BUILDING A BOND WITH CITIZENS ON FACEBOOK

BUILDING A BOND WITH CITIZENS ON FACEBOOK - Aalborg Allerød MSc in Economics and Business Administration Marketing Communications Management Department of Marketing Copenhagen Business School January 2011 Anna Sofie Brill Jørgensen Supervisor: Luisa Gorgone STU: 181,545 / 79.8 pages

EXECUTIVE SUMMARY As the municipal brand in general is facing the challenges of a bad reputation as a service provider and a decreasing interest in the local democracy this thesis offers valuable knowledge about how to use Facebook as a platform for getting closer to and creating a bond with the citizens by being on Facebook. To restore or create the bond between a municipality and its citizens via Facebook, the social medium must be used appropriately. Through a case study of the Facebook pages of two municipalities, Allerød and Aalborg, it is evaluated how appropriately the two municipalities have approached Facebook both stra- tegically and in practice, and what the potential is for the Facebook pages to work as a platform for a municipal- citizen bond. The frame of analysis is brand personality theory by Jenni- fer Aaker (1997) also described by Tilde Heding, Charlotte F. Knudtzen and Mogens Bjerre (2009). From this offset, the groundswell mindset by the authors Charlene Li and Josh Ber- presence. The findings are more specifically centered on the approach, the goals and the target au- i- y one Facebook page and since they are focused on pushing out one- way communication, i.e. information without having defined one or more target audiences, their Facebook page is evaluated to have a limited chance of being a platform that Allerød can use in creating a municipal- citizen brand. Aalborg on the other hand has chosen a project or campaign- based approach with several i- nions and activate them in the local community which is an appropriate offset for using Fa- cebook. Even though the research shows that they have not defined their target audience, they have a better chance of getting closer to the citizens in Aalborg through Facebook as their pages naturally targets smaller audiences. The research is specifically focused on the municipal side of using Facebook to strengthen the municipal brand through the relationship with the citizens. It calls for further research to find out how citizens actually bond with municipalities through Facebook and how other social media can be used in municipalities to improve the general reputation.

TABLE OF CONTENTS 1 Introduction... 7 1.1 Problem statement and research questions... 9 1.2 Delimitation... 9 1.3 Thesis progression... 11 2 Methodology... 13 2.1 Scientific standpoint... 13 2.1.1 The influence of pre- understanding... 13 2.2 Research design... 14 2.2.1 Case study design... 14 2.2.2 Cases... 15 2.3 Primary Data collection... 15 2.3.1 The structure of the interviews... 16 2.3.2 Interview guide... 17 2.3.3 Interviews with municipalities... 17 2.3.4 Interviews with experts... 19 2.3.5 Survey with stakeholders using Facebook... 20 2.3.6 Observations of the case municipalities behavior on Facebook... 23 2.3.7 Order of data collection... 23 2.4 Secondary data collection... 23 2.5 Validity and reliability of research... 24 3 Theoretic framework... 28 3.1 Brand personality... 29 3.1.1 Municipalities as brands and citizens as consumers... 30 3.1.2 Basic assumptions... 30 3.1.3... 31 3.1.4 The brand- consumer relationship... 32 3.1.5 Limitations of the brand personality approach... 33 3.2 The groundswell mindset... 34

3.2.1 The POST- method... 35 3.2.2 Strategy and objectives... 36 3.2.3 People: Target audience... 37 3.2.4 Limitations to the groundswell mindset... 39 4 Analytic strategy... 40 5 Allerød... 42 5.1 Presentation of Allerød... 42 5.2... 43 5.2.1... 44 5.2.2... 44 5.2.3... 46 5.3... 46 5.3.1... 46 5.3.2... 47 5.3.3... 48 5.4 Evaluation of approach, goals and target audience... 48 5.4.1... 48 5.4.2... 51 5.4.3 ook... 55 5.5... 57 6 Aalborg... 58 6.1 Presentation of Aalborg... 58 6.2... 60 6.2.1... 61 6.2.2... 62 6.2.3... 63 6.3... 64 6.3.1... 64 6.3.2... 66 6.3.3... 67

6.4 Evaluation of approach, goals and target audience... 68 6.4.1... 68 6.4.2... 70 6.4.3... 73 6.5 Building a relationship between Aalborg and the citizens on Facebook... 75 7 Discussion... 78 7.1 Experimental approach... 78 7.2 Resources for facebook pages... 79 7.3 Facebook in a time of cost cutting... 81 8 Recommendations... 82 8.1 Allerød... 83 8.2 Aalborg... 84 8.3 Municipalities in general... 85 9 Conclusion... 88 10 Perspectives... 91 10.1 Measuring how personality is perceived... 91 10.2 Social media in general... 92 10.3 Brand communities... 92 11 Literature... 93 11.1 Books and articles... 93 11.2 Web... 93 List of figures... 97 List of appendices... 98 Appendix 1-16... 99

CHAPTER 1: INTRODUCTION

Chapter 1: Introduction 1 INTRODUCTION Municipalities have a reputation of being secretive and closed as the common perception is that it is difficult to get help and answers. This has often been in the limelight of the press - we often hear stories in the news about how municipalities do not live up to standards and how municipalities in general are perceived to be more trouble than help. Since the municipal reform in 2007 there has been even more focus on the municipal service and how the municipalities are managed only making the general image of municipalities worse. The municipalities have to do better than good in order to change the publi Furthermore, the municipal government is facing a decrease in participation in the municipal elections. At the last election in 2009 the participation dropped about four percentage points from 69.5 % in 2005 to 65.8 % (Danmarks Statistik kommunalvalg 2010). Especially the participation of young people is low; 55 % of people aged 18-21 did not show up to vote. Municipalities face a lack of interest in local democracy and need to get closer to the citizens in order to engage them. In addition, municipalities are as all other organizations experiencing how the internet is enabling people to communicate and find information online. Citizens have expectations to how the municipalities should carry out their online presence. Citizens today count on being able to find out what they need about municipal services and decision making on the internet, and in addition they want to be able to get the most basic services via online self- service. This is a challenge for the municipalities, but at the same time offers a lot of possibil- ities. Citizens who are requesting services digitally is less costly than servicing citizens face to face and with the constant pressure on public expenses and t zero- growth plan for municipal expenses (Dr.dk Indland 2010), digitalization must be a top priority. As communication via online tools has become more and more acceptable, municipalities can use the currently available digital tools to cut costs. social network site Facebook is the preferred online network of the Danes with more than 2.5 million profiles (Dwarf 2010). People use Facebook to keep in touch with friends and networks but also brands, companies and municipalities are on Facebook with a Facebook 7 / 218

Chapter 1: Introduction This newsfeed also shows updates of a pe becomes part of a place of trust where people communicate with each other about their personal experiences, problems and what matters to them in general. A person can also a page, comment on it and share the content if he or she has an opinion about shown in the newsfeed, once again making it visible for others that this person has an interest in the given municipality. With Facebook pages municipalities therefore have the possibility of being part of a conversation and send out messages and content relevant for citi- g- es. In other words, being present on Facebook is a way to brand one self, also for a municipality. In my opinion, the mere fact that you are on Facebook signals openness and an interest in communicating with the citizens. Facebook centers on the social element, i.e. two- way conversation and sends strong brand impressions out to the public by having a municipal Fac e- book page. But there has been limited research about this topic. However, a successful Facebook presence requires an understanding of mechanisms of the social network site. People can write whatever they want on the Facebook page of a municipality and if they do not live up to the expectations of citizens who use Facebook, it can cause serious damage to the already challenged municipal brand. Facebook in an organiza- tional context must be managed by skilled people, but municipalities do not necessarily have Facebook experts employed who can ensure a beneficial presence. In addition, municipalities have other aspects on top of this to consider as they are public organizations The expectations to some of the larger commercial brands who have led the way for companies on Facebook so far are different, as municipalities are financed by tax money. omentum (KL.dk 2010b) shows that municipalities are increasingly moving onto Facebook. So far 48 % of the 98 m u- nicipalities have an official page or other pages focused on a special topic or project. Fur- 8 / 218

Chapter 1: Introduction thermore, 2/3 of the respondents have answered that they will join Facebook in the coming year. They have seen the potential of reaching their citizens with their messages, both politi- cally and service oriented. Many municipalities are testing Facebook to see how it works and what the benefits are. My concern in this context is that, as any other communication chan- nel, municipalities have to recognize that their behavior on Facebook sends signals about what the municipal brand stands for and if and how citizens want to be related to it. There have been few studies in how public organizations can strengthen or weaken their brand by having a page on Facebook, which makes it interesting to approach this field of research. Facebook offers a way of getting closer to the citizens by having a conversation with them on their preferred online social networking site and this offers great branding potential. 1.1 PROBLEM STATEMENT AND RESEARCH QUESTIONS The introduction leads me to the problem statement, research interest and the purpose of this thesis. Facebook is a communication tool that municipalities can use to post content and act in a way that brands the municipality with the potential of changing the general bad i m- age of municipalities into something positive. This can also be achieved by opening up to the citizens in conversation about the topics that matters to them. Thus, I focus my research into the field of municipalities on Facebook based on the following problem statement: How does a municipality act appropriately on a Facebook page to establish and build a bond between the municipality and its citizens? What behavior on Facebook is recommended in the literature, by experts and by Fa- cebook users? What should a municipality focus on expressing through Facebook to create a bond with citizens? 1.2 DELIMITATION I have decided to study how municipalities use Facebook pages and if and how it in turn can citizens. My approach is the branding angle. I therefore delimitate the research from dealing with what democratic or political effect it might have that municipalities go on Facebook. If 9 / 218

Chapter 1: Introduction municipalities can get a positive branding effect by being on Facebook there is a possibility it can affect the current local democracy, however this is not the purpose of the current branding research. In addition, I have chosen to focus on the issue of the municipality in this matter.. I am interested to find out how municipalities can find out about their challenges and how they can work with the challenges concerning Facebook. This is the reason why I find it necessary to go into depth with the municipal side of the relationship in this thesis. Furthermore, I will focus on a certain delimitated stakeholder group the citizens. Municipal a- mental task to ensure the welfare of the citizens per definition. Other stakeholders such as the state, the press, companies, newcomers and tourists are very important stakeholders for the municipalities as well, each in their own way. However, regarding Facebook I assume that the first priority is the citizens and therefore this is where I will have my starting point. I also choose to delimitate this thesis to focus only on the functionality of Facebook pages as my initial research has shown that this is what most municipalities are using Facebook for each in their own way. Facebook ads or using Facebook groups or other functions are not as much used for branding purposes or in a municipal context, which is why my research will revolve around pages. In addition, this thesis will study pages that municipalities are respon- sible for creating and maintaining. It would be too extensively to consider pages created by citizens or other stakeholders about municipalities and this does not match my research interest as I will concentrate on what municipalities can do on Facebook to strengthen their brand, not what others are doing. Finally, I will not be looking at how suitable Facebook is as a technology in itself. I will evaluate how appropriately the municipalities use this technology as this lies within the current research interest. 10 / 218

Chapter 1: Introduction 1.3 THESIS PROGRESSION The structure of the thesis is as showed in the following figure. Figure 1: Thesis progression Ch. 1 Ch. 2 This chapter introduces the field of research as well as problem statement, research questions and limitations. Based on the research questions, this chapter explains the research methods and design including scientific standpoint, case study approach and data collection method. Ch. 3 Here the basic assumptions of the theories are described and the arguments for the links between the theories chosen for the thesis are presented. Ch. 4 In this chapter the data collected is presented and analyzed in the theoretic framework; first the case of Allerød and then the case of Aalborg. Ch. 5 Ch. 6 In chapter five, the findings from the analyses is discussed and the recommendations following from this are given. Finally, chapter six provides the conclusions based on the findings and reflections of the previous chapters. In the end, perspectives for further research is presented. 11 / 218

CHAPTER 2: METHODOLOGY

Chapter 2: Methodology 2 METHODOLOGY In this chapter I will explain the scientific standpoint that I am working from as well as the method for data collection. The scientific fundament is the starting point for the choice of research methods which will also be described in this chapter. The section ends with an illu- stration of the methodology. 2.1 SCIENTIFIC STANDPOINT The starting point for this thesis is a social constructivist view in which knowledge is recog- nized as a product created by social processes (Fuglsang & Olsen 2004). This approach is part of what Fulgsang et al. (2004) characterizes as complex idealisms. Together with movements such as phenomenology and hermeneutic, constructivism belongs in the interpretive paradigm, which is different from the positivistic paradigm (positivism and critical rationalism) or the scientific realisms (systems theory, critical realism, Marxism etc.). These views have a respective realistic approach and a more critical research interest (Fuglsang & Olsen 2004). The interpretive paradigm on the other hand regards truth as dependant on time and space of the object or phenomenon studied, and research is a dynamic process. This stands in op- position to the positivistic paradigm which seeks to uncover the objective truth about a phenomenon or subject. As the fundamental goal of this thesis is to reach a better understanding of how the use of Facebook is connected to branding of municipalities, the interpretive paradigm and the s o- cial constructivist approach offers the best scientific fundament because context relational interpretation based results will be regarded as valid. 2.1.1 THE INFLUENCE OF PRE- UNDERSTANDING ok as research field is a result of my interest in and work experience with the subject. I recognize that I am influenced as a researcher by my pre- understanding of this field. I have an interest in Face- book rooted in my curiosity for the technology behind it and the functions of communicating with friends and acquaintances through it. I use Facebook on a daily basis and a lot of the people I know do the same. Furthermore, from working with municipal communication in 13 / 218

Chapter 2: Methodology LGDK) communications department the last two years, I have In line with the interpretive method I have chosen, it is a given that it is impossible to study an object unaffected by pre- understanding. I will be highly conscious of my pre- understanding of the field when I do my research to make sure it does not have an effect on the thesis making the analyses biased. On the other hand my experience and knowledge is useful in structuring my research as I already have a certain basic understanding of the su b- ject. In the next section I will explain the practical aspects of the research for this thesis. 2.2 RESEARCH DESIGN It is necessary to discuss and decide on the grounds for the research as it must be considered what implications the methods can have on the results (Fuglsang & Olsen 2004; An- dersen 1998). Furthermore the research must support the research interest which in this case is the use of Facebook in municipalities and how this affects the municipal brand. Below I will give an account of my considerations regarding methodology and purposes for the data collection. 2.2.1 CASE STUDY DESIGN Following from the empirical character of my research, based on the research approach and the scope and time limit I am subject to, I choose a deductive, multiple case study design for my research (cf. Andersen 1998: 165ff). A deductive approach means concluding on cases based on theoretical principles instead of inductively making conclusions based on the empirical research. The multiple case study design is chosen over a single case study, as I aim to achieve knowledge from more thar- ences between Danish municipalities a study of only one municipality is too narrow. A single case study would on the other hand give me the option of examining a specific municipality more in depth, but that would according to Yin (1989, in Andersen 1998: 165ff) only be relevant if the goal is to e.g. test a theory on a specific case, if the case in question is extreme or unique or if I wanted to reveal the effects of a certain phenomenon on a specific organiza- 14 / 218

Chapter 2: Methodology tion. In this thesis the aim is to produce knowledge about two cases which can inspire and be an eye opener to other municipalities using or thinking about using Facebook. Thus, the multiple case study approach offers the best platform (Andersen 1998) In the paragraph below the choice of cases are described. 2.2.2 CASES After researching a variety of municipalities on Facebook, I have chosen to focus on the municipalities of Allerød and Aalborg. Throughout this thesis the two municipalities will only be have different approaches to being on Facebook as will be evident from the details given throughout the thesis. In short, Allerød has one central Facebook page and Aalborg has several project- centered pages. Furthermore, the municipalities are located in different parts of Denmark and they are different in size both by population and employees. These are the details in focus for this thesis other points of differences, for example political, are left out as they are not within the focus or scope of the research as also pointed out in the delimitation section. I acknowledge that Allerød and Aalborg is not representative for all municipalities in Denmark, but I am of the opinion that the examples can capture some of the general differences in how municipalities are present on Facebook which can act as an eye opener to other municipalities. The details about the further investigation of the research topic are explained in the section about data collection below. 2.3 PRIMARY DATA COLLECTION To ensure that the research topic is looked at from all angles relevant, the primary data collection will include data collection from four perspectives: The head of communication in the municipalities of Allerød and Aalborg Two communication consultants who work with municipalities and their presence on Facebook, and one expert in branding and social media The opinion of Facebook users about municipalities on Facebook 15 / 218

Chapter 2: Methodology I will interview the managers of communication and the communication and branding consultants. The opinion of Facebook users will be researched through a survey. Finally I will observe the behavior and activity on the Facebook pages in focus. This approach mixes qualitative and quantitative data, which according to Andersen (1998) is appropriate and highly useful for case studies. The two types of data will complement each other; the qualitative research, i.e. interviews, will go into depth with the options for branding through Facebook; and the quantitative data, i.e. the survey, will outline and show the broader tendencies in the public that must be considered when using social media in municipalities; the observations will function as a fundament to hold the interviews up against. The data collection in general will focus on the current situation and thus be a snapshot of the situation instead of for example a dynamic study over time as my research interest concerns the current situation and not the development in municipalities use of Facebook (cf. Andersen 1998). The interviews and the survey will be explained further in the paragraphs below. 2.3.1 THE STRUCTURE OF THE INTERVIEWS The five interviews have the purpose of gaining insight to the problem area from the perspectives of the people working with it either from the municipal angle or as a consultant external to the municipalities. The chosen interview approach is semi- structured (Andersen 1998:205; Hair et al. 2009:182). The semi- structured approach refers to the method of working with a set of questions, i.e. an interview guide, while still making room for relevant new ideas, thoughts or topics that comes forward through the interview (Kvale 2000). The alternative would be a more strict approach in which no deviations from the prepared questions of the interviewer would be allowed. This would limit the research from gaining important points not within the exact limits of the questions and I thus find the semi- structured approach most appropriate (An- dersen 1998; Kvale 2000). The interviews can furthermore be characterized as experience or informant interviews, as (Hair et al. 2009). This data collection technique deviates from the classic in- depth, research interview that aims to gain in 16 / 218

Chapter 2: Methodology data (Hair et al. 2009:158; Kvale 2000:19). It is not the point of the interview to reach an understanding of the underlying personal motives for the respondents but instead get the professional perspective on the research topic which is why the experience interview form is the best suited for the purpose. Furthermore it ensures consistency enough to conclude on the findings of my analyses. 2.3.2 INTERVIEW GUIDE To make sure I get answers to the topic in question, I will use an interview guide as earlier mentioned. The guide is structured in three parts according to the recommendations by Kvale (2000): 1) Introduction and creating comfort zone by asking opening questions in the form - questions 2) Going in to depth with the research area - questions 3) Summarizing and closing the interview During the interview, I will ask questions that encourage the respondent to elaborate further on special areas or to get direct answers about special details. This can be done by probing which can result in asking leading questions. I will strive to avoid this as this can bias the research (Hair et al. 2009; Kvale 2000; Bryman & Bell 2003). Also Kvale (2000) recommends periods of silence that allows for the respondent to think and further comment on a question if necessary. A part of the interview guide will focus specifically on the concepts of the theoretic perspectives of the thesis that will be explained in the theory section. The questions will directly re- late to branding and the use of social media in order to have a specific link to the theoretic tools that creates the fundament for the later analysis. 2.3.3 INTERVIEWS WITH MUNICIPALITIES The first part of my primary data collection is interviews with the managers of communication in each of the two municipalities Lotte Birk, Aalborg, and Jesper H. Pallesen, Allerød. 17 / 218

Chapter 2: Methodology They are expected to have the knowledge relevant to my research interest, i.e. the overview and understanding of the use of Facebook and the possible strategies behind. The interviews will be carried out by telephone as this was a necessary choice because of the limitations of time and resources. I recognize that this means an absence of visual aids and that the respondents and I cannot use regular cues such as body language or face expres- sions to guide the conversation (Andersen 1998; Hair et al. 2009; Bryman & Bell 2003). The advantage of telephone interviews is on the other hand that it will reduce the possibility of my personal characteristics or appearance affecting the respondents or vice versa. To prevent misunderstandings, I will give as many verbal cues as necessary and make sure to ask declaring questions if I have any doubt that my questions are not understood thoroughly (cf. Hair et al. 2009). Also periods of silence as Kvale recommends (2000) is especially important in the situation of interviewing by phone where I cannot tell from body language if the respondent has nothing further to add. As mentioned by Andersen (1998) interviewing people in leading positions can be proble- expe- rience as well as the responsibility to act as a representative of the organization resulting in diplomatic statements. It is thus necessary that I am well prepared and make sure to argue for the purpose of my research as this can increase the likeability of more honest answers (1998). It must also be mentioned that it takes a lot of skills to be a good interviewer (Hair et al. 2009; Kvale 2000). There are several criteria such as listening and probing skills that are pri- marily attainable through experience that I as a student do not have. This can affect the interviews, but considering the extensive preparations and my experience from earlier aca- demic projects I trust the interviews will still be of adequate quality. See Appendix 1 and Appendix 3 for the interview guides for each municipality. 18 / 218

Chapter 2: Methodology 2.3.4 INTERVIEWS WITH EXPERTS As the second part of my primary data collection I will interview three experts in communication and branding. This ensures that I get the experience from projects about branding and being on Facebook from other angles than the ones of the municipalities. The first respondent chosen is chief consultant at LGDK, Jesper Brieghel. His job is among other things to advise municipalities about being present on Facebook. He is chosen because he has great insight into the communication and branding issues facing the municipalities in general. Secondly, I have chosen the managing director Rasmus Møller- Nielsen from the Danish consultant company Komfo as a respondent. Komfo are the authors of a Facebook guide for the municipalities (Komfo nyhedsstrøm 2010; Komfo 2010) and have worked with several municipalities. Rasmus Møller- Nielsen is, as Jesper Brieghel, assumed to be knowledgeable about the situation for the municipalities at the moment and what can be done with Facebook in that context. Furthermore Rasmus Møller- Nielsen works with other customers than municipalities which can prove to give insights different to the ones of Jesper Brieghel. Finally, I will interview tactical activator Morten Saxnæs from the Danish advertising agency PeopleGroup. He works in the division called DialoguePeople and is skilled and experienced in using several social media platforms, including Facebook, for a variety of customers. He is a relevant respondent as he works with creating dialogue in an advertising and branding context and hence, he is focused more directly on the branding issues than the two other experts that work more from a corporate communication starting point. Each of these three interviews will be carried out face to face in the offices of each of the three experts. This has the advantages, as mentioned above, that both the respondent and I can use body language, facial expression and other visual cues to guide the conversation. The disadvantages however is that the face to face form can affect the behavior of the respondents for example they might not feel comfortable with the situation. Furthermore, as experts they can have a tendency to prove themselves and try to lead the interview in their preferred direction (Andersen 1998; Kvale 2000). Also, my lack of interviewer skills, as ex- 19 / 218

Chapter 2: Methodology and their reactions in general. Evaluating both the advantages and disadvantages the interview form is still suitable as I will get a great opportunity to get in- depth knowledge of how municipalities can work with branding and Facebook. The statements from the three experts will add a broader perspective to the research and raise the quality of the analysis I will be able to make. The interview guide for experts can be seen in Appendix 5, 7 and 9. 2.3.5 SURVEY WITH STAKEHOLDERS USING FACEBOOK The last part of the primary data will be collected through a survey, i.e. a questionnaire, with the purpose of defining the broader tendencies in the expectations and perceptions of The survey has been chosen over a focus group or personal interview approach that would have aimed at understanding the deeper personal opinions of stakeholders. As pointed out by Kvale (2000) a qualitative approach such as interviews is not suitable when the research aims to show the state of the mind of bigger groups of people because it would be too time consuming and require substantial resources to interview everyone. Thus, in a case study as the current a survey offers getting opinions from a larger sample than if I had chosen a focus group or personal interviews. 2.3.5.1 TARGET POPULATION According to Facebakers, a website with the newest statistics on Facebook, there are over 2.5 million Danish Facebook users (Facebakers Denmark). The survey will target the general majority of Danes on Facebook; those in the age of 25-34 (Ibid.). The sampling design will resemble a snowball sampling method characterized by respondents referring other people as possible respondents (Hair et al. 2009:323) and is chosen especially according to the research objective, the time frame, resources and scope (cf. Ib- id.:324f). 20 / 218

Chapter 2: Methodology 2.3.5.2 THE STRUCTURE OF THE SURVEY As can be seen from Appendix 13 (the survey questions) I use structured close- ended questions as I wish to get answers to specific questions and because this reduces the amount of effort the respondents need to use (Hair et al. 2009). I recognize that this means that some answers can more or less forced. However, there will be comment boxes for the questions asking about behavior intentions and opinions, giving the respondents the option of elabo- rating on their answer. This opens up for interesting insights valuable for the research. Open questions have furthermore been left out because the respondents are assumed to have very different knowledge and prerequisites for answering questions about municipalities. Furthermore, the survey is built upon questions with different scales to get the answers necessary. The nominal scale is suitable for getting the precise answers that I need to understand what type the respondents are both in his/her use of Facebook, age, location etc. as this can be used for tabulation. Ordinal scales are appropriate for measuring state of mind which is exactly the purpose of the survey. The ordinal scales are to be more specifically de- signed as five- point, free choice scale as they include a neutral response choice (Hair et al. 2009:354f). The survey includes both a Likert- scale that measures a mental or behavioral belief by asking to what extent the respondent agrees/disagrees (Ibid.:370f); and a behavior intention scale that asks if the respondents would or would not interact with a municipality on Facebook. As earlier stated, some respondents may not have enough knowledge to answer either positive or negative. Therefore, in order to ensure the results are not skewed, the neutral response will be given as an option in all questions except in the probability scale as this is important for the analysis. The questions will concern the general opinion of how Facebook users expect municipalities to behave and some will regard the theoretic concepts that will be described later in this thesis. I could have chosen to ask more specifically about the two municipalities studied but this would limit the respondents to people that have knowledge and an opinion about exactly these municipalities. Instead I have chosen to let the survey focus on municipalities in general which will then be related to Allerød and Aalborg. This delimitation is also connected to my research focus that concentrates on the municipal side of the research area as stated under limitations earlier. 21 / 218

Chapter 2: Methodology Some of the questions are asked to be able to see if there are remarkable links e.g. between political standpoint and expectations to municipalities level of activity on Facebook. There- fore I do not expect all results to be equally relevant to my analyses. The results of the survey and their relevance also depend on the results of the other parts of the data collection, and thus, only the relevant results will be presented in the analyses. 2.3.5.3 TESTING THE SURVEY Before the survey gets distributed it will be tested by a small group of people in the age of 25 to 34 which resembles the users of Facebook. A test is obligatory to ensure that the questions are understandable, that the questions are in a logical order and that the survey has an acceptable length etc. (Andersen 1998). 2.3.5.4 DISTRIBUTION The survey will be created electronically and distributed online via a link. The link to the survey will be send out by e- mail and I will continuously post the link on my Facebook profile, to my followers on Twitter, and my connections on LinkedIn as well as encourage my contacts to forward it to everyone they know 1. Since I have contacts to people in the bigger cities in Denmark Copenhagen, Aalborg, Århus and Odense this distribution method is assumed to be effective and thus most suitable in this situation where I am not in possession of con- tact data like professional research companies or companies with a customer database are. The risk when using Facebook and other social media for distribution of the survey is that the people using these media the most will answer the questions and the results will thus be biased as the ones using these media less will not be fairly represented. This is a necessary risk that must be taken as the current distribution method is the one where I reach the most people. The reach of the survey is important since the starting point of the research is the municipalities in general which makes it of high priority to get answers from people located in differ- 1 Twitter is a microblogging service where you do not have to be friends with which updates you follow (Wiki- pedia Twitter 2010). LinkedIn is an online network similar to Facebook but for professional contacts (Wikipedia LinkedIn 2010). 22 / 218

Chapter 2: Methodology ent parts of Denmark this because it is assumed that the opinions will vary from location to location. 2.3.6 OBSERVATIONS OF THE CASE MUNICIPALITIES BEHAVIOR ON FACEBOOK Finally, I will observe and gather data about how each municipality behaves on Facebook with the purpose of holding this up against how each head of communication describes their Facebook presence in the interview. In doing so, I draw on the twenty- o- graphy, the so- called netnography. A netnographic study typically revolves around an internet community and serves the purpose of gaining consumer insights much like from a survey or a focus group (Kozinets 2002; Heding et al. 2009). However, I will only be quantitatively studying the number of posts and likes, i.e. clicks on the like- button of posts, and who authors the posts. I will supply these observations with considerations about the type of content of the posts. These studies will make the fundament for my evaluation of each munici- 2.3.7 ORDER OF DATA COLLECTION I have chosen to collect all primary data at the approximately same time. There are both advantages and disadvantages of for example carrying out the interviews after the survey results but in order to prevent my research from the bias this might cause, I have chosen not to. 2.4 SECONDARY DATA COLLECTION Besides from the primary data collection involving the people directly involved with the area of research, I will draw on secondary data that makes up the fundament for parts of the analyses. This includes: Statistical data about the municipalities for example from Danmarks Statistik (Statis- tics Denmark) Research reports and statistics about Facebook use in Denmark and about Facebook use in general for example from Komfo (2010; 2009), Dwarf (2010) and KL (KL.dk 2010a) 23 / 218

Chapter 2: Methodology These facts and results of examinations of the Danish population and Facebook match the social constructivist view that is the fundament for data collection. Society is at the same time socially constructed and objectively real, thus these quantitative secondary data are relevant to include in my research. When both primary and secondary data about these research area have been collected I have the data required for the analysis that will lead to an answer to the problem statement for this thesis. However the quality of the research must be assessed which is the focus of the section below. 2.5 VALIDITY AND RELIABILITY OF RESEARCH Every method has advantages and disadvantages and the choices made for the current research have consequences. As the scientific point of departure is social constructivist I do not strive to find one objective truth about the topic of research. The results of the data collection are affected by the context of data collection which involves the macro perspective, i.e. the broader societal context, as well as a micro perspective, i.e. the situation of the interview or the setting in which the survey is answered. As a researcher I recognize this as an inevitable circumstance that is forming the results which has also been explained in the sections above. However, an assessment of the quality of data is in order. Quality is dependent on validity and reliability. The relationship between these parameters can be illustrated as in the figure below. 24 / 218

Chapter 2: Methodology Figure 2: The relationship between reliability and validity (Hellevik 1980:158 in Andersen 1998:110) Validity determines to what extent the data has been collected correctly. The model shows how the validity of data is dependent on the how the theoretically defined concepts match the variables set up in the data collection. The theoretic concepts, as will be explained further in the theory- section, are defined by acknowledged and peer- reviewed author, which adds to the validity of definition. Furthermore, the validity of definition for the current data collection is found to be relatively high as I work with the theoretic concepts used in the analysis directly in both interviews and survey. Furthermore I believe the right variables are included in the research as they are chosen on the basis of preliminary research, data and knowledge of the field of research. Reliability concerns how much the results of my research can be trusted and if the results were the same if the research were repeated. As touched upon in earlier sections it can be argued that the reliability is limited as I only work with two municipalities. The ideal study would have taken more municipalities into consideration and thus come closer to a more generalizable result than the current case study aims for. The aim for the case studies will thus be to present both results specific to the municipalities of Allerød and Aalborg and to discuss how they to some extent can be useful for other municipalities without generalizing. The reliability of the survey is acceptable however limited. As only a small sample of respondents is attainable, because I do not have access to databases that can give me a representative sample of the population in Denmark, I will only use the data from this survey to point 25 / 218

Chapter 2: Methodology me in the direction of what Danes are likely to think about social media on a certain set of parameters. Even though I cannot do a perfect survey I still find it interesting to see what tendencies it will show. The municipalities can put all their efforts into social media that they want if the target audience cannot see the purpose of the municipalities being on Face- book it makes little sense to put resources in it. To sum up, the research methods have an acceptable degree of validity considering the general challenges to the research made as part of this thesis as well as the theoretical defini- tions that are prepared for the data collection. In the next chapter the theoretical tools and concepts will be explained and I will account for the way they will be used when analyzing the data. 26 / 218

CHAPTER 3: THEORETIC FRAMEWORK

Chapter 3: Theoretic Framework 3 THEORETIC FRAMEWORK This chapter serves the purpose of giving an overview and explaining the theoretic perspectives that I will apply to my research. First I will briefly introduce the two perspectives and how they are interlinked. After this short introduction I will in more detail describe each To examine the data described in chapter 2 (methodology) it makes sense to start in brand personality theory as described by Heding et al. (2009) (1997) work rand personalities. Brand personality theory centers on the bond between a brand and an individual which fits the current research interest in to how municipalities can bond with their citizens. Municipalities do not have a product to sell in the same way commercial companies do, but it is imperative that they create a relationship with their citizens. Brand personality theory will thus be the offset and the frame within which I am interested in how a municipality can use Facebook to create a brand- consumer (municipality- citizen) bond that will strengthen the overall municipal brand. In order to evaluate how a municipality can make Facebook a platform for creation of a bond with citizens it is necessary to study how a municipality approaches Facebook and behaves on a page. In order for Facebook to used appropriately. Hence, within the frame of Facebook as a tool to create and strengthen y, I will evaluate how the two municipalities that are in focus, (2009) mindset about taking part in the groundswell, i.e. the conversation online about a brand. Groundswell is a strategic mindset that can help companies and organizations to understand and be a part of the conversation about their brand online. Furthermore, it offers a set of guidelines and recommendations to follow in order to avoid the pitfalls of using groundswell technologies, i.e. social media such as Facebook. The link between brand personality theory and the groundswell mindset is that there cannot be a brand- consumer (municipality- citizen) bond, unless symbolic value is communicated and perceived. For symbolic value to be communicated and perceived on Facebook the 28 / 218

Chapter 3: Theoretic Framework communication must follow the basic principles of the groundswell or the communication will not get through to the citizens. Thus, in order for Facebook to function as a platform from which Allerød and Aalborg can create bonds with citizens through communication of symbolic value, they need to use Facebook as a groundswell technology and what this en- tails. This link is illustrated in the figure below. Figure 3: The link between brand personality theory and the groundswell mindset (own make from Heding et al. 2009:128) Behavior on Facebook following the mindset of the groundswell Symbolic brand value Symbolic brand value Brand personality Personality of stereotypical consumer Adds personality and symbolic value This short introduction leads me on to a fuller description of first, brand personality theory and second, the groundswell mindset. 3.1 BRAND PERSONALITY If a person proudly describes himself or herself as a citizen of a municipality it is an expression of loyalty towards the municipality and thus a sign of brand strength. This connection 29 / 218

Chapter 3: Theoretic Framework between the municipality and the citizen is an example of the dyadic relationship that the brand personality approach centers on which is why this approach has been found suitable as the frame of analysis in present thesis. The brand personality theory as proposed by Aaker The Dimensions of Brand Personality (1997) and described by Heding, Knudtzen and Brand Management Research, Theory and Practice(2009) will make up the theoretic grounds for analysis and the key elements and how I will use them are described in the following paragraphs. I begin by stating how I will work with municipalities as brands. 3.1.1 MUNICIPALITIES AS BRANDS AND CITIZENS AS CONSUMERS In this thesis I view and define municipalities as brands. To be specific, I look at the municipality as a whole, i.e. both as a service provider and a democratic institution. It is my assumption that citizens of a municipality will have relations to both the service and the democratic outputs in their lifetime and thus, both aspects are included. The next paragraphs will explain how the brand personality views brands and brand- to the definition of a municipality as a brand, as citizens are consumers both in the sense of consuming services such as day- care, elderly care, housing support etc., but also as consumers of democratic services such as information about local politics, decision making and similar. I will not distinguish between citizens as consumers of one or the other types of services; After this initial explanation of my understanding of the research objects, I will describe basic assumptions of brand personality theory. 3.1.2 BASIC ASSUMPTIONS The central point of branding theory is what makes consumers choose one product over another. In the personality approach it is believed that the consumer will choose the brand part of the movement from an economic approach in brand management towards a more consumer oriented and consumer- dialog based approach. The underlying assumption is that consumers use brands that offer them something for their identity or expression of self 30 / 218

Chapter 3: Theoretic Framework through the symbolic benefits of the brand. If the personality of the brand matches what the consumer needs for his or her construction of self or as an expression of self to others, the consumer will choose this brand over others (Heding et al. 2009; Aaker 1997). Thus, a brand should aim to signal a brand personality that matches what consumers want for their personality. In a municipality- citizens context the above description means that in order for a municipality to create a bond with citizens, the municipality should communicate symbolic value that is attractive to the citizen in his or her self- construction. Relating the basic assumptions to being on Facebook, it makes good sense to talk about personality as Facebook can be characterized as a medium in which the individual promotes its personality through status updates, personal photos, content sharing etc. (Dwarf 2010)r- sonality on Facebook is through pages the individu preferences in music, politics, brands and for example a municipality, other people are able approach suitable as it has its starting point in how individuals use brands to construct their own personality also in a Facebook setting. 3.1.3 NALITY DIMENSIONS The brand personality approach has especially found recognition with the work of Jennifer Aaker in 1997 w the set of human characteristics associated with a brand (Aaker 1997:347) and the article describes her ro- e- able, Consciousness, Emotional stability, Openness/Cultured (Smit et al. 2003:24) research suggests:... that consumers perceive that brands have five distinct personality dimensions: Sincerity, Excitement, Competence, Sophistication and Ruggedness(Aaker 1997:353). These five dimensions each consist of a set of facets that describe the dimensions in more detail to add to the accuracy of the understanding of e.g. sincerity. 31 / 218

Chapter 3: Theoretic Framework r- sonality is perceived i.e. through the behavior, attitudes, physical appearance, beliefs etc. with it. Aaker (1997) mentions for example the treatment the customer gets from the com- u- tion of the brand, packaging etc. Thus, both human characteristics and product- related attributes contrib (Aaker 1997; Smit et al. 2003). In my research I focus on how communication has importance for the construction of brand personality as I am focusing on the possibility of making Facebook a platform for the creation of municipal brand personality and a municipal- citizen bond. Thus, the thesis is delimitated from focusing on other attributes. I am aware that this leaves me with a limited picture of the brand personality but I find that this analytical delimitation suits the research interest and makes it more focused. 3.1.4 THE BRAND- CONSUMER RELATIONSHIP As earlier outlined, the center of attention in brand personality theory, from the viewpoint of the marketer, is creating a brand- consumer relationship. If a brand has a personality, i.e. can be linked to a set of human characteristics, it is more likely that the consumer will form an emotional bond with it (Heding et al. 2009). The brand personality approach is characterized by a dyadic relationship between the brand and the consumer, both affecting the other. According to Heding et al. (2009) this line of thought is closely related to the brand relationship approach in which a brand must have a personality in order for there to be a consumer- brand relationship (Ibid.:172). However, the relational approach differs by viewing the consumer as an existential being. The objective of research in the brand relationship approach is n- sumption in a marketing context. This research interest departs in a phenomenological scientific standpoint that dictates only qualitative measures (Heding et al. 2009). Compared to the brand personality approach, the relational approach more strongly views the brand as 32 / 218