "Framing European politics: A content analysis of press and television news." Journal of communication 50.2 (2000): 93-109.
Sen, S., & Bhattacharya, C. B. (2001). Does doing good always lead to doing better?
Consumer reactions to corporate social responsibility. Journal of marketing Research, 38(2), 225-243.
23 Schmeltz, L. (2012). Consumer‐ oriented CSR communication: focusing on ability or
morality?.Corporate Communications: An International Journal, 17(1), 29-49. http://dx.doi.org/10.1108/13563281211196344
Tengblad, S., & Ohlsson, C. (2010). The framing of corporate social responsibility and the globalization of national business systems: A longitudinal case study. Journal of Business Wagner, T., Lutz, R., & Weitz, B. (2009). Corporate Hypocrisy: Overcoming the Threat of Inconsistent Corporate Social Responsibility Perceptions. Journal Of Marketing, 73(6), 77-91. http://dx.doi.org/10.1509/jmkg.73.6.77
Wang, A. (2007). Priming, framing, and position on corporate social responsibility. Journal of Public Relations Research, 19(2), 123-145.
24 Appendix I: Codebook
Codebook
V1. In which newspaper is the article published? 1. NRC Handelsblad
2. De Volkskrant 3. De Telegraaf 4. Algemeen Dagblad
V2. Which supermarkets are mentioned in the article? 1. Hard discount stores (e.g. Lidl or Aldi)
2. Soft discount stores (e.g. Albert heijn or Jumbo) 3. Ecological supermarkets (e.g. Marqt or Ecoplaza) 4. No specific supermarkets are named
5. Ecological supermarkets and Soft discount stores 6. Ecological supermarkets and hard discount stores 7. Hard discount stores and soft discount stores
D. What is the tone of the article towards the CSR initiative(s) of a specific supermarket or the industry as a whole?
Tone of the article towards the supermarket/supermarket industry will be coded with the help of the following question, these can be used by the coder as a tool or assistance. When reading the article the overall ‘feel’ or tone was also taken in to account (concerning factors such as the title of the article).
Negative tone
1. Does the article name inconsistencies of the CSR policy (or policies) discussed? 2. Does the article name inconsistencies between CSR communication of supermarkets
25 3. Is the cost of biological/organic/fairtrade products in supermarkets criticised in the
article?
4. Is the availability or accessibility of biological/organic/Fairtrade products in supermarkets criticised?
Neutral Tone
5. Does the article state facts about a CSR initiative without judging the value of said initiatives?
6. Does the article give positive and negative arguments regarding the CSR initiative without choosing a ‘winner’?
Positive Tone
7. Is the CSR initiative explicitly labelled as good, useful, or practical? 8. Is the CSR initiative labelled as a ‘step in the right direction’?
9. Does the article state that supermarkets are increasingly involved in CSR initiatives, or offer a larger amount of biological/organic/Fairtrade products than earlier years? 10. Does the article state that supermarkets have an adequate/satisfactory amount of
biological/organic/Fairtrade products in stock? V3- What is the tone of the article?
Negative Neutral Positive
E. Which frame or frames are present in the article? More than one answer possible.
Responsibility frame
‘When an issue is framed according to the responsibility frame it is presented in such a way that the responsibility or blame for the cause or the solution is placed on political authorities, individuals or groups’ (Dirikx & Gelders, 2010).
26 V4-Is the responsibility of the supermarket(s) in question towards the environment or
society stated, suggested, or implied?
V5- Is a conflict with established norms or values, suggested, or implied?
V6- Does the article state, suggest, or imply that there is a problem or issue which can or should be resolved by the actions of actors?
V7- Does the article state, suggest, or imply that there is an issue or problem which the actions of an actor have caused?
If the answer to one or more of these questions is yes the article contains the responsibility frame.
Conflict frame
‘Articles that reflect the conflict frame emphasize conflicts between parties/individuals and stress the points of divergence between the opponents’ (Dirikx & Gelders, 2010).
V8- Does the article name inconsistencies in the CSR initiative/approach of a supermarket?
V9- Are two or more actors with different opinions quoted in the article? V10- Are different actors with conflicting viewpoints quoted in the article? V11- Are conflicting viewpoints mentioned in the article?
If the answer to one or more of these questions is yes the article contains the conflict frame Economic consequence frame
‘The economic consequences frame emphasizes the manner in which an issue will economically affect people’ (Dirikx & Gelders, 2010).
V12- Is the economic benefit or disadvantage of a CSR initiative stated in the article?
Q13- Is the current, future, or past economic climate mentioned in the article? Q14- Is the economic position or status of an actor mentioned in the article? If the answer to one or more of these questions is yes the article contains the economic consequence frame.
27 Morality frame
‘The morality frame presents situations from a religious/moral angle. As it is difficult for journalists to give a moral message while adhering to the journalistic norm of objectivity, this is often bypassed by quoting others’ (Semetko and Valkenburg, 2000; Dirikx & Gelders, 2010)
Q15- Do one or more actors in the article label a certain issue or person ‘good’ or ‘bad’ explicitly?
Q16- Are moral implications, complexities, and characteristics of a certain issue are mentioned in the article?
Q17- Are guidelines and/or rules on expected behaviour are mentioned in the article?
If the answer to one or more of these questions is yes the article contains morality frame
Q18- Which actor or actors are present (directly or indirectly quoted or named) in the article?
o Actor(s) affiliated with supermarket
o Actor(s) affiliated with a non-governmental organisation o Scientist(s)
o Journalist(s)
o Actor(s) affiliated with a commercial institution o Supplier(s) of the food industry (for example farmers) o Consumer(s)
o The ‘consumentenbond’ o No external actors are present
28 Appendix II: Intercoder reliability
