Crisis communication and media
:
Does information source (citizen journalism, news media,
organizational communication) and information form (offline
versus online) have an influence on the perceived credibility
of media reporting on crisis communication?
Master’s thesis
Graduate School of Communication Master’s programme Communication Science Corporate Communication track Student name: Roy van der Hoorn Student number: 10423583 Supervisor: Arie den Boon Date: 26th June, 2015
2
Abstract
This study examined the possible effects of different information sources (organizational communication, news media, and citizen journalism) and information form (traditional media versus new media) on the perceived level of credibility of media reporting on crisis communication. In order to answer this, an experiment using an online survey (N = 218) was conducted using information source and form, and brand attitude as the independent variable to measure the effects on perceived source credibility while using the same crisis communication strategy, justification. Results indicate that there are no general proclamations to be made that would indicate a difference between the information source, form, and brand attitude on perceived credibility. There is a small to moderate effect when making the comparison blog versus newspaper article. It was found that blogposts appear to be less credible, which is in line with previous research. Having a positive- or negative brand attitude only has a small effect on perceived source credibility. The type of information does not generally have an influence on the perceived credibility of media reporting.
The results bring a number of theoretical and practical implications regarding crisis communication. This study is one of the first analysing social media using a specific crisis communication strategy. It is a first step in a rather complex framework of different source types, source forms and crisis communication strategies. As this study shows, there is no difference in perceived credibility between traditional media and new media using the justification strategy. Organizations do not have to worry or have doubts whether their message gets across. They can safely use this strategy in order to communicate with the public, as the acceptance of the message is not dependant on the information form.
3
Introduction
The way we receive information regarding a crisis has changed rapidly the past few years. In the far past, the public had to rely on traditional news sources such as newspapers or television news programs to receive their information. With the rise of online and social media as part of daily life for many people (Chung, Nam & Stefanone, 2012; Ellison, Steinfeld & Lampe, 2007), new means of communication have been established. This affects the way communication regarding a crisis is perceived and spread. The effectiveness of crisis communication strategies has been contested the past few years as online and social media increase rapidly in
popularity. The continuing growth of social media results in an increased use of online information sources. A shift in medium use has caused the need for organizations to adapt and while integration is advancing, there are still many
obstacles and uncertainties as with all new phenomena and technologies (Bergemann & Bonatti, 2011; Liu, Austin & Jin, 2011; Utz, Schultz & Glocka, 2013).
Organizations have embraced online and social media as a key communication channel for marketing, PR and crisis communication (Bergemann & Bonatti, 2011; Utz et al., 2013). This is mostly due to the widely availability of the medium and because it allows for rapid exchange of information (Lovejoy, Waters & Saxton, 2012; Page, 2014). It makes it easier for organizations to convey their message to the public directly, without the intervention of news media or opinion leaders
(Bergemann & Bonatti, 2011).
On the other hand, this raises questions on the credibility of these messages (Chung et al., 2012; Johnson & Kaye, 2010; Kim & Johnson, 2009). Since no news media is able to intervene, messages corporations send are not validated by
4 social media also brings in the possibility to spread the word more easily, even for smaller stakeholders. It provides a platform for ordinary citizens, to serve as
journalists and move from content consumers to producers (Chung et al., 2012; Lin, 2014). Blog posts are a good example of this. Regular consumers place content online on their own without the help from traditional media or companies. Today, its reach is for example dependent on how many times an article or item is shared, and might focus less on the source. So for example when an organization posts an online message regarding a crisis that concerns the organization, how does the public perceive this message, as opposed to the same message being sent out from a third party? In other words, is there a difference to be found in the source?
How and why social media is able to affect crisis communication
management is extremely valuable information for professionals, as the credibility of social media is in many areas still uncertain. On top of that, getting a product to stand out in the crowd is increasingly more difficult. Marketing is now more challenging than ever with an increasing population but also an increase in the diversity of products. A cluttering of ads and commercials causes companies to see it as a big concern to differentiate themselves from others (Downey, 2002; Koernig & Boyd, 2009). This research aims to explore and understand in what ways source credibility exists between different types of media, how it can be affected, and what the
influence is of brand attitude. This can be formulated into the following research question:
RQ: Does information source (citizen journalism, news media, organizational
communication) and information form (offline versus online) have an influence on the perceived credibility of media reporting on crisis communication?
5
Theoretical background
This chapter includes describing the validity and necessity for this research, as well as aiming to provide a solid background of the different concepts. These concepts are divided into independent and dependent variables, and defining them is necessary to understand their role in this research. Each of them will be addressed individually first, then a connection will be made in order to form the different hypothesis. Finally, the conceptual model is presented to help visualize the process.
Information source & information form
Information sources can be described as inputs for a production process (Ratchford, Talukdar & Lee, 2001). The need for information has been in our system for generations, dating back to cavemen who wrote symbols on walls as a way of communication and provide information for others. Between then and now, evolution led to the discovery of many new technologies that help fulfill our needs for
information. The transferal of information went from word-of-mouth to more general platforms reaching a wider audience, also known as the era of mass communication. Information was made available for the masses with newspapers only to reach an even bigger audience with television. The information in mass media became the only contact many consumers have with a diverse array of information (McCombs, 1972). Sending out messages was solely reserved for big news organizations, large corporations that do advertising, and governments. Mass media was used to influence public opinion, as most issues and concerns that attracts our interest are not
amendable to direct and personal experiences (McCombs, 2013; Murch, 1971). This left the general thought that the information being sent out was credible, valuable,
6 and timely information with few arguments to doubt the source (Heath, Liao, & Douglas, 1995).
The past decades, digitalization took place and information availability increased greatly. With the upcoming of the internet, consumers can place content more easily and get a message across without the need for large amounts of funds or big corporations (González-Herrero, & Smith, 2008). The internet has become a major source of information, following the digitalization of the world. (Ratchford, et al., 2001; Westerman, Spence, & van der Heide, 2014). Each type of source has its own characteristics, advantages and disadvantages. Each source can present valuable information for users, but each operate on a very different level. For certain sensitive topics, official news distribution is (deliberately) slowed down, while the internet and social media possibilities make it possible that the information is more swiftly
distributed (Westerman et al., 2014). For the basis of this research, five different sources are defined. The oldest medium newspaper articles was take into account first. Newspapers have a long established history. Next to newspaper articles written by journalists, another way of getting information is straight from organizations, through the use of press releases. Besides the traditional media as an information source, online-, and in specific, social media has increased possibilities for how people are able to send and receive information (Fox, 2011; Pepitone, 2010; Westerman et al., 2014). Two well-known examples of social media are Facebook, and blog posts. Facebook for its popularity worldwide, and blog posts because of the possibility to easily setup an own personal page and create and distribute content for the world to see. Social media have become part of everyday life for many, and it is making for a very important and useful communication channel during
7 organizational crises (Ellison, Steinfeld, & Lampe, 2007; Utz, Schults, & Glocka, 2013). The focus is text-based only, so comparisons can be made more easily.
Not only is there a distinctness to be found between different types of information sources, one can also distinguish different forms of information. As Willemsen, Neijens and Bronner (2012) describe, a difference can be seen in online media due to the fact that it has the power for user-generated content to be made. People that were first solely known to be an ‘audience’, suddenly have the possibility to write content themselves and participate more interactively (Rosen, 2006;
Willemsen et al., 2012). Thus, for this first step in explaining the different concepts, information sources are divided into two forms: online- and offline media. While research is extensive on both traditional- and online media in terms of perceived credibility and brand attitude, research that considers both types of media is scarce.
Getting back to the introduction, each type of information source is a potential candidate for organizations to communicate through in case of a crisis. A separation between online- and offline was made. Now that the different types of media have been elaborated, a clear definition of crisis communication must be given.
Crisis communication
The definition of an organizational crisis can be best described as an unpredictable event that results into a conflict between an organization and other stakeholders, having an impact on the performance of the organization and generate negative outcomes (Coombs, 2012). For organizations it is imperative that they are able to counter these negative outcomes and improve organizational performance,
8 hence crisis management has an important role to maintain the organization’s
sustainability.
Crisis management is even more important than ever due to stakeholder activism and advances in communication technologies such as the continued popularity of social media and microblogging sites (Austin, Liu & Jin, 2012; Coombs, 2012; Lenhart, Purcell, Smith & Zickuhr, 2010). Classical crisis
communication theories neglect the role of the new online medium as they focus more on the traditional crisis type versus communication strategy (Utz et al., 2013). Social media and the internet in general are increasingly being seen as the most reliable source for news, and its popularity is mostly due to the direct availability of the medium and its high level of interactivity (Seltzer & Mitrook, 2007; Taylor & Perry, 2005). Social media play a fundamental role when it comes to the negotiation and dynamics of crises (Utz et al., 2013).
A fair amount of research on crisis communication effects has been done (Coombs, 2008; Jin & Liu, 2010; Lerbinger, 2007; Yang, Kang & Johnson, 2010), though there is still much to learn on this concept. This is primarily because research on incorporating online media is just emerging (Austin et. al, 2012) and thus more in-depth or longitudinal research has not yet been formed. There are some fairly new models such as the Social-Mediated Crisis Communication model (SMCC)
developed by Jin and Liu (2010), although such designs have not yet proven their full potential use in the field (Liu, Austin & Jin, 2011). First of all, the experiment using this model included only one small target group, college students, in a follow-up of university-related communication crisis situations. Additional experiments with a greater diversity of publics have not yet been conducted. The model has been tested
9 during the crisis response phase, but not pre-crisis or in the stage of recovery of the organization.
Crisis communication is an essential form of communication for a lot of companies, as it requires to adapt a part of the organizational strategy to fit to an actual event. According to Coombs (2007), one can see three major types of crises. The first being the accidental crisis, in which the action an organization takes leads to a crisis, but this was unforeseen. The second crisis described by Coombs (2007) is the victim crisis, in which the organization itself is (also) a victim of the crisis and is not to be seen as directly responsible for the past events leading to a crisis. The third and final mentioned crisis type is the preventable crisis, in which the organization’s faulty action lead to a crisis. This could be due to unlawful proceedings such as ignoring the moral code or not abiding the law.
Crisis communication strategies
There has been a lot of research on the different stakeholders that could be involved in a crisis, and in which different ways an organization can respond to a crisis (Barton, 1993; Egelhoff & Sen, 1992). Coombs (1995) has focused a lot more in the ways organizations communicate during a crisis, by researching the different techniques to respond and calls it crisis communication response strategies. Coombs (2007) has continued this concept of crisis communication in a model that offers an evidence-based foundation for all sorts of crisis communication related issues. This theory is called the Situational Crisis Communication Theory (SCCT). As the name suggests, it gives pr professionals and communication specialists several ways in which they can react to a crisis, and in what way they can reach the relevant stakeholders best. The SCCT provides an explanation for how key factors of a
10 situation can influence characteristics of the crisis and the reputation. Ultimately, it provides an understanding of how crisis managers can prevent their organization’s reputation from being destroyed. The theory relies on the fact that stakeholders will try and find reasons (attributes) for the crisis to have happened (Coombs, 2007). This is especially the case for events that are negative by nature or responsibility lies with the organization (Weiner, 1985). A person could for instance experience anger or sympathy towards the organization, the core emotions of the Attribution Theory, which is the work of Heider (1958), Kelley and Michela (1980) and Weiner (1985). These researchers have defined that attributes are individual’s explanations for the causes of success and failure.
Coombs recognizes two types of response strategies: primary- and secondary strategies. The primary being deny and rebuild. Denying the crisis, can be directly attacking another involved party, shifting the blame. It can also be denying when the organization comes up with an excuse. Rebuild means compensating for the crisis, or apologizing for it and thus taking responsibility. Secondary strategies consists of reminding of how well the organization did in the past, or using flattering, in which the organization tries to persuade others by stating that the organization’s intentions were good. Benoit (1997) also developed a theory for crisis response strategies, and while in general the different strategies are linked, they are named differently. For example, reducing strategies matches diminish of Coombs (1997). Apology is labeled as corrective action by Benoit, while Coombs uses the term compensation.
The justification strategy is an interesting choice as a crisis communication style for organizations, as they aim to set the records straight in their communication messages. These types of messages respond to a crisis in a way that the organization tries to be justifiable in their actions. But does this strategy work effectively no
11 matter the source of the message? Consumers are inclined to seek and accept
information as genuine when they believe the source to be true. Whether a message containing the justification-strategy is successful or not, depends on how the message is received by the audience. While previous research mainly does research of
different crisis strategies on a macro-level (Benoit, 1997; Coombs, 1995), none focus on a specific crisis communication strategy on the micro-level. Thus, this research is based on a sole response strategy, the justification method. Before the first
hypothesis can be formulated, it is imperative to get a hold of the concept source
credibility. The aim of using a communication strategy is to improve overall image
and reputation of an organization. This can only be achieved if the message is seen is credible. All strategies from Coombs are to be found in the table 2 in the Appendix A. The table of Benoit is listed in Appendix A, table 3.
Credibility
How and why social media are able to affect successful crisis communication management is extremely valuable information for professionals, as the credibility of social media as an information source is in many ways still uncertain (Chung et al., 2012). Johnson and Kaye (2010) argue for example that web-based information often does not reach sufficient levels of factual substantiation or analysis, because there is a lower pressure-level for those that are able to post content, no matter the validity. Its information lead critics to argue against the credibility of the internet because of its easy access, and lack of standards for publishing content online (Andie, 1997; Newhagen & Levy, 1997). Information might be biased and this takes hybrid news sources such as blogs and other user-generated content into question (Cassidy, 2007). However, there are a lot of contrasting claims on this concept. For example, Utz
12 (2013) claim that traditional media are seen as more credible, while other researches display the opposite, positing that publics in certain cases actually assign a higher level of credibility to social media coverage than to traditional mass media (Kim & Johnson, 2009; Procopio & Procopio, 2007; Sweetser & Metzgar, 2007). Contrasting research on this matter make it to be an interesting phenomena to research further in-depth.
Source credibility is a form of attitude towards the message source (Gunther, 1992). Whether a message is accepted by its receiver depends partially on the characteristics of the sender. The term source credibility is an already established theory that elaborates on how communication effects can be dependent on the perceived credibility of the source (Sherif, Sherif & Nebergall, 1965; Rhine & Severance, 1970; Gunther, 1992). It can be defined as judgments made by a perceiver concerning the believability of a communicator (O’Keefe, 1990). Source can have different interpretations, depending on the perspective in which it is used. Sundar and Nass (2001) distinguish two types of interpretations of the definition of source. It could either be ontological of psychological, where the ontological approach means that a source is defined by its function or what it does. More specifically, scholars have made a distinction between an internal source or the communicator originating the message and an external source or the medium transmitting the message. Academic studies on this topic emerged during the World War 2, when propaganda was used by many countries, including the United States, to try and influence public opinion to increase support for war and raise morale among citizens. Building on the traditional communication theories such as the send-message-receive theory, Hovland (1951) continued to build on existing communication models and researched the influence of source credibility on
13 communication effectiveness. It is known that the credibility of all types of
communication relies heavily on the perceived credibility of the source of the message (Lowry, 2013).
Measuring source credibility has been done in a lot of ways in the past, using all sorts of definitions and sub-dimensions. Some scholars view message source credibility as a construct consisting of three sub-dimensions (Paul & Wang, 2011), while others add more dimensions to it (Willemsen, Neijens & Bronner, 2012). Paul and Wang (2011) believe source credibility to be constructed from trustworthiness, expertness, and attractiveness. Willemsen, Neijens and Bronner (2012) show that source credibility can consist of even more sub-dimensions, such as the level of knowledge, honesty, and sincerity. Over the past years, source credibility was measured and scales kept on improving. For this research, an advanced and adapted scale was used based on these past versions. While source credibility is researched extensively (Paul & Wang, 2011; Sunder & Nass, 2001; Westerman et al., 2014; Willemsen et al., 2012), most scholars base results focusing on a single medium or information type. This research will add to that, using multiple sources, in both online- and offline media.
Brand attitude
In addition to researching credibility depending on the different sources and forms, each consumer has some pre-existing attitude towards a brand they are familiar with or have an actual experience with. As the diversity as well as the number of products keep on increasing, it is important for organizations to stand out in the crowd. In order for companies to sell their products, a likeable brand has to be created that customers are loyal to. Organizations have focused much on branding
14 itself the past years. Companies even use in-store experiences for their customers in order to differentiate themselves from others (Pine & Gilmore, 1998). However, what effect does brand attitude have when it is already at a certain level, on the perceived credibility of a message? A positive brand attitude could work as a buffer when exposed to bad publicity. For example, an organization may face a negative issue that is spread through different media. When an organization responds, does it matter if the consumer who reads the crisis communication already has a positive, or negative attitude towards the brand? In the next section, all of the concepts combined together will be displayed and hypotheses will be formulated.
Construct of concepts
Now, with all the separate concepts, it is time to put them all together and show their relations. The evolution in mass communication makes it possible for consumers to get news information in countless of ways, through many types of information sources. Each source has its own characteristics, advantages and disadvantages. Organizations are constantly on the look-out for ways to get their message across, and to reach the audience. In case of promotion, but also in the situation of a crisis. Crisis communication in the far past was manageable more easy for organizations as they would often only have to send out one message using one type of information source (newspapers). Now, with online media and the
construction of user-generated content becoming more mainstream, it is a lot more difficult for them to reach the masses the way organizations want to. When a crisis appears, the organization having to deal with the crisis often makes use of
justification-strategy. This is where credibility comes in. Offline media in mass communication was up to many years ago the only way of reaching the audience and
15 credibility of the source was almost never doubted (Westerman et al., 2014).
However, with the increased usage of online and social media, the number of sources increase and there are more ways of getting information, leading to questions
regarding credibility of the source. This leads to the very first hypothesis to be tested in this research:
H1: Offline media has a higher level of source credibility than online media. Not only is there a possible difference to be found in traditional and new media, but all sorts of information sources altogether also could have an effect on the perceived credibility of the receiver. Are direct messages, in the form of a press release, perceived to be less credible as a source than newspaper articles using the same justification strategy? To answer this, the following hypothesis was formulated:
H2: Direct messages from an organization are perceived to be less credible than
news media messages while containing the same justification crisis strategy.
While blog posts have been around for some years now, they are more used as a tool for citizen journalism lately. A good example of this is recent issues between freedom fighters and dictators in countries such as Egypt, Libya and Iran. Seeing as some major social media sites are being blocked when large parts of the population are dissatisfied, people turn to blogs as a way of communicating with the outside world, reporting on the events happening nearby. This makes for an
interesting view of perspective, as blogs are a very quick way of communication and no news media, or even government is able to intervene since blog posts are not posted all on the same web domain. This raises the question if there is a difference to be found between these two sources. It is formulated into the following
16
H3: A news media message is perceived to be more credible than a blog post while
containing the same justification crisis strategy.
As discussed, brand attitude as a pre-selected point of view towards a
company, could lead to differences in perceived credibility. It would make sense that a crisis strategy is more effective, if the brand is already liked by a consumer before the actual crisis event takes place. To analyse this, the final hypothesis has been established as follows:
H4: Consumers with a prior positive attitude towards a brand perceive a higher level
of credibility towards a source than consumers with a prior neutral- or negative brand attitude.
17
Method
To measure the different variables, an online experiment using a survey is conducted. The experiment was chosen because it can reveal causal relations and has the practical advantage that it can process a large number of respondents in a
relatively small amount of time (Boeije, ‘t Hart & Hox, 2009). Its financial costs are kept to a minimum and it can be spread relatively easy (Vaske, 2011).
Sample
For this experiment, a convenience sample was used. The survey was created using questionnaire tool Qualtrics and published using Facebook reaching friends, co-students, co-workers and other relatives. The only criteria for respondents was to meet the age requirement of 18, and to know the fast-food company McDonalds. All respondents met these criteria (N = 241). Several respondents used the sharing-option and shared it across their network as well. A total number of 241 people participated in the survey of which 23 did not finish the survey, leaving 218 fully completed results to be taken into further account. Respondents ranged in age from 18 up to 65 and their average age was 32.44 years (SD = 12.49). Of the respondents, 72.9% (N = 159) was female, leaving the remaining 27.1% (N = 59) to be male. Finally,
educational level was measured. The majority of the respondents was higher
educated (hbo or higher). 29.8% of the respondents finished or is currently following a hbo degree, 26.1% finished or is currently attending a wo (academic) degree, and 27.1% has or attends to an mbo degree. The remainder of 17% had an degree in or attended high school or vmbo. The complete socio-demographic info can be found in Appendix D, table 4.
18
Design
Different conditions have been used to help simulate the different situations of exposure. The experiment had a 2 (online versus offline) x 3 (organisational communication, news organisation message, citizen journalism) between-participants design. Respondents were randomly assigned to one of five different conditions, creating an estimated 20% chance to be placed in a certain condition. The conditions were divided as follows: Press release ( N = 40), news article nu.nl (N = 41), blog post (N = 47), news article De Telegraaf (N = 46), Facebook-page article (N = 44). Two news items were created to display the difference between online (nu.nl) and offline (Telegraaf) media. For citizen journalism, a fictive blog post was created. Finally, for the organizational communication, a press release was created. See table 1 directly below for an in-depth schematic view of the experimental design.
Table 1: Experimental design
T1 M T2
O1 R - Condition 1 (press release) O2
O1
O1
O1
O1
R - Condition 2 (news article nu.nl) R - Condition 3 (blog post)
R - Condition 4 (news article De Telegraaf) R - Condition 5 (Facebook page)
O2 O2 O2 O2 IV DV T1-2 = Timeslot M = Condition (Randomized, R) X = Intervention (manipulation) O1-2 = Measuring moment IV = Independent variable DV = Dependent variable
19
Measuring variables
Independent variables
The independent variable information source was tested by exposure to either one of the following manipulated conditions: citizen-journalism article, newspaper
article, or press release from an organization. Added to that, these are again
categorized into a secondary variable; offline versus online media. To check and see if differences in effects can be measured between old media (physical newspapers) and new media (online news), both types were included in the online experiment.
Attitude towards the brand was also measured before applying the stimulus.
Dependent variables
To measure source credibility, different cases were studied where credibility was measured using different scales. For this specific research, an adapted version of Meyer’s Newspaper Credibility Index (1988) and five-point bipolar items from the research by Willemsen, Neijens and Bronner (2012) were used as they were found to have the strongest reliability. These measure trustworthiness (e.g. dishonest-honest, insincere-sincere), α = 0,91; and perceived expertise (e.g. inexperienced-experienced, unknowledgeable-knowledgeable), α = 0.94. These items were to some extent also used in Gaziano and McGrath’s (1986) research, displaying these items are well-represented in communication science. For this research, a total of five items were selected to measure credibility. For example, respondents were asked to give their opinion on the article they were exposed to by answering questions such as: “To
what extent do you agree to the article being sincere”, and “To what extent do you believe the article to be trustworthy”. Answer possibilities were categorized to a
20 To analyse if the adapted scale also worked for this research, a principal
component analysis (PCA) with varimax rotation was conducted. A one-factor solution emerged, accounting for 71.7% of the variance. This can be seen as one component has an Eigenvalue higher than 1 (3.587) and the scree plot indicating an inclination after the first component. This makes credibility a suitable scale to be constructed from the single items that measure credibility. Then, a reliability test was conducted in order to check if the to be constructed scale is actually a reliable way of measuring credibility. Cronbach’s alpha was measured for the five different items. The results show that Cronbach’s alpha for credibility scale was 0.90, which can be described as a very reliable scale (Appendix D, table 5). The scale would not increase in reliability any further if an item would be deleted from it.
Brand attitude was measured using a 7-point Likert scale, which was then categorized into negative, neutral, or positive in order to measure the appropriate hypothesis and help interpreting results easier.
Procedure
An online questionnaire was presented to the public and informed consent was used before the start of the survey to ensure the understanding between the respondent and the researcher. As stated before, respondents were randomly assigned to one of five different conditions. The survey took an average of five minutes to complete and consisted of 19 different questions, including questions regarding the credibility of the article they were exposed to, their opinion on McDonalds before and after exposure to the manipulation, as well as general questions such as age, sex, and educational level. The Dutch language was used during the entire survey, as it is the native language of both the researcher and the respondents that were targeted,
21 making it easier to find the suitable participants without having the concerns of items being misread due to language barriers. The complete survey copy can be found in Appendix B. The stimuli materials can be found in Appendix C.
Manipulation
The messages were each created trying to maintain a neutral perspective, without the use of positive- or negative framing. The subject and content in each source article was practically the same, so that differences in perceived credibility could be attributed specifically to the source, rather than differences in between messages. Research by Jones et al. (2003) suggests that sources that are not credible cause people to fail to engage with the contents of the message, without regard for the message being framed positively or negatively.
To initiate a possible difference between online and offline sources, the decision was made to add one original newspaper to the questionnaire. Due to practical limitations, all of the questions had to be asked online through Qualtrics. One of the conditions however, was manipulated to look like an offline, newspaper version article. Respondents were told that the newspaper article from the Dutch newspaper ‘De Telegraaf’ was scanned for the purpose of this questionnaire.
As the manipulative story, fast-food company McDonalds was chosen as the topic in each of the different sources. At the start of the survey, respondents were asked if they knew this company to confirm whether they have a pre-selected opinion on the firm. All of the respondents knew the company (i.e. by name at least). Then, the general opinion of the firm was measured, before the manipulation and after, to check for any differences in effect after being exposed to one of five conditions. The story of the message was that an employee of McDonalds was unsatisfied with her working condition and this led to an incident on the work floor. She misbehaved,
22 provoked her manager and then physically damaged one of the McDonalds
restaurants. The story itself contained the justification strategy from McDonalds, claiming that the company has taken appropriate action and that the employee was fired for her misconduct. Then, the company stated that each employee gets paid at least the minimum salary that is in accordance with the law. The same content is to be found in each condition, although the overall style is edited to fit the source. For instance, the blog post will have a more informal style of writing to match its
medium. After the manipulation, respondents were asked if they thought the message was credible or not, by answering five five-point items measuring credibility.
23
Results
In this part, the results will be displayed to answer the hypotheses. The statistical measuring tool SPSS version 21 was used to analyse the data provided by the Qualtrics survey. In order to test the different hypotheses, independensamples t-tests and a one-way Analysis of Variance (ANOVA) was conducted. Beforehand, each analysis had to be tested for equality of variances, by using Levene’s F test. For the one-way ANOVA, the post-hoc test Bonferroni was chosen to check for exact differences in between groups.
Testing hypotheses
H1: Offline media has a higher level of source credibility than online media.
The first hypothesis posited that offline media has a higher perceived source credibility than online media. To test this, an independent-sample t-test was
conducted to compare the level of perceived credibility in both offline- and online media. Analysis was done using a 95% confidence interval. First, Levene’s Test for Equality of Variances was conducted. Results displayed that equal variances can be assumed (F = 2.88, p = .091). There was a difference in the scores for offline media (M = 2.33 , SD = 0.93) and online media (M = 2.14, SD = 0.83 ) in terms of
perceived credibility. However, this result is not significant, t (216) = -1.37, p = 0.174. Offline and online media do not differ in terms of perceived credibility. Thus, this first hypothesis is rejected. See table 7 in Appendix D for the SPSS output.
H2: Direct messages from an organization are perceived to be less credible than
24 The second hypothesis stated that direct organizational communication is perceived to be less credible than news media messages. To analyze this, another independent-sample t-test was conducted. Again, Levene’s F test for Equality of Variances was conducted and shows that equal variances can be assumed (F = 2.95,
p = 0.088). There was a difference between the perceived credibility of press releases
(M = 2.28, SD = 0.784) and the perceived credibility of news media messages (M = 2.34, SD = 0.93). However, this result is also not significant, t (124) = 0.367, p = 0.714. There is no difference in perceived credibility when exposed to either a press release or a news article. Messages that are sent out directly by companies and delivered to the consumer, are not perceived as less credible than news articles created by journalists that contain the same crisis communication strategy. The second hypothesis is rejected. See table 8 in the Appendix for the SPSS output.
H3: A news media message is perceived to be more credible than a blog post
while containing the same justification crisis strategy.
The third hypothesis posed that blog posts with news information (citizen journalism) are perceived to be more credible than news media messages published by journalists. This was again tested using an independent-sample t-test. Levene’s F showed that equal variances could not be assumed (F = 6.09, p = 0.015). There was a statistical significant difference in the scores for news media (M = 2.34, SD = 0.93) and the blog (M = 1.94, SD = 0.76), t (114) = 2.70, p = 0.008, 95% CI [0.11, 0.69]. A news article is seen as more credible than a blog post containing the same
justification strategy. As there is a difference to be noticed, the null hypothesis is rejected. Further, the effect size is calculated with Cohen’s effect size value (d = 0.47), suggesting a small to moderate practical significance. This hypothesis is accepted. See table 9 in the Appendix for the SPSS output.
25
H4: Consumers with a prior positive attitude towards a brand perceive a higher
level of credibility towards a source than consumers with a prior neutral- or negative brand attitude.
The fourth hypothesis posed that consumers with a positive attitude towards a brand before exposure to the stimulus, have a higher level of perceived credibility towards a source containing a justification strategy than those who possess a neutral- or negative brand attitude. In order to measure this, a one-way ANOVA was
conducted to check and see if there are any differences between these groups. The majority of the sample had a positive brand attitude towards McDonalds (Appendix D: Table 6). A check for homogeneity of variances was done to see if equal
variances may be assumed. Finally, a Post-hoc test was added to the test to see which specific groups differed from each other. Levene’s F for homogeneity of variances showed an insignificant result (F (2, 215) = 5.16, p = 0.276), meaning that equal variances can be assumed for this analysis. There was a significant difference between groups as determined by the one-way ANOVA, F (2, 215) = 5.16, p = 0.006. The size of the effect is however very small (η2 = 0.05). Still, because the test was significant, a post-hoc test is conducted using Bonferroni’s correction.
Consumers who are positive regarding McDonalds have a higher level of perceived credibility (M = 3.02, SD = 0.91) than consumers that are neutral, (M = 2.74, SD = 0.73) or negative, (M = 2.58, SD = 0.89). Comparisons using this test indicated that the mean score for positive consumers’ perceived credibility was significantly different than the mean score for negative consumers’ perceived credibility,
(Mdifference = 0.44, p = 0.007). There was no significant difference between those who
are negative, and those who are neutral regarding McDonalds (Mdifference = -0.16, p =
26
Conclusion & Discussion
In this final chapter, the results will be interpreted and the main research question will be answered. After that, the answers will be discussed to relate back to the theory as well as their possible practical implications. The chapter ends with limitations of this study and eventual suggestions for future research will be posed.
Does information source (citizen journalism, news media, organizational communication) and information form (offline versus online) have an influence on the perceived credibility of media reporting on crisis communication? While
predictions were clear, the results did not show each expected effect. First off, offline media does not have a higher perceived credibility than online media. Consumers do not see traditional media as more credible than new media. This is contrasting findings of some scholars such as Kim and Johnson (2009), Procopio & Procopio (2007), and Sweetser & Metzgar (2007). This study shows that when the message is constant but the medium is different, responses in terms of perceived credibility are not changed. This is important for example in organizations which would question in which medium their message would be better understood and accepted. Credibility does not seem dependent on the difference between offline and online media.
Secondly, it was questioned if direct messages from organizations could be seen as less credible as opposed to news media items when communicating the justification crisis strategy. Again, this difference is not supported in these findings. This shows that it does not matter whether the message is presented as a journalistic article or as a press release. Consumers do not seem to concerned about possible manipulations of newspaper articles any more than press releases. Relating back to the literature review, Schultz, Utz and Göritz (2011) also found positive and negative results when it comes to perceptions and credibility of online media, but they did not
27 focus on traditional offline media such as newspapers. They argue that it could have something to do with the specific attitude towards a brand. This study again has a very strong brand used in the stimuli, which makes the brand attitude very high. The majority of the respondents find McDonalds a very likable brand and perceive it as mainly positive. This high reputation can create a halo effect, which protects the organization even when negative publicity is involved (Coombs & Holladay, 2006).
Thirdly, there is a small to moderate effect to be found when comparing blog versus news media. In this study it was found that blog posts appear to be less credible. This result builds on previous theories by Schultz (2011) that sources that have a more established connection with the receiver, tend to be perceived as more credible than unfamiliar sources. In this study, a fictive blog was chosen so that none of the participants would have a pre-selected attitude towards the blog. This confirms previous research where unfamiliar senders are more critically received.
Finally, while significant, the effect on credibility by having a positive- or negative brand attitude is too small to generalize without the consideration for other possible factors.
Going back to the original research question, information source does not generally have an influence on the perceived credibility of media reporting. A justification strategy’s effectiveness does not appear to have any mentionable significant effects on the level of credibility dependent on the information source or form. There are some nuances to be found as is the case with most proclamations, but perceived credibility in communication messages remains equal no matter the
28
Practical implications and future research
The results bring a number of theoretical and practical implications regarding crisis communication. This study is one of the first analysing social media using a specific crisis communication strategy. It is a first step in a rather complex
framework of different source types, source forms and crisis communication strategies. As this result shows, there is no difference in perceived credibility between traditional media and new media using the justification strategy.
Organizations do not have to worry or have doubts whether their message gets across when using this strategy in order to communicate with the public, as the acceptance of the message is not dependant on the information form. Secondly, there is no dissimilarity between press releases and news articles in terms of perceived credibility. When the message is the same, organizations can either be the source themselves, or let the media send out the message to reach the audience. This implicates that organizations do not need to fear that credibility will decline once a message is mediated through a different source.
In terms of theoretical implications, this study shows a very reliable scale that measures perceived source credibility, which is useable on both traditional, and online media.
There are a few limitations that can be observed during this research. A manipulation check was added to see if participants could see for themselves what kind of article they were exposed to. This was done by adding the question: “What
type of item did you just read?”, where respondents had to choose between the five
different conditions. While manipulation checks succeeded, it is notable to share that not everyone participating the survey answered this question correctly. Some
29 some viewed the press release page from McDonalds to be a news article as well. This could have minor influences on the results. For future research on this topic, it might be a good idea to actually double-check or make pages more detailed, or let them visit a (fake) web page themselves in order to diminish this issue.
Another limitation to be addressed is the target group. Due to time
limitations, a convenience sample was used using the snowball-technique in which the questionnaire was sent out to Facebook friends, co-students and co-workers, and was shared several times so the network reached further. This could cause for the results to contain relativity more co-students that study communication science as well. Also, the average age being around 32, adults were targeted but no specific age group was selected and/or categorized into different groups. In future research, different age groups could be categorized to check and see if differences occur in outcomes. It would make sense if teenagers take information from Facebook or blog posts more seriously, than adults or seniors would. On the other hand, adults or elderly might see offline media as a more credible media platform than online media.
As said in the discussion section of this study, a possible halo effect can be seen due to the fact that the brand chosen for the study was a very likeable brand (Appendix D: Table 6). This could cause an automated defence mechanism that defends the brand in case of negative exposure. In turn, it could deflect any possible influences on perceived credibility. When a brand is liked, the user is likely to be less critical about the content itself, resulting in a relatively good level of credibility regardless of the source or form the message appeared in. It would be recommended for future studies to make a discrepancy between brands based on their reputational status. Also, measurements of crisis-communication strategies only focused on the justification strategy. It seems likely that when for example Coombs’ (1997)
30 deniability strategy is used, credibility will be questioned faster. Further research could be useful by taking the present micro-level study towards a more macro-level and take into account several other crisis communication strategies.
Finally, the information sources were concentrated on text-based items only. It is seemingly more difficult to expand this research but it could add value when comparing text-based sources to more visual sources, such as pictures or even video’s. The characteristics of said source types might have an influence that could build upon this study.
31
References
Andy, T. (1997). ‘Why web warriors might worry’, Columbia Journalism Review,
36(2). 35-39.
Austin, A., Fisher Liu, B. & Jin, Y. (2012) How audiences seek out crisis information: Exploring the social-mediated crisis communication model. Journal of
Applied Communication Research, 40(2), 188-207. doi: 10.1080/00909882.2012.654498
Barton, L. (1993). Choosing the right words: The development of guidelines for the selection of the “Appropriate” crisis-response strategies. Management
Communication Quarterly, 8(4), 447-476. doi: 10.1177/0893318995008004003
Benoit, W. (1997). Image repair discourse and crisis communication. Public
Relations Review, 23(2), 177-186. doi:10.1016/S0363-8111(97)90023-0
Bergemann, D. & Bonatti, A. (2011). Targeting in advertising markets: implications for offline versus online media. RAND Journal of Economics, 42(3), 417-443. doi: 10.1111/j.1756-2171.2011.00143.x
Boeije, H., ‘t Hart, H. & Hox, J. (2009). Onderzoeksmethoden. Den Haag: Boom Onderwijs.
Coombs, W. (2012). Ongoing crisis communication: Planning, managing, and responding (3rd ed.). Thousand Oaks, CA: Sage
Chung, J., Nam, Y. & Stefanone, M. (2011). Exploring online news credibility: The relative influence of traditional and technological factors. Journal of
Computer-Mediated Communication, 17(2), 171-186. doi: 10.1111/j.1083-6101.2011.01565.x
Coombs, W. (1995). Choosing the right word. The development of guidelines for the selection of the ‘appropriate’ crisis response strategies. Management Communication
Quaterly, 8(1), 447–476.
Coombs, W., & Holladay, S. (2006). Unpacking the halo effect: Reputation and crisis management. Journal of Communication Management, 10(1), 123–137
Coombs, W. (2007). Protecting organization reputations during a crisis: The development and application of Situational Crisis Communication Theory.
32 Coombs, W. (2009). Ongoing Crisis Communication: Planning, managing and
responding. LA: Sage.
Egelhoff, W. & Sen, F. (1992). An information-processing model of crisis management. Management Communication Quarterly, 5(4), 443-484. doi: 10.1177/0893318992005004003
Ellison, N., Steinfield, C., & Lampe, C. (2007). The benefits of Facebook “friends”: Social capital and college students’ use of online social network sites. Journal of
Computer-Mediated Communication, 12(1), 1143–1168.
Fox, S. (May 12, 2011). The social life of health information, 2011. Pew Research Center’s Internet & American Life Project: Washington, D.C. Retrieved April 20, 2015 from
http://www.pewinternet.org/∼/media//Files/Reports/2011/PIP_Social_Life_of_Healt h_Info.pdf
Gaziano, C. & McGrath, K. (1986). Measuring the concept of credibility. Journalism
Quarterly, 63(3), 451-462. doi: 10.1177/107769908606300301
González-Herrero, A., & Smith, S. (2008). Crisis communications management on the web: How Internet-based technologies are changing the way public relations professionals handle business crises. Journal of Contingencies and Crisis
Management, 16(1), 143–153.
Gunther, A. (1992). Biased press or biased public? Attitudes towards media coverage of social groups. Public Opinion Quarterly, 56(2), 147-167. doi: 10.1086/269308 Heath, R., Liao, S. & Douglas, W. (1995). Effects of perceived economic harms and benefits on issue involvement, information use and action: A study in risk communication. Journal of Public Relations Research, 7(2), 89-109. doi:10.1207/s1532754xjprr0702_01
Heider, F. (1958). The psychology of interpersonal relations. NY: Wiley.
Jin, Y. & Liu, B. (2010). The blog-mediated crisis communication model: Recommendations for responding to influential external blogs. Journal of Public
33 Johnson, T. & Kaye, B. (2010a). Choosing is believing? How web gratifications and reliance affect internet credibility among politically interested users. Atlantic Journal
of Communication, 18(1), 1-21. doi:10.1080/15456870903340431
Kelley, H. & Michela, J.(1980). Attribution theory and research. Annual Review
Psychology, 31(1) , 457–501. doi: 10.1146/annurev.ps.31.020180.002325
Koenig, S. & Boyd, T. (2009). To catch a tiger or let him go: The match-up effect and athlete endorses for sport and non-sports brands. Sport Marketing Quarterly,
18(1), 25-37.
Kim, D. & Johnson, T. (2009). A shift in media credibility: Comparing internet and traditional news sources in South Korea. International Communication Gazette,
71(4), 283-302. doi:10.1177/1748048509102182
Lenhart, A., Purcell, K., Smith, A. & Zickuhr, K. (2010). Social media & mobile internet use among teens and young adults. PewResearchCenter reports. Retrieved on May 25th 2015, http://www.pewinternet.org/.
Lerbinger, O. (1997). The crisis manager: facing risk and responsibility. Mahwah, NJ: Lawrence Erlbaum.
Lin, J. (2014). The effects of gratifications on intention to read citizen journalism news: The mediating effect of attitude. Computers in Human Behavior, 36(1), 129-137. doi: http://dx.doi.org/10.1016/j.chb.2014.03.054
Lowry, P., Wilson, D. & Haig, B. (2013). A picture is worth a thousand words: Source credibility theory applied to logo and website design for heightened credibility and consumer trust. International Journal of Human-Computer
Interaction, 30(1), 63-93. doi: 10.1080/10447318.2013.839899
McComas, K. & Trumbo, C. (2001). Source credibility in environmental health - Risk controversies: Application of Meyer’s Credibility Index. Risk Analysis, 21(3), 467-480.
Meyer, P. (1988). Defining and measuring credibility of newspapers: Developing an index. Journalism Quarterly, 65(3), 567-588. doi: 10.1177/107769908806500301 Murch, A. (1971). Public concern for environmental pollution. Public Opinion
34 Newhagen, J. & M. Levy (1997). ‘The Future of Journalism in a Distributed Communication Architecture’. The Electronic Grapevine: Rumor, Reputation, and
Reporting in the New On-Line Environment. Mahwah, NJ: Lawrence Erlbaum.
O’Keefe, D. (1990). Persuasion: Theory and research. Newbury Park, CA: Sage. Paul, C. & Wang, Y-C. (2011). The influences of electronic word-of-mouth message appeal and message source credibility on brand attitude. Asia Pasific Journal of Marketing and Logistics, 23(4), 448-482. doi: 10.1108/13555851111165020
Pearson, C. & Clair, J. (1998). Reframing crisis management. The Academy of
Management Review, 23(1), 59-76. doi: 10.5465/AMR.1998.192960
Pepitone, J. (March 10, 2010). Twitter users not so social after all. CNNMoney.com. Retrieved March 14, 2010 from
http://money.cnn.com/2010/03/10/technology/twitter_users_active/index.htm?hpt=M id
Pine, B. & Gilmore, J. (1998). Welcome to the experience economy. Harvard
Business Review, 76(4), 97-105.
Procopio, C. & Procopio, S. (2007). Do you know what it means to miss New Orleans? Internet communication, geographic community, and social capital in crisis.
Journal of Applied Communication Research, 35(1), 67-87. doi: 10.1080/00909880601065722
Ratchford, B., Talukdar, D. & Lee, M. (2001). A model of consumer choice of the internet as an information source. International Journal of Electronic Commerce,
5(3), 7-21. doi:10.1080/10864415.2001.11044217
Rhine, R. & Severance, L. (1970). Ego-involvement, discrepancy, source credibility and attitude change. Journal of Personality and Social Psychology, 16(2), 175-190. doi: 10.1037/h0029832
Rosen, J. (2006). The people formerly known as the audience. Retrieved from http://archive.pressthink.org/2006/06/27/ppl_frmr.html on June 1st, 2015.
Seltzer, T. & Mitrook, M. (2007). The dialogic potential of weblogs in relationship building. Public Relations Review, 33(2), 227–229. doi:10.1016/j.pubrev.2007.02.011
35 Sherif, C., Sherif, M. & Nebergall, R. (1965). Attitude and attitude change. Philadelphia: Saunders.
Sundar, S. & Nass, C. (2001). Conceptualizing sources in online news. Journal of
Communication, 51(1), 52-72.
Sweetser, K. & Metzgar, E. (2007). Communicating during crisis: Use of blogs as a relationship management tool. Public Relations Review, 33(3), 340-342. doi: 10.1016/j.pubrev.2007.05.016
Taylor, M. & Perry, D. (2005). Diffusion of traditional and new media tactics in crisis communication. Public Relations Review, 31(2), 209–217. doi:10.1016/j.pubrev.2005.02.018
Thompson, J. (1990). Ideology and modern culture: Critical social theory in the era
of mass communication. Cambridge, UK: Polity Press.
Utz, S., Schultz, F. & Glocka, S. (2013). Crisis communication online: How medium, crisis type and emotions affected public reactions in the Fukushima Daiichi nuclear disaster. Public Relations Review, 39(1), 40–46. doi:10.1016/j.pubrev.2012.09.010 Westerman, D., Spence, P. & van der Heide, B. (2012). A social network as information: The effect of system generated reports of connectedness on credibility on Twitter. Computers in Human Behavior, 28(1), 199–206. doi:10.1016/j.chb.2011.09.001
Westerman, D., Spence, P. & van der Heide, B. (2014). Social media as information source: Recency of updates and credibility information. Journal of
Computer-Mediated Communication, 19(1). 171–183. doi:10.1111/jcc4.12041
Willemsen, L., Neijens, P. & de Ridder, J. (2011). “Highly recommended!” The content characteristiscs and perceived usefulness of online consumers reviews.
Journal of Computer-Mediated Communication, 17(1), 19-38.
doi:10.1111/j.1083-6101.2011.01551.x
Willemsen, L., Neijens, P. & Bronner, F. (2012). The ironic effect of source credibility on the perceived credibility of online product reviewers. Journal of
Computer-Mediated Communication, 18(1), 16-31, doi:10.1111/j.1083-6101.2012.01598.x
36 Yang, S.-U., Kang, M., & Johnson, P. (2010). Effects of narratives, openness to dialogic communication, and credibility on engagement in crisis communication through organizational blogs. Communication Research, 37(4), 473-497.
Vaske, (2011). Can weighting compensate for sampling issues in internet surveys?
Human Dimensions of Wildlife: An International Journal, 16(3), 200-215. doi:
10.1080/10871209.2011.571230
Weiner, B. (1985). An attributional theory of achievement motivation and emotion.
Psychological Review, 92(4), 548-573. http://dx.doi.org/10.1037/0033-295X.92.4.548
37
38 Table 3 – Crisis Communication strategies by Benoit (1997)
39
Appendix B: Survey
Original survey was presented in Dutch. Questions below only refers the most relevant questions for measuring the variables in this study. For a full version of the survey, please visit:
https://uvacommscience.qualtrics.com/SE/?SID=SV_eQmD1NIHCD1lj6d or the bit.ly shortened URL: http://bit.ly/1BGvG3p
QMacA and QMacB (Before stimulus and after)
40
Appendix C: Stimuli material
Condition 1: Press releasePersbericht – Een medewerkster van McDonalds is ontslagen nadat zij zich onbehoorlijk gedroeg op de werkvloer en schade aanbracht aan het restaurant.
Een medewerkster van McDonalds Zaandam heeft zich vorige week ernstig misdragen op de werkvloer. Tijdens een geschil met de filiaalmanager over de werktijden en het loon laaide de discussie op waarna de betreffende medewerkster gooide met etenswaren. Ook bracht zij vernielingen aan het interieur van het restaurant aan. Ze moest vervolgens onvrijwillig uit het pand worden verwijderd door collega’s.
In Nederland bestaat er een wettelijk minimumloon voor betaald werk. McDonalds houdt zich altijd aan de afspraken die in de cao staan vermeld. De schade die het restaurant heeft opgelopen zullen worden verhaald op de nu ex-medewerkster.
41
Condition 2: online news article (nu.nl)
ZAANDAM – Een medewerkster van fastfoodketen McDonalds is vorige week ontslagen nadat zij door het lint ging en een collega aanviel.
Vorige week zaterdag zorgde een medewerkster voor veel onrust op de werkvloer. De discussie begon tussen de medewerkster en haar manager over het late doorwerken en de verplichting om de werkvloer netjes achter te laten na sluitingstijd. De medewerkster klaagde tegen de manager, waarop de manager haar aansprak. Vervolgens vloekte de medewerkster over het lage loon dat zij ontving van haar baas. Hierna escaleerde de situatie snel en werd er zelfs met etenswaren en kassa’s door het restaurant gegooid. De medewerkster moest met harde hand uit het restaurant verwijderd worden door meerdere collega’s.
De oorzaak van de discussie betrof de ontevredenheid van de medewerkster over de werkomstandigheden bij de fastfoodketen. Overuren worden vaak niet uitbetaald en dat resulteerde dit keer in een fysiek gevecht. McDonalds medewerkers verdienen veelal het minimumloon wat is vastgelegd in de cao. McDonalds gaat de schade proberen te verhalen op de inmiddels ex-werkneemster van het bedrijf. © ANP
42
Condition 3: Blog post
Eten bij de McDonalds is niet altijd even gezellig.
Vorige week kreeg een medewerkster het aan de stok met een manager op de werkvloer. Ze begon te discussiëren over de vele onbetaalde overuren die ze moest maken, en de verplichting om na sluitingstijd aanwezig te moeten zijn tot het gehele restaurant schoon is. De discussie werd zelfs zo heftig, dat de medewerkster laaiend werd op haar baas en met etenswaren en kassa’s gooide. Hierop werd de medewerkster het gebouw uitgezet.
McDonalds betaalt overuren nooit uit en betaalt haar medewerkers niet meer dan het minimumloon, wat is vastgelegd in de cao van werknemers. McDonalds gaat de schade proberen te verhalen op de inmiddels ex-werknemer van het bedrijf.
43 Condition 4: News article De Telegraaf
ZAANDAM – Een medewerkster van fastfoodketen McDonalds is vorige week ontslagen nadat zij door het lint ging en een collega aanviel.
Vorige week zaterdag zorgde een medewerkster voor veel onrust op de werkvloer. De discussie begon tussen de medewerkster en haar manager over het late doorwerken en de verplichting om de werkvloer netjes achter te laten na sluitingstijd. De medewerkster klaagde tegen de manager, waarop de manager haar aansprak. Vervolgens vloekte de medewerkster over het lage loon dat zij ontving van haar baas. Hierna escaleerde de situatie snel en werd er zelfs met etenswaren en kassa’s door het restaurant gegooid. De medewerkster moest met harde hand uit het restaurant verwijderd worden door meerdere collega’s.
De oorzaak van de discussie betrof de ontevredenheid van de medewerkster over de werkomstandigheden bij de fastfoodketen. Overuren worden vaak niet uitbetaald en dat resulteerde dit keer in een fysiek gevecht. McDonalds medewerkers verdienen veelal het minimumloon wat is vastgelegd in de cao. McDonalds gaat de schade proberen te verhalen op de inmiddels ex-werkneemster van het bedrijf. © ANP
44 Condition 5: Facebook-page item
45
Appendix D: SPSS Outputs
Table 4: Socio-demographic variables sample
Socio-demographic variables Frequency %
Gender (N = 218, M = 32.44, SD = 12.50) Male 59 27.1% Female 159 72.9% Educational level Elementary 0 0% Vmbo/Mavo 20 9.2% Havo 12 5.5% Vwo 5 2.3% Mbo 59 27.1% Hbo 65 29.8% Wo 57 26.1%
Table 5: Reliability scale perceived source credibility
Cronbach’s Alpha Cronbach’s Alpha based on standardized items
N of items
0.901 0.901 5
0 ≤ alpha < 0.60 = unreliable scale
0.60 ≤ alpha < 0.80 = moderately reliable scale 0.80 ≤ alpha ≤ 1 =reliable scale
Table 6: Brand attitude
Mac brand attitude (recoded) Frequency %
(N = 218)
Negative brand attitude
58 26.6%
Netural brand attitude 54 24.8%
46 Table 7: Testing hypothesis 1: Offline versus online media in terms of perceived source credibility
Credibility N Mean Std. Deviation
online 173 2.1387 0.83063
offline 45 2.3333 0.92932
Levene’s test for Equality of Variances F Sig. t df Mean difference Equal variances assumed 2.878 0.091 -1.365 216 -0.19461 Equal variances not assumed -1.278 63.485 -0.19461
Table 8: Testing hypothesis 2: News media versus press releases in terms of source credibility
Credibility Frequency Mean Std.
Deviation Std. Error Mean (N = 218) News media 86 2.3372 0.92813 0.10008 Press release 40 2.2750 0.78406 0.12397
Levene's Test for Equality of Variances F Sig. t df Sig. (2-tailed) Mean Difference Std. Error Difference Credibility (news vs. Press) Equal variances assumed 2.950 0.088 0.367 124 0.714 0.06221 0.16944 Equal variances not assumed 0.390 89.048 0.697 0.06221 0.15933
47 Table 9: Testing hypothesis 3: Blogs verses news media in terms of source credibility
Table 10: Testing hypothesis 4: attitude on credibility
ANOVA Sum of Squares df Mean Square F Sig. Between Groups 7.717 2 3.858 5.158 0.006 Within Groups 160.818 215 0.748 Total 168.535
48
Descriptives
Credibility
N Mean Std. Deviation Std. Error
95% Confidence Interval for Mean Lower Bound
Negatief over McDonalds 58 2,5828 ,89465 ,11747 2,3475
Neutraal over McDonalds 54 2,7444 ,72935 ,09925 2,5454
Positief over McDonalds 106 3,0189 ,91027 ,08841 2,8436
Total 218 2,8349 ,88128 ,05969 2,7172
95% Confidence Interval for Mean
Minimum Maximum
Upper Bound
Negative regarding McDonalds 2,8180 1,00 5,00
Neutral regarding McDonalds 2,9435 1,40 4,40
Positive regarding McDonalds 3,1942 1,00 5,00
Total 2,9525 1,00 5,00
Post-Hoc Test
Dependent Variable: Credibility Bonferroni
(I) HER_MacA (J) HER_MacA
Mean Difference
(I-J) Std. Error Sig.
Negative attitude Neutral attitude -,16169 ,16355 ,972
Positive attitude -,43611* ,14125 ,007
Neutral attitude Negative attitude ,16169 ,16355 ,972
Positive attitude -,27442 ,14460 ,177
Positive attitude Negative attitude ,43611* ,14125 ,007
Neutral attitude ,27442 ,14460 ,177
Post-Hoc test, confidence interval.
Dependent Variable: Credibility Bonferroni
(I) McDonalds attitude (pre) (J)
95% Confidence Interval Lower Bound Upper Bound
Negative attitude Neutral attitude -,5563 ,2329
Positive attitude -,7769 -,0953
Neutral attitude Negative attitude -,2329 ,5563
Positive attitude -,6233 ,0745
Positive attitude Negative attitude ,0953 ,7769
Neutral attitude -,0745 ,6233
