Description
Market research is any organized effort to gather information about markets or customers. It is a very important component of business strategy.
Case Study for Investigative Management and Consumer Research on the Internet
Abstract This paper introduces the notion of Investigative Research on the Internet (IRI) and conceptualises its processes through the principle of streaming. It discusses the similarities and differences between IRI and netnography and considers various aspects of the IRI process, including site selection, sampling, data collection and analysis. It is argued that streaming can help to understand the processes involved in conducting netnographic research. Moreover, it is suggested that streaming is a more appropriate way to conceptualise some internet-based studies that do not conform to netnographic or ethnographic ideals. Three international empirical cases are used to illustrate the application of IRI and streaming in research on international workers, consumer cultures and on emerging business phenomena.
Keywords: Internet, online, virtual, ethnography, netnography, qualitative research, streaming analysis
2
1. Introduction
The development of web 2.0 and interactive technologies offer numerous opportunities for qualitative management and consumer research. Researchers may use instant messaging services and chat rooms (virtual spaces facilitating synchronous communication) to conduct interviews and focus groups (Fielding et al., 2008; Grossnickle and Raskin, 2001). These enable fieldworkers to collect data through real-time interaction, although there are various other sources that facilitate asynchronous data collection, focusing on archived material (Hewson et al., 2003; Dholakia and Zhang, 2004). This paper advances existing knowledge on online methods by introducing the notion of investigative internet research, or Investigative Research on the Internet (IRI), and distinguishing such qualitative, exploratory studies from other forms of online research. It is argued that IRI shares many of the characteristics of netnography (Kozinets, 2002) and other forms of online ethnography (Hine, 2000, 2005), but that not all IRI can or should be called netnography. We conceptualise the processes involved in such internet-based research using the notion of streaming, which can inform future netnographic studies as well as other IRI that does not conform to ethnographic ideals. Finally, the paper demonstrates, through three international empirical cases, how IRI and streaming can be applied in research on international hospitality workers, consumer cultures and networks, and more broadly on emerging hospitality business phenomena.
3
We concentrate on three sources of data: 1) message boards or forums, which enable users to engage in asynchronous exchanges or postings on different topic threads; 2) blogs, narratives and opinion pieces written by individuals, to which others may respond and which may also provoke ongoing asynchronous exchanges; and 3) websites provided by commercial and non-commercial organisations or individuals, offering news, reviews and either lay or professional commentary. We discuss the investigative processes, including navigation in and across multiple 'field sites', sampling, analysis and interpretation, through which researchers can exploit these three sources of data in their studies.
The paper contributes to knowledge in several ways. Firstly, we illustrate how internet research and streaming can be applied to address issues relevant to hospitality managers, academics working in broader disciplines, alongside policy makers. Secondly, and more broadly, it uses research on hospitality to conceptualise qualitative, investigative, internet-focused research that is applicable to other intellectual contexts. Consequently, rather than simply importing concepts from other disciplines, hospitality research creates opportunities to contribute to wider fields of inquiry.
2. Conceptualising investigative research on the internet
It is important to begin by distinguishing IRI from other types of internet-based methods of data collection, for example, text mining. These approaches use software that searches the internet and attempts to identify patterns and 4
relationships between textual information (Miller, 2005). IRI may use specialist software, and it will inevitably employ search engines; however, it does not rely on textual material alone, and the researcher has a much greater role in sourcing, analysing and making connections between sources of information.
IRI has a number of features. Firstly, as the term investigative implies, it is inherently analytical, exploratory and seeks to simultaneously discover and construct. This implies that the research process attempts to unearth existing pieces of information, which emerge in a variety of ways, for example through objects, actions, images, texts, relationships, interactions, spaces and institutions. Moreover, the researcher or investigator connects these to other pieces of data that may not seem immediately related. Secondly, the research act is a dynamic process, responsive to changes in the fieldwork 'site', but also requiring sensitivity towards the individuals who provide information, towards relationships and interactions during the fieldwork, and towards the various sources of data, which may not be predetermined at the study's outset. Thirdly, because the research is dynamic, it also has to be context sensitive, taking account of the subtle idiosyncrasies of the virtual spaces through which researchers navigate. Researchers also have to be mindful of the social, cultural and sub-cultural factors entangled in the offline worlds in which data emerges. Finally, as noted above, in IRI the researcher is not a neutral entity, but is central to the research process, making decisions at every moment about what is relevant, what is excluded and which lines of inquiry are pursued. IRI thus shares many of the features of online 5
ethnography (Hine, 2000; Kozinets, 2002; Wittel, 2000); consequently, it is useful to discuss these research approaches in further detail, before introducing the notion of streaming to conceptualise the processes of conducting IRI.
3. Ethnography and netnography
Our discussion primarily draws on and engages with netnography, rather than other conceptualisations of internet-based ethnography (e.g. Hine, 2000; 2005), partly because Kozinets (2002; 2010) has attempted to define its processes in detail, but also because the concept clearly distinguishes it from traditional ethnography. Moreover, netnography, which was originally developed for marketing and consumer research, is rapidly becoming an accepted concept in business research, increasingly being applied to the study of hospitality (e.g. Watson et al., 2008).
Netnography, like its antecedent, ethnography, is not a single method or approach; it involves a range of investigative strategies. The lack of space prevents a detailed discussion of ethnography, but it is useful to briefly consider some of its key features, which helps to understand netnography's underpinning principles. Lugosi (2009), synthesising earlier discussions, suggests ethnography seeks to gain an emic understanding of groups and individuals through immersion in their activities over extended periods of time. The emerging insights are context sensitive and embrace the complexity of human phenomena. The researcher, rather than being a detached neutral 6
entity in the research process, has a central role in knowledge generation. This means fieldworkers have to be critically sensitive about their relationships with informants and the data they develop. Ethnographers often use some variation of participant observation, although this is often complemented by interviews, the analysis of written material, and, increasingly, by visual methods (see e.g. Pink, 2007). Historically, the insights gained through ethnography have been communicated through rich textual accounts, although these, too, have been enhanced by the use of technology and visual methods.
Netnography also attempts to understand, rather than measure, but Kozinets (2010, p.5) highlights several features that distinguish it from ethnography. Firstly, interaction is different than in 'traditional', face-to-face ethnography. Engagement can be with a much wider range of geographically dispersed people, and participation can take a number of different forms in online environments. Data can be gathered without any interactions but engagement may also range from brief exchanges to longer and deeper interactions, using a variety of communications e.g. text, sounds, images and other 'virtual' expressions such as emoticons. Secondly, rather than relying on field notes and creating textual accounts, netnography gathers and uses a wider range of data, which varies in terms of focus and quality. This means the researcher has to broaden her or his repertoire in gathering, interpreting and integrating vastly different data. These are themes echoed by others distinguishing between traditional and virtual ethnographies (e.g. Wittel, 2000).
7
Kozinets (2010, p. 89) outlines six criteria for selecting sites for study, arguing that they should be 1) relevant to the research focus and questions, 2) active, with recent and regular communications, 3) interactive, having flows of communication between participants, 4) substantial, in terms of numbers of users, 5) heterogeneous, involving different participants, and 6) data-rich. Kozinets (2002) also proposes a number of procedures to facilitate the process of conducting online research and identified several stages: entry into the 'community', data collection utilising a variety of textual and visual material, analysis and interpretation, as well as member feedback, which requires an overt approach. However, this was challenged by Langer and Beckman (2005) who suggest adopting a covert approach and Beaven and Laws (2007) who propose conducting observational or passive netnography. Informed consent is seen to be unnecessary when there is an openly available archive of online material or when forums are publicly accessible (Beaven and Laws, 2007).
Kozinets' clear guidance helps to shape the direction and processes of online research, but there remain several areas of debate for fieldworkers, which make it difficult to think of all IRI as netnography. The first concerns the centrality of community in netnographic research. The notion of community has continued to evoke critical debate, and there have been extensive discussions about its appropriateness in understanding the multiple relationships that exist in cyberspace (see Jones, 1998). Contemporary debates have shifted away from traditional conceptions of community, based, for example, on structures and stable roles, to more fluid ones, suggesting 8
that people continually (re)construct and perform notions of shared identity and belonging (Guimarães, 2005).
Kozinets (2010) acknowledges the problematic nature of 'community', but maintains that the focus of netnography is on 'online communities', i.e. groups constructed and maintained in virtual worlds. He also advocates the increased use of netnography, in conjunction with traditional forms of ethnography, to study 'communities online', i.e. groups that exist offline, but also interact through the web. According to Kozinets (2010), communities involve a minimum number of people, who identify with others in the community and engage in ongoing, meaningful interactions with other members. This can be used to describe some of the social configurations and interactions relevant to hospitality management research, but it is problematic to designate individuals with shared interests or involvement in particular work and leisure activities as being part of a community.
Contemporary ethnographers, engaged in 'multi-sited' fieldwork, have maintained that ethnography continues to offer insights into the complex sets of actions through which individuals construct and articulate relationships and notions of self/other (see e.g. Falzon, 2009; Guimarães, 2005). Extended engagement with multiple 'informants' provides opportunities to understand behaviours, values and norms that enables researchers to define and thus construct individuals as belonging to a community. Netnography, and ethnography in and of virtual environments, therefore becomes a productive as well as an analytical act. Consequently, if we acknowledge the central role 9
of the researcher in constructing a notion of community, then we have to recognise that the fieldworker's version of community may not be shared by the people being represented.
A final challenge, which relates to the previous point, concerns the implicit assumptions about the nature of knowledge and understanding gained through netnographic research. The central role of the researcher, and her or his extended engagement with individuals offering insights into the nuances of interactions, norms, values and identities, is one of the defining characteristics of both netnography and ethnography (Hine, 2000, 2005; Leander and McKim, 2003). Netnography, like ethnography, thus stresses a particular epistemological ambition in assuming that immersion into a group and their culture can help fieldworkers develop emic perspectives. However, if notions of community are fluid and highly contestable, and engagement with informants may be fragmented and asymmetric, not all the interactions and relationships can provide detailed ethnographic data, a point acknowledged by its practitioners (Hine, 2000, 2005; Markham and Baym, 2009). Moreover, not all IRI may set out to or claim to have gained such ethnographic insights, and describing such research as netnography is therefore potentially problematic.
Calling all IRI 'netnography' is an example of concept creep - a simultaneous extension and reinterpretation of netnography. Kozinets (2002), Hine (2000) and others recognise that netnography, and other forms of online ethnography, are flexible approaches, involving a range of methods and 10
techniques, and are thus open to critical reconsideration. This paper engages in the reinterpretation of netnography by proposing the notion of streaming to help understand the processes involved in such internet research. However, IRI and streaming offer a broader conceptualisation of qualitative, exploratory internet-based studies. This is necessary and useful for two reasons: firstly, rather than simply, and uncritically, inflating the netnography concept, this paper defines a key aspect of such online research, thus helping researchers conduct netnographic studies. Secondly, distinguishing the processes involved in IRI from the concept of 'netnography' helps researchers to understand and conduct research via the internet, which may not conform to ethnographic ideals.
4. Investigative research on the internet as streaming
We use 'streaming' as a metaphorical device to conceptualise the data gathering and analysis processes in which researchers trace relationships between actors and agencies through the internet. The notion of streaming may initially seem to be limited, as is any other metaphor: it may create an image of a defined path to be followed and imply a singular direction, something which does not reflect well the multi-spatial and multi-directional nature of the internet and the movement of its users. Streaming echoes Hine (2000) and Leander and McKim's (2003) focus on mobility and stresses the fluidity of relationships and interactions. Streaming should be thought of as a dynamic, creative process, when the researcher opens new directions for data gathering and analysis as he or she moves through and across different 11
sites. As ethnographers working in multi-sited environments have argued (Hine, 2005; Leander and McKim, 2003; Falzon, 2009; Wittel, 2000), fieldworkers connect, perform and effectively construct the field site. Even if they establish one set of connections, thus creating a 'stream' through which they move, they can retrace their movement, move backwards, identify new divergences and create alternative connections - streams.
At the core of streaming is the notion of connectedness and the importance of mobility. Urry (2007) is one of many commentators to stress that contemporary society is characterised by increased movement both as people move around physically but also as information circulates rapidly, facilitated by the growing prevalence of technology (see also Hannam et al., 2006). Consequently, individuals are less likely to be members of stable social groups, based on strong social ties; instead, people are connected, often through loose ties, to a wide range of other people through networks that stretch across geographical boundaries (Urry, 2007). Streaming can thus be thought of as a research aim to follow the networks of relationships through which information about people, places and experiences flows. Just as Hine (2000, p.35) noted, rather than remaining on a single site or focusing on one context, internet research should use movement between sites as an analytical device. The key issue is that researchers attempt to understand how relationships and flows of information operate through virtual media. Movement across different virtual spaces in data gathering and analysis also implies that streaming is more likely to examine interactions and
12
representations during extended periods and across several time periods, rather than only at a single point in time.
A further feature of streaming is its inductive nature, involving simultaneous data collection and analysis (Lofland and Lofland, 1995). Concurrent data gathering and open, inductive analysis are necessary as representations and interactions occur across multiple discussion threads or even websites. In many cases, even when interactions take place on a single forum, they include links to other sites. Examining relationships and interactions through virtual networks inevitably leads to choices that researchers make about which sites, texts, discussions or representations of people, places, events or individuals should be traced further. These have the potential to become what we refer to as key streams, which lead to new lines of inquiry. Within traditional ethnography, 'key informants' have often played a crucial role in gaining access, mediating on fieldworkers' behalf and providing information. The notion of key streams collapses the distinction between people, sites and texts, as they all have the potential to be crucial reference points researchers may draw upon once, but may return to repeatedly, to gain insights. There may also emerge more minor streams, located elsewhere, that are also useful but less so than key ones.
Streams have an elaborative function, providing illuminating illustrations of behaviours, events, beliefs, values etc., thus helping to understand the dimensions and dynamics of particular phenomena. They also have a connective function, linking to further sites, documents, representations etc. 13
where issues can be explored further. It is also useful to distinguish between key streams, which are the primary sources from which streaming begins and sub-streams that may exist within, or extend from, these key streams. These may be the discussion threads on a site, or representations by individuals e.g. in the form of personal narratives, informal commentaries, or lengthier, formal discourses, identified by the researcher as relevant based on the examination of another site or text.
Finally, the process of following posts or threads and moving between different websites also helps to identify a final key feature of streaming, which is that it involves snowball sampling that may be complemented by other forms of criterion sampling, within a broader purposive sampling approach (see Patton, 2002). Internet interactions are driven by user generated material, and users can create new linkages between other users, sites and threads of discussion; therefore, it is important, in principle, to let the data and the emerging themes shape the sample size, rather than it being defined too narrowly at the outset of the research. Ideally, data collection will stop when the researcher arrives at a point of theoretical saturation, when no new themes emerge from the data (Strauss and Corbin, 1998). However, the specificity of the study's aim and objectives and resource restraints, for example in terms of funding, time, project life etc. will often impose other limits on the study. Consequently, the research may come to focus principally on a specific forum, a limited number of threads - based on some research question driven criteria, a fixed time period, a limited number or network of users or even an individual, which is used as a particular case. 14
5. Streaming in practice
5.1 Streaming in research on employee experiences
Background and purpose Migrant workers make up a large part of the hospitality workforce and their experiences are frequently shaped by their immigrant status, language skills and their ability to mobilise social capital: they are frequently confined to low paid, low status jobs and hospitality offers particular career pathways, while denying others (Janta et al., 2011). The first study used to illustrate streaming is part of a wider project examining Polish migrant employees' experiences in the hospitality and tourism sector. The study applied a mixed-methods approach to allow an understanding of respondents' lives at work and beyond it, including their interactions, aspirations, work conditions, job scopes and demographic profiles. However, we concentrate here on the qualitative internet research dimension of the study. A key objective of the research was to examine migrant networks and how networking, in virtual environments in particular, mediated their employment experiences. More specifically, it sought to understand how migrants supported each other and represented their experiences of mobility and hospitality employment to other current and future migrants.
Data collection and streaming There is a high usage of social networking sites and discussion forums among Poles in the UK (Janta et al., 2011), and the public nature of these 15
forums prompted the researcher to engage in observational or passive netnography (Beaven and Laws, 2007). Streaming utilised purposive sampling and began by identifying the key streams, using the six criteria, outlined earlier, to select appropriate sites. Initially, the site's traffic and activity were prominent criteria. Three important sites were identified: mojawyspa.pl (my island) with 23,000 threads, ang.pl, a site dedicated to those learning English, with more than 72,000 threads, and gazeta.pl - the biggest forum with more than 7,400 different forums on it and more than 102 million written posts. These became the key streams, and the starting point for streaming was a specific virtual space - i.e. the forum, from which networks of relationship and interactions were traced.
Within these key streams it was then necessary to identify sub-streams, which were the relevant message boards and threads, found by analysing headings and, where possible, by using the sites' internal search engines. At the level of sub-streams, essential sampling criteria were increasingly the relevance of posts to the research objectives rather than just the traffic. Examples of topics included 'job seeking' and 'applying for positions', 'hotel work' and 'other jobs in hospitality'. The relevance of some streams was obvious, for example: 'Working in a hotel? Discuss it!' or 'Why are educated Poles washing pots in the UK??' on gazeta.pl or 'Manchester-waiter-a few questions
' on mojawyspa.pl. However, other titles were less obvious, for example 'Are you planning to return to Poland one day?' on gazeta.pl and others discussing migrants' future plans, but these made repeated references to hospitality work. Approximately 100 existing threads, launched between 16
2004 and 2008, were downloaded, printed and analysed in this research. Some continued for only a few days while others until 2010.
Key emerging questions that drove the processes of streaming, as noted in the previous paragraph, concerned the thread topics and subsequent interactions. Examining the threads led to other relevant issues and helped answer such questions as who uses the space and for what purpose. For example, a thread launched on gazeta.pl, entitled 'How is it getting a job after returning to Poland?', had an intensive beginning, with 95 posts within 3 days (11-14.04.2006). Tracing this stream helped to understand the development of migrant careers over extended periods, including the experience of return and the possibility to utilise the human capital developed during work in the UK. Other streams showed how migrants interacted, building their social networks. For example, a thread 'Anyone going to Bournemouth?' on ang.pl was launched in 2005 and the last post was written in January 2010.
Indicative emerging themes identified through streaming Streams of inquiry helped to understand specific aspects of instrumental and emotional support directly related to hospitality employment. For example, one reply to a posting entitled 'Interview - waitress - on Friday help me!!!' (ang.pl, 27.05.2008) provided twenty examples of possible interview questions such as 'What do you like about waitressing?' and 'Are you good at multi-tasking?' Other aspects of support revealed through tracing key substreams incorporated reflections on hospitality work. Examples included streams entitled 'Why are qualified Poles washing pots in the UK?' 17
(02.08.2006), which discussed the experience of having to take jobs for which they were overqualified. Others were less philosophical and more prescriptive, such as the stream entitled 'Don't come! No work here!', where posts revealed the bleak situation of Polish migrants: 'Only those who have worked in the UK before and those who speak English well will find a job even such as 'on the sink' [i.e. low status kitchen porter positions] - this is what it looks like now!' (gazeta.pl, 30.01.2007).
Not all the postings were so negative; some sub-streams also revealed information about migrants' positive work-related learning experiences. For example, a thread entitled: 'A waiter in England' (gazeta.pl, 25.11.2004), started by a user seeking advice on learning English while working in foodservice, included a number of constructive postings by other users:
It is hard work, especially when it gets busy and you need to deal with a number of things at the same time. But it gets better, for example when they ask you where you come from and they don't want to believe that you come from a non-English speaking country and they appreciate your English
Another reply, posted on the same day, also revealed themes of emotional support from fellow migrants and positive work-based encounters. It also helped identify further streams of inquiry concerning the role of formal education in migrants' experiences:
18
Don't give up! I don't know why [the other site users] are scaring you! I work behind a bar, maybe it's not the same as being a waiter but it is not simple either! (sometimes) I have been learning English for 4 years and I don't have a FCE [First Certificate in English] (yet) but I don't have problems in communicating. So don't put him off! People are really understanding!
Significance, applications and contributions to knowledge Streaming, using particular online forums as initial points of focus, was important in meeting the study's objectives because these were spaces of interaction dedicated to a particular social group. However, these spaces were used by individuals from a wide range of geographical areas and it would have been very difficult to reach such a disparate network of individuals through conventional methods. In some cases, sites, and threads within them, concentrated on specific geographical areas or activities e.g. hospitality work or language learning, which helped to focus in on specific subgroups, based on location or activity. Finally, streaming was useful because it helped to follow particular lines of enquiry extending from these key streams.
By conducting internet-based research on this and related topics, through streaming, hospitality academics can contribute to knowledge outside their subject areas, thus helping to raise the profile and influence of the research community. More specifically, migration has become a sensitive political 19
topic, especially in the current economic climate, so understanding their experiences, for example relating to adjustment, learning, access to work and social inclusion/exclusion has clear implications for policy makers regarding the type of migration that should be encouraged and how migrants can be supported to make positive contributions to society. Furthermore, understanding how migrants perceive their work, understanding their career aspirations, appreciating their processes of learning, particularly through hospitality work has clear relevance to practitioners who devise human resource strategies concerning recruitment, training and development as well as retention.
5.2 Streaming in research on consumer experiences
Background and purpose Recent years have seen the development of 'foodie' culture and the rise of enthusiastic consumers using the internet to communicate their meal experiences and their passion for food (cf. Watson et al., 2008). Blogs describing restaurant visits are narratives that weave together a complex set of events into a coherent story and thus help to understand the consumer experiences restaurants offer. Such forms of consumption and their representation through blogging can be considered a form of serious leisure (Stebbins, 1992), which is used by people to construct their identities and build communities of consumption (Watson et al., 2008). Many food bloggers are examples of Stebbins' (1992) skilled amateurs: they become self-aware participants in their chosen activity and develop the ability to 20
articulate their emotions and express their opinions. The second case illustrating streaming is a study of blogging, with particular reference to meal and restaurant experiences. Given the central role of identity construction and communities of consumption within food blogging, a key objective of the research was to understand how internet users articulate their identities and engage with networks to mediate their food related experiences.
Data collection and streaming This study also adopted a passive or observational form of online research, although permission was sought from the blogger to study the site. Streaming and the identification of key streams began with a broad search of the internet using 'food blogs' as the core search term. From the millions of returns, irrelevant ones, for example those focusing primarily on personal cooking, were rejected. As in the previous case, the sampling for key and sub-streams was purposive, and a specific blog was selected because it was substantial, relevant and active. The key stream, Grab Your Fork (grabyourfork.blogspot.com/, referred to hereafter as GYF) was established in 2004. The blogger, Helen Yee, posts regularly and responds to each comment individually. GYF won of the 'Best Restaurant Review Blog' award in 2006, and was a finalist in 2004, 2005, 2007 and 2008 in various categories for food blog awards. In addition, it was listed fourth in a worldwide top 10 food blogs by The Times in 2008. Yee is based in Sydney, Australia but comments are made by people from all around the world. The study focused on posts made over 24 months during 2006 and 2007. In total this covers 460 posts plus 2774 comments by other visitors, but 15 key sub21
streams discussing 'memorable meals' were identified within this process of streaming. One example of these key sub-streams is her description of a visit to Sydney's famous Tetsuya's restaurant in April 2005, which received 51 comments, the latest as recent as September 2009.
In this study the starting point for streaming was the individual, Helen Yee, and the study traced her experiences, interactions and relationships extending from her blog. Key themes and emerging issues driving, and emerging from, streaming included presentations of self, particularly as they are articulated through her descriptions of food and eating. Presentations of self were also important issues in other users' posts. Additional important themes were her extending networks, the reciprocal transactions involved, and the shared codes and meanings in interactions.
Indicative emerging themes identified through streaming The construction of identity and articulations of self were central features of food blogging and Yee was candid about her passion for food and why she blogs: 'An unbridled commitment to share, to comment and to espouse the joys of food are the main reasons why many of us blog' (Helen Yee, GYF, 27. 01.2010). Her commentaries also revealed other aspects of her identity, for example, her use of English points to her constructed class status: 'The roast suckling pig embodies all three in glorious gormandizing hedonistic glory' (Helen Yee, GYF, 02.02.2006). Moreover, her descriptions of eating offal and other animal parts reveal information about her ethnicity and how it defines food habits, for example, 'if there are three particular idiosyncrasies 22
about Asian palates it's our love of bones, skin and fat' (Helen Yee, GYF, 02.02.2006).
Beyond Yee's commentaries, it was useful to trace her interactions through her blog to better understand how communities of consumption were articulated through online interaction. There are numerous visitors to Yee's blog, some simply 'lurking' (passively observing rather than participating) or making transient visits, but she has developed relationships with several posters. Between them they have established interactional repertoires and codes. For example, they appear to be familiar with technical culinary terms, including 'galette', 'confit' and 'quenelle', which are used without explanation. Moreover, Yee and followers of the blog regularly perpetuate interactional routines with comments and descriptions of their meal experiences or of reading other people's meal experiences:
Sorry I meant "yee sang" in my earlier comment. It's a bit hard to type and drool at the same time as you know. (Oslofoodie, GYF, 02.02.2006)
Tracing the interactions, which occur in physical and virtual settings, also reveals the potential role of Yee and her extending network as informed critics who influence purchasing decision making:
[...] next time try the wat tan hor and fried radish cake which are both very good and if having the ice kacang make sure 23
you ask for them to put enough rose water. you'll see they put corn in there's [sic]. some pics of the dishes i've had are available to look at [link attached] in my food slideshow. (SimonLeong, GYF, 16.10.2006)
Comments such as this, from 'SimonLeong', not only identify subtle details of the meal experience, they also draw upon, and reveal, aspects of their own identity. References to heritage and personal experience are thus entangled with commentary on various aspects of the service operation:
I am Burmese and new to Sydney. I have been to several other Burmese restaurants in Asia, UK and US, and I must stay I was rather impressed with the food here. It is authentic indeed and the service from Victor and his staff were [sic] impeccable. (Cho, GYF, 16.10.2006)
Furthermore, as the quotes from 'SimonLeong' above and 'fooDcrazEE' below highlight, interactions between Yee and other blog followers provide links to further sites and discussion threads, which helps the researcher illustrate key points, identify other users and trace further streams.
What a feast! The YEE SANG is different from Malaysia though. Take a look here for a post of Malaysian Yee Sang. [Link attached] (fooDcrazEE, GYF, 02.02.2006)
24
Significance, applications and contributions to knowledge A streaming approach, in this case taking the individual blogger as the starting point from which to extend investigations, was central to meeting the study's objectives for several reasons. Firstly, focusing on a committed skilled amateur, such as Yee, provides a rich starting point for examining how individuals express notions of identity through serious leisure. Yee is an extreme case because of her nuanced appreciation of foodways and her commitment to expressing her engagement. Following Yee thus helps to generate a wide set of conceptual themes that are useful when examining the behaviour of other skilled amateurs. Secondly, Yee is a common point of reference, which other similar enthusiasts use when articulating their own identities and expertise. Linked to this, Yee's blog is an expressive space where performances of self are mediated and others can develop interactional routines and connect to a broader social network of likeminded individuals.
Applying this technique to such social activities also has a number of implications for academia and practice. An analysis of food blogs, again, provides opportunities for hospitality academics to make contributions to knowledge in other fields, for example in leisure studies about emerging forms of serious leisure, or to social scientific debates about identity, networked sociality and its relationship to food. Researching food blogging also has implications for hospitality marketing and operations management. Key individuals in networks disseminate information, shape tastes and influence purchasing decisions. Therefore, identifying influential individuals, 25
like Helen Yee, understanding their motivations, modus operandi and relationships with other consumers provides clear marketing opportunities. Moreover, their critical commentary on decor, food and service gives managers insights into consumer perceptions of their commercial propositions.
5.3 Streaming in research on emerging hospitality trends and phenomena
Background and purpose The final illustration of streaming takes as its focus an emerging trend or phenomenon in commercial hospitality. This is the most ambiguous of all three types of streaming because trends may actually lack clear definition, and exploratory research offers a way to clarify them and their dynamics. The use of streaming to conceptualise the research process is particularly appropriate in these cases because, although it may be informed by an ethnographic sensibility and desire to gain emic insights, it is problematic to call it netnography. Emerging phenomena are amorphous in nature and there may not be a clear community to which they can be ascribed.
Some hospitality phenomena such as 'homestay' (Lynch et al., 2009) have evolved to a stage where their manifestations in different international contexts are easy to identify. However, there are many other emerging ones that are only now being investigated and understood. One such phenomenon are 'rom' and 'kert' venues, which developed in Budapest, Hungary in the 26
beginning of the 21st century. The term 'rom', meaning 'ruin' in Hungarian, refers to hospitality venues operating in dilapidated buildings in Budapest. 'Kert' means 'garden' and was used in reference to outdoor venues. The two terms were frequently used in combination as rom were located in the open courtyard of venues, but the two should be treated as separate categories of hospitality operation. No research had been conducted on these types of operation so a key objective of this study was to understand what the 'propositions of hospitality' were in these venues and how this genre of hospitality emerged.
Data collection and streaming Tracing the evolution of the rom/kert venues helps to illustrate streaming that may, at various stages, involve focus on individuals and specific places, but ultimately focuses on the phenomenon. Initial broad investigations began with general internet searches on commentary that included references to one venue, the Szimplakert, which was considered by many commentators to be pioneering. Moreover, it was important to examine what other venues were discussed in relation to or in reference to Szimplakert. Identifying similar venues and descriptions of them helped to understand these venues, not as individual operations, but as points of reference in a constellation of venues.
An important aspect of streaming was identifying key streams, which included cultural commentators and virtual outlets where commentary was located. Three key streams were identified, again, though the focused use of search engines. Index.hu is an established media site, which has a number of writers 27
producing regular features on culture, politics and urban affairs. One commentator, András Földes, wrote several articles for Index.hu on the emergence and growth of the rom/kert phenomenon. He also collaborated on a guide, distinguishing venues according to customers' 'neediness'/'needlessness', and discussed elsewhere the struggles venues faced in the 2005 summer season. Szimpla also maintained a press centre on their website, which stored scanned copies of articles about the Szimplakert that usually discussed it alongside other, related venues. Index.hu, Földes and the Szimpla repository, cross-referenced each other and became the key streams used to trace the evolution of the rom/kert phenomenon.
Key sub-streams were subsequently identified. These were used to pursue the emerging themes in the research including the venues' defining operational characteristics, variations on the rom/kert model, and the collaborations and conflicts between different stakeholders influencing venues' survival. Importantly, the information gained from Földes, Index and Szimpla's archived material helped to understand not only the venues as individual commercial entities but how they existed in relation to their social, cultural, physical and political environments.
Additional important streams were traced through other sites and individual blogs that made reference to venues and to the rom/kert phenomenon. For example, professional reviews and lay commentaries made repeated reference to particular design features and cultural events hosted in the 28
venues. Some websites contained a single article, followed by comments from different users giving their own accounts of venues. The small number of active users, who engaged in dialogue, made it problematic to treat such idiosyncratic commentaries as authentic representations. Therefore, as Hine (2000) suggests, it was necessary to consider individual commentaries and representations in relation to others, which may not directly acknowledge, engage with or respond to them. Streaming, using the phenomenon as the focal point from which to extend inquiry was therefore necessary in constructing a rich story of the rom/kert phenomenon. Interpreting different representations in relation to others also became a form of triangulation. Several attempts were made to contact some of the bloggers and writers who provided insightful accounts, but none were successful. Nevertheless, the study was augmented by visits to venues, the collection of photographic data, and recorded interviews with operators.
Indicative emerging themes identified through streaming The emergence of this hospitality genre can be attributed to other venues operating in Budapest in the 90s, but commentators agree that a pioneering example of a romkert was the Szimplakert, which opened in 2002. The first Szimplakert was housed in the courtyard of a building awaiting demolition. This venue, including the use of courtyards of empty residential buildings, became the template which was reproduced, albeit slightly differently, in subsequent years. In 2003, two similar venues opened in the district; and by 2004 numerous operators had created their own version of the rom/kert. More recently, numerous other venues have opened in surrounding city districts 29
and several have opened on rooftops rather than in courtyards (see Lugosi et al., 2010).
Tracing different representations of venues and their histories also helped to understand the hospitality proposition of the venues and how these propositions were shaped by the social, economic, political and physical environment. For example, the decaying nature of the urban fabric, aging population, rising 'rent gap' between the potential value of the land and the income actually generated from rents led to the emptying of numerous buildings. Many of the venues were established because buildings awaiting demolishment could be occupied by operators, and urban decay, therefore, gave these venues distinct identities. Some of the districts were going through a process of social and economic regeneration. Several operators exploited pauses in the regeneration, for example, while developers and local authorities clashed over development plans. Other operators actively contributed to an area's regeneration by maintaining a 'cultural scene' (see Lugosi et al., 2010).
Tracing multiple representations of the venues revealed how distressed, mismatched tables and chairs, table football, lampions and artistic decorations were fundamental to the propositions of hospitality in these spaces. Closely linked to this were the various cultural events hosted by the venues, including book launches, talks and debates, fashion shows and art exhibitions. Moreover, tracing consumer representations of the venues,
30
through streaming, helped to understand how consumers experienced the venues. For example:
[...] last time I was at Szimpla, a huge norwegian contingent was there - love norwegians dont' [sic] get me wrong, but not the kind who feel the need to paint their flags on their faces and ladies loo was filled with Aussies. Actually, foreigners don't bother me, as long as they are the right kind (a bit boho, can string an intelligent sentence together etc) but those trendy fake-nailed, hair-straightened, designer clothes wearing idiots that would be better at the White Party or whatever its [sic] called. Leave the romkocsmak to the rest of us who probably wont' [sic] be spending 3 hours to get ready to go there. (pestcentric.com, 08.05.2007 [emphasis added])
These types of commentaries provide critical, though fragmented, information about consumer perceptions of the venues and their customer profiles, including how both have changed. Moreover, such reflections on the scene, offered insights into commentators' (sub)cultural values and their sense of identities.
Significance, applications and contributions to knowledge The use of streaming was particularly important in understanding the evolution of this genre and the propositions involved. Because of its fragmented, evolving nature, identifying key and sub-streams helped to 31
create a composite picture of rom/kert hospitality. Beyond the transactions of food and drink, multiple cultural activities were entangled in the genre. Therefore, only by tracing multiple representations, which included those of cultural commentators, critics, journalists, operators and customers, could the complex dynamics of its development be understood.
The examination of the rom/kert and similar trends and phenomena has a number of implications. Firstly, understanding this interaction between venues and urban transformation contributes to debates among academics working in areas of geography and social policy, and, potentially, policy makers, concerning the multiple contributions hospitality can make in urban regeneration. Secondly, for commercial operators, investigating the dynamics of this genre of hospitality can inform their business strategies in occupying unorthodox physical spaces and utilising internet-based consumer-toconsumer interaction to understand consumer tastes and create demand. Furthermore, investigating this genre can inform operational policies concerning music, decoration and cultural activities used to define their commercial propositions. Pursuing these lines of inquiry, through streaming, again, provides hospitality researchers opportunities to contribute to wider academic, cultural, commercial and policy agendas.
6. Conclusion
This paper has discussed netnography and introduced the notion of streaming to conceptualise a related investigative process on the internet. As 32
we argued, online research might be informed by the principles of netnography, and, more broadly, ethnography, but it may be problematic to call all such investigative internet research 'netnography'. In short, all netnography involves streaming, but not all IRI involving streaming is netnography. We introduced three studies to demonstrate different aspects of streaming, including the identification and tracing of key and sub-streams. We suggested that the researcher can begin tracing streams of inquiry from three initial points of focus: 1) a specific virtual space, 2) an individual, and 3) an emerging trend or phenomenon. Table 1 summarises potential themes and issues to be considered in the different focuses of streaming, which were highlighted in the three cases.
Table1. Initial focus and potential investigative themes in streaming Initial focus S pa ce Individual Potential • Topics of discussion?• Performances of selves themes and • Uses of space • Extending networks issues in • Users of space • Codes and shared streaming meanings in • Interactions, interactions including collaborations and • Transactions and conflicts reciprocity
• • • •
Phenomenon Key actors/agencies involved in creation of phenomenon Phenomenon's characteristics Evolution of phenomenon Conflicts associated with phenomenon development
Such investigative techniques can be used within exploratory studies, which have individual value (cf. Stebbins, 2010); but, they may also be deployed within broader data mining exercises. IRI can be used to identify both key 'concepts' (e.g. types of behaviour, relationships, venues, (sub)cultural codes etc.) and 'contexts' (i.e. specific sites or other online sources of data) (Lau et
33
al., 2010). These can then be analysed quantitatively to discover new patterns and test hypotheses (ibid).
It is important to acknowledge the challenges and limitations associated with the investigative approaches discussed here. Engaging in unobtrusive research limits the data to that which occurs naturally. It also raises a number of ethical questions, concerning covert research, invasion of privacy and the absence of informed consent (cf. Lugosi, 2006). Even if greater interaction is sought, engaging with active users has fundamental sampling biases, and although online investigative research can provide access to many sections of the population that may otherwise remain invisible, it still cannot account for how 'passive' users receive or react to online material. Finally, as with many other types of qualitative research, online investigations may be criticised for their lack of generalisability, although multiple investigative studies can be 'concatenated' (Stebbins, 2010), and, as noted above, IRI can become the first stage in a confirmatory, quantitative exercise.
Regardless of whether it is used within an exploratory study, or as part of a broader data mining exercise, we have argued that engaging in this type of online research has broad applications in examining the experiences of employees, consumers and in developing human resource, marketing and operational strategies. Moreover, as the cases highlight, such research offers hospitality academics opportunities to contribute to broader disciplinary and policy debates. The ubiquity of the internet and the access it provides to data will undoubtedly lead to its increasing use in the future among academics. 34
However, a remaining challenge, certainly for the readership of this journal, is to translate these investigative techniques so commercial operators can also explore and exploit the internet to develop sustainable business practices.
7. References
Beaven, Z. and Laws, C. (2007), "'Never Let Me Down Again': Loyal customer attitudes towards ticket distribution channels for live music events: A netnographic exploration of the US leg of the Depeche Mode 2005-2006 World Tour", Managing Leisure, Vol. 12, pp. 120-142.
Dholakia, N. and Zhang, D. (2004), "Online qualitative research in the age of e-commerce: Data sources and approaches", Forum: Qualitative Social Research, Vol. 5 No. 2, Art. 29, available at:http://nbnresolving.de/urn:nbn:de:0114-fqs0402299 (accessed 10 August 2010).
Falzon, M. A. (2009) (Ed.), Multi-Sited Ethnography, Ashgate, Abingdon.
Fielding, N. G., Lee, R. M. and Blank, G. (2008) (Eds.), The Handbook of Online Research Methods, Sage, London.
Grossnickle, J. and Raskin, O. (2001), Handbook of Online Marketing Research, McGraw-Hill, Blacklick, OH.
35
Guimarães, M. J. L. (2005), "Doing anthropology in cyberspace: Fieldwork boundaries and social environments", in Hine, C. (Ed.), Virtual Methods: Issues in Social Research on the Internet, Berg, Oxford, pp. 141-156.
Hannam, K., Sheller, M. and Urry, J. (2006), "Editorial: Mobilities, immobilities and moorings", Mobilities, Vol. 1, pp. 1-22.
Hewson, C., Yule, P., Laurent, D. and Vogel, C. (2003), Internet Research Methods: A Practical Guide for the Social and Behavioural Sciences, Sage, London.
Hine, C. (2000), Virtual Ethnography, Sage, London.
Hine, C. (2005) (Ed.), Virtual Methods: Issues in Social Research on the Internet, Berg, Oxford.
Janta, H., Ladkin, A., Brown, L. and Lugosi, P. (2011), "Employment experiences of Polish migrant workers in the UK hospitality sector", Tourism Management, Vol. 32, pp. 1006-1019.
Jones, S. G. (1998), "Information, internet, and community: Notes toward an understanding of community in the information age", in Jones, S.G. (Ed.), Cybersociety 2.0: Revisiting Computer-Mediated Communication and Community, Sage, London, pp.1-35.
36
Kozinets, R.V. (2002), "The field behind the screen: Using netnography for marketing research in online communities", Journal of Marketing Research, Vol. 39, pp.61-72.
Kozinets, R. V. (2010), Netnography: Doing Ethnographic Research Online, Sage, London.
Langer, R. and Beckman, S. C. (2005), "Sensitive research topics: Netnography revisited", Qualitative Market Research: An International Journal, Vol. 8, pp. 189-203.
Lau, K.N., Lee, K.H. and Ho, Y. (2010), "Text mining for the hotel industry", in Enz, C.A. (Ed.), The Cornell School of Hotel Administration Handbook of Applied Hospitality Strategy, Sage, Thousand Oaks, CA, pp. 66-84.
Leander, K. M. and McKim, K. K. (2003), "Tracing the everyday 'sitings' of adolescents on the internet: A strategic adaptation of ethnography across online and offline spaces", Education, Communication and Information, Vol. 3, pp. 211-240.
Lofland, J. and Lofland, L. H. (1995), Analysing Social Settings: A Guide to Qualitative Observation and Analysis, 3rd ed., Wadsworth, Belmont, CA.
37
Lugosi, P. (2006), "Between overt and covert research: Concealment and revelation in an ethnographic study of commercial hospitality", Qualitative Inquiry, Vol. 12, pp. 541-561.
Lugosi, P. (2009), "Ethnography, ethnographers and hospitality research: Communities, tensions and affiliations", Tourism and Hospitality: Planning and Development, Vol. 6, pp. 95-107.
Lugosi, P., Bell, D. and Lugosi, K. (2010), "Hospitality, culture and regeneration: Urban decay, entrepreneurship and the 'ruin' bars of Budapest", Urban Studies, Vol. 47, pp. 3079-3101.
Lynch, P., McIntosh, A. and Tucker, H. (2009) (Eds), The Commercial Home: An International Perspective, Routledge, London.
Markham, A. N. and Baym, N. K. (2009) (Eds), Internet Inquiry: Conversations about Method, Sage, London.
Miller, T. W. (2005), Data and Text Mining: A Business Applications Approach, Pearson, Upper Saddle River, NJ.
Patton, M. (2002), Qualitative Research and Evaluation Methods, 3rd ed., Sage, Thousand Oaks, CA.
Pink, S. (2007), Doing Visual Ethnography, 2nd ed., Sage, London. 38
Stebbins, R.A. (1992), Amateurs, Professionals and Serious Leisure, McGillQueens's University Press, Montreal.
Stebbins, R. A. (2010), "The Internet as a scientific tool for studying leisure activities: exploratory Internet data collection", Leisure Studies, Vol. 29, pp. 469- 475.
Strauss, A. and Corbin, J. (1998) The Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory, 2nd ed., Sage, Thousand Oak, CA.
Urry, J. (2007), Mobilities, Polity Press, Cambridge.
Watson, P. Morgan, M. and Hemmington, N. (2008), "Online communities and the sharing of extraordinary restaurant experiences", Journal of Foodservice, Vol. 19, pp. 289-302.
Wittel, A. (2000), "Ethnography on the move: From field to net to internet", Forum: Qualitative Social Research, Vol. 1 No. 1, Art. 21,http://nbnresolving.de/urn:nbn:de:0114-fqs0001213 (accessed 10 August 2010).
39
doc_599677782.docx
Market research is any organized effort to gather information about markets or customers. It is a very important component of business strategy.
Case Study for Investigative Management and Consumer Research on the Internet
Abstract This paper introduces the notion of Investigative Research on the Internet (IRI) and conceptualises its processes through the principle of streaming. It discusses the similarities and differences between IRI and netnography and considers various aspects of the IRI process, including site selection, sampling, data collection and analysis. It is argued that streaming can help to understand the processes involved in conducting netnographic research. Moreover, it is suggested that streaming is a more appropriate way to conceptualise some internet-based studies that do not conform to netnographic or ethnographic ideals. Three international empirical cases are used to illustrate the application of IRI and streaming in research on international workers, consumer cultures and on emerging business phenomena.
Keywords: Internet, online, virtual, ethnography, netnography, qualitative research, streaming analysis
2
1. Introduction
The development of web 2.0 and interactive technologies offer numerous opportunities for qualitative management and consumer research. Researchers may use instant messaging services and chat rooms (virtual spaces facilitating synchronous communication) to conduct interviews and focus groups (Fielding et al., 2008; Grossnickle and Raskin, 2001). These enable fieldworkers to collect data through real-time interaction, although there are various other sources that facilitate asynchronous data collection, focusing on archived material (Hewson et al., 2003; Dholakia and Zhang, 2004). This paper advances existing knowledge on online methods by introducing the notion of investigative internet research, or Investigative Research on the Internet (IRI), and distinguishing such qualitative, exploratory studies from other forms of online research. It is argued that IRI shares many of the characteristics of netnography (Kozinets, 2002) and other forms of online ethnography (Hine, 2000, 2005), but that not all IRI can or should be called netnography. We conceptualise the processes involved in such internet-based research using the notion of streaming, which can inform future netnographic studies as well as other IRI that does not conform to ethnographic ideals. Finally, the paper demonstrates, through three international empirical cases, how IRI and streaming can be applied in research on international hospitality workers, consumer cultures and networks, and more broadly on emerging hospitality business phenomena.
3
We concentrate on three sources of data: 1) message boards or forums, which enable users to engage in asynchronous exchanges or postings on different topic threads; 2) blogs, narratives and opinion pieces written by individuals, to which others may respond and which may also provoke ongoing asynchronous exchanges; and 3) websites provided by commercial and non-commercial organisations or individuals, offering news, reviews and either lay or professional commentary. We discuss the investigative processes, including navigation in and across multiple 'field sites', sampling, analysis and interpretation, through which researchers can exploit these three sources of data in their studies.
The paper contributes to knowledge in several ways. Firstly, we illustrate how internet research and streaming can be applied to address issues relevant to hospitality managers, academics working in broader disciplines, alongside policy makers. Secondly, and more broadly, it uses research on hospitality to conceptualise qualitative, investigative, internet-focused research that is applicable to other intellectual contexts. Consequently, rather than simply importing concepts from other disciplines, hospitality research creates opportunities to contribute to wider fields of inquiry.
2. Conceptualising investigative research on the internet
It is important to begin by distinguishing IRI from other types of internet-based methods of data collection, for example, text mining. These approaches use software that searches the internet and attempts to identify patterns and 4
relationships between textual information (Miller, 2005). IRI may use specialist software, and it will inevitably employ search engines; however, it does not rely on textual material alone, and the researcher has a much greater role in sourcing, analysing and making connections between sources of information.
IRI has a number of features. Firstly, as the term investigative implies, it is inherently analytical, exploratory and seeks to simultaneously discover and construct. This implies that the research process attempts to unearth existing pieces of information, which emerge in a variety of ways, for example through objects, actions, images, texts, relationships, interactions, spaces and institutions. Moreover, the researcher or investigator connects these to other pieces of data that may not seem immediately related. Secondly, the research act is a dynamic process, responsive to changes in the fieldwork 'site', but also requiring sensitivity towards the individuals who provide information, towards relationships and interactions during the fieldwork, and towards the various sources of data, which may not be predetermined at the study's outset. Thirdly, because the research is dynamic, it also has to be context sensitive, taking account of the subtle idiosyncrasies of the virtual spaces through which researchers navigate. Researchers also have to be mindful of the social, cultural and sub-cultural factors entangled in the offline worlds in which data emerges. Finally, as noted above, in IRI the researcher is not a neutral entity, but is central to the research process, making decisions at every moment about what is relevant, what is excluded and which lines of inquiry are pursued. IRI thus shares many of the features of online 5
ethnography (Hine, 2000; Kozinets, 2002; Wittel, 2000); consequently, it is useful to discuss these research approaches in further detail, before introducing the notion of streaming to conceptualise the processes of conducting IRI.
3. Ethnography and netnography
Our discussion primarily draws on and engages with netnography, rather than other conceptualisations of internet-based ethnography (e.g. Hine, 2000; 2005), partly because Kozinets (2002; 2010) has attempted to define its processes in detail, but also because the concept clearly distinguishes it from traditional ethnography. Moreover, netnography, which was originally developed for marketing and consumer research, is rapidly becoming an accepted concept in business research, increasingly being applied to the study of hospitality (e.g. Watson et al., 2008).
Netnography, like its antecedent, ethnography, is not a single method or approach; it involves a range of investigative strategies. The lack of space prevents a detailed discussion of ethnography, but it is useful to briefly consider some of its key features, which helps to understand netnography's underpinning principles. Lugosi (2009), synthesising earlier discussions, suggests ethnography seeks to gain an emic understanding of groups and individuals through immersion in their activities over extended periods of time. The emerging insights are context sensitive and embrace the complexity of human phenomena. The researcher, rather than being a detached neutral 6
entity in the research process, has a central role in knowledge generation. This means fieldworkers have to be critically sensitive about their relationships with informants and the data they develop. Ethnographers often use some variation of participant observation, although this is often complemented by interviews, the analysis of written material, and, increasingly, by visual methods (see e.g. Pink, 2007). Historically, the insights gained through ethnography have been communicated through rich textual accounts, although these, too, have been enhanced by the use of technology and visual methods.
Netnography also attempts to understand, rather than measure, but Kozinets (2010, p.5) highlights several features that distinguish it from ethnography. Firstly, interaction is different than in 'traditional', face-to-face ethnography. Engagement can be with a much wider range of geographically dispersed people, and participation can take a number of different forms in online environments. Data can be gathered without any interactions but engagement may also range from brief exchanges to longer and deeper interactions, using a variety of communications e.g. text, sounds, images and other 'virtual' expressions such as emoticons. Secondly, rather than relying on field notes and creating textual accounts, netnography gathers and uses a wider range of data, which varies in terms of focus and quality. This means the researcher has to broaden her or his repertoire in gathering, interpreting and integrating vastly different data. These are themes echoed by others distinguishing between traditional and virtual ethnographies (e.g. Wittel, 2000).
7
Kozinets (2010, p. 89) outlines six criteria for selecting sites for study, arguing that they should be 1) relevant to the research focus and questions, 2) active, with recent and regular communications, 3) interactive, having flows of communication between participants, 4) substantial, in terms of numbers of users, 5) heterogeneous, involving different participants, and 6) data-rich. Kozinets (2002) also proposes a number of procedures to facilitate the process of conducting online research and identified several stages: entry into the 'community', data collection utilising a variety of textual and visual material, analysis and interpretation, as well as member feedback, which requires an overt approach. However, this was challenged by Langer and Beckman (2005) who suggest adopting a covert approach and Beaven and Laws (2007) who propose conducting observational or passive netnography. Informed consent is seen to be unnecessary when there is an openly available archive of online material or when forums are publicly accessible (Beaven and Laws, 2007).
Kozinets' clear guidance helps to shape the direction and processes of online research, but there remain several areas of debate for fieldworkers, which make it difficult to think of all IRI as netnography. The first concerns the centrality of community in netnographic research. The notion of community has continued to evoke critical debate, and there have been extensive discussions about its appropriateness in understanding the multiple relationships that exist in cyberspace (see Jones, 1998). Contemporary debates have shifted away from traditional conceptions of community, based, for example, on structures and stable roles, to more fluid ones, suggesting 8
that people continually (re)construct and perform notions of shared identity and belonging (Guimarães, 2005).
Kozinets (2010) acknowledges the problematic nature of 'community', but maintains that the focus of netnography is on 'online communities', i.e. groups constructed and maintained in virtual worlds. He also advocates the increased use of netnography, in conjunction with traditional forms of ethnography, to study 'communities online', i.e. groups that exist offline, but also interact through the web. According to Kozinets (2010), communities involve a minimum number of people, who identify with others in the community and engage in ongoing, meaningful interactions with other members. This can be used to describe some of the social configurations and interactions relevant to hospitality management research, but it is problematic to designate individuals with shared interests or involvement in particular work and leisure activities as being part of a community.
Contemporary ethnographers, engaged in 'multi-sited' fieldwork, have maintained that ethnography continues to offer insights into the complex sets of actions through which individuals construct and articulate relationships and notions of self/other (see e.g. Falzon, 2009; Guimarães, 2005). Extended engagement with multiple 'informants' provides opportunities to understand behaviours, values and norms that enables researchers to define and thus construct individuals as belonging to a community. Netnography, and ethnography in and of virtual environments, therefore becomes a productive as well as an analytical act. Consequently, if we acknowledge the central role 9
of the researcher in constructing a notion of community, then we have to recognise that the fieldworker's version of community may not be shared by the people being represented.
A final challenge, which relates to the previous point, concerns the implicit assumptions about the nature of knowledge and understanding gained through netnographic research. The central role of the researcher, and her or his extended engagement with individuals offering insights into the nuances of interactions, norms, values and identities, is one of the defining characteristics of both netnography and ethnography (Hine, 2000, 2005; Leander and McKim, 2003). Netnography, like ethnography, thus stresses a particular epistemological ambition in assuming that immersion into a group and their culture can help fieldworkers develop emic perspectives. However, if notions of community are fluid and highly contestable, and engagement with informants may be fragmented and asymmetric, not all the interactions and relationships can provide detailed ethnographic data, a point acknowledged by its practitioners (Hine, 2000, 2005; Markham and Baym, 2009). Moreover, not all IRI may set out to or claim to have gained such ethnographic insights, and describing such research as netnography is therefore potentially problematic.
Calling all IRI 'netnography' is an example of concept creep - a simultaneous extension and reinterpretation of netnography. Kozinets (2002), Hine (2000) and others recognise that netnography, and other forms of online ethnography, are flexible approaches, involving a range of methods and 10
techniques, and are thus open to critical reconsideration. This paper engages in the reinterpretation of netnography by proposing the notion of streaming to help understand the processes involved in such internet research. However, IRI and streaming offer a broader conceptualisation of qualitative, exploratory internet-based studies. This is necessary and useful for two reasons: firstly, rather than simply, and uncritically, inflating the netnography concept, this paper defines a key aspect of such online research, thus helping researchers conduct netnographic studies. Secondly, distinguishing the processes involved in IRI from the concept of 'netnography' helps researchers to understand and conduct research via the internet, which may not conform to ethnographic ideals.
4. Investigative research on the internet as streaming
We use 'streaming' as a metaphorical device to conceptualise the data gathering and analysis processes in which researchers trace relationships between actors and agencies through the internet. The notion of streaming may initially seem to be limited, as is any other metaphor: it may create an image of a defined path to be followed and imply a singular direction, something which does not reflect well the multi-spatial and multi-directional nature of the internet and the movement of its users. Streaming echoes Hine (2000) and Leander and McKim's (2003) focus on mobility and stresses the fluidity of relationships and interactions. Streaming should be thought of as a dynamic, creative process, when the researcher opens new directions for data gathering and analysis as he or she moves through and across different 11
sites. As ethnographers working in multi-sited environments have argued (Hine, 2005; Leander and McKim, 2003; Falzon, 2009; Wittel, 2000), fieldworkers connect, perform and effectively construct the field site. Even if they establish one set of connections, thus creating a 'stream' through which they move, they can retrace their movement, move backwards, identify new divergences and create alternative connections - streams.
At the core of streaming is the notion of connectedness and the importance of mobility. Urry (2007) is one of many commentators to stress that contemporary society is characterised by increased movement both as people move around physically but also as information circulates rapidly, facilitated by the growing prevalence of technology (see also Hannam et al., 2006). Consequently, individuals are less likely to be members of stable social groups, based on strong social ties; instead, people are connected, often through loose ties, to a wide range of other people through networks that stretch across geographical boundaries (Urry, 2007). Streaming can thus be thought of as a research aim to follow the networks of relationships through which information about people, places and experiences flows. Just as Hine (2000, p.35) noted, rather than remaining on a single site or focusing on one context, internet research should use movement between sites as an analytical device. The key issue is that researchers attempt to understand how relationships and flows of information operate through virtual media. Movement across different virtual spaces in data gathering and analysis also implies that streaming is more likely to examine interactions and
12
representations during extended periods and across several time periods, rather than only at a single point in time.
A further feature of streaming is its inductive nature, involving simultaneous data collection and analysis (Lofland and Lofland, 1995). Concurrent data gathering and open, inductive analysis are necessary as representations and interactions occur across multiple discussion threads or even websites. In many cases, even when interactions take place on a single forum, they include links to other sites. Examining relationships and interactions through virtual networks inevitably leads to choices that researchers make about which sites, texts, discussions or representations of people, places, events or individuals should be traced further. These have the potential to become what we refer to as key streams, which lead to new lines of inquiry. Within traditional ethnography, 'key informants' have often played a crucial role in gaining access, mediating on fieldworkers' behalf and providing information. The notion of key streams collapses the distinction between people, sites and texts, as they all have the potential to be crucial reference points researchers may draw upon once, but may return to repeatedly, to gain insights. There may also emerge more minor streams, located elsewhere, that are also useful but less so than key ones.
Streams have an elaborative function, providing illuminating illustrations of behaviours, events, beliefs, values etc., thus helping to understand the dimensions and dynamics of particular phenomena. They also have a connective function, linking to further sites, documents, representations etc. 13
where issues can be explored further. It is also useful to distinguish between key streams, which are the primary sources from which streaming begins and sub-streams that may exist within, or extend from, these key streams. These may be the discussion threads on a site, or representations by individuals e.g. in the form of personal narratives, informal commentaries, or lengthier, formal discourses, identified by the researcher as relevant based on the examination of another site or text.
Finally, the process of following posts or threads and moving between different websites also helps to identify a final key feature of streaming, which is that it involves snowball sampling that may be complemented by other forms of criterion sampling, within a broader purposive sampling approach (see Patton, 2002). Internet interactions are driven by user generated material, and users can create new linkages between other users, sites and threads of discussion; therefore, it is important, in principle, to let the data and the emerging themes shape the sample size, rather than it being defined too narrowly at the outset of the research. Ideally, data collection will stop when the researcher arrives at a point of theoretical saturation, when no new themes emerge from the data (Strauss and Corbin, 1998). However, the specificity of the study's aim and objectives and resource restraints, for example in terms of funding, time, project life etc. will often impose other limits on the study. Consequently, the research may come to focus principally on a specific forum, a limited number of threads - based on some research question driven criteria, a fixed time period, a limited number or network of users or even an individual, which is used as a particular case. 14
5. Streaming in practice
5.1 Streaming in research on employee experiences
Background and purpose Migrant workers make up a large part of the hospitality workforce and their experiences are frequently shaped by their immigrant status, language skills and their ability to mobilise social capital: they are frequently confined to low paid, low status jobs and hospitality offers particular career pathways, while denying others (Janta et al., 2011). The first study used to illustrate streaming is part of a wider project examining Polish migrant employees' experiences in the hospitality and tourism sector. The study applied a mixed-methods approach to allow an understanding of respondents' lives at work and beyond it, including their interactions, aspirations, work conditions, job scopes and demographic profiles. However, we concentrate here on the qualitative internet research dimension of the study. A key objective of the research was to examine migrant networks and how networking, in virtual environments in particular, mediated their employment experiences. More specifically, it sought to understand how migrants supported each other and represented their experiences of mobility and hospitality employment to other current and future migrants.
Data collection and streaming There is a high usage of social networking sites and discussion forums among Poles in the UK (Janta et al., 2011), and the public nature of these 15
forums prompted the researcher to engage in observational or passive netnography (Beaven and Laws, 2007). Streaming utilised purposive sampling and began by identifying the key streams, using the six criteria, outlined earlier, to select appropriate sites. Initially, the site's traffic and activity were prominent criteria. Three important sites were identified: mojawyspa.pl (my island) with 23,000 threads, ang.pl, a site dedicated to those learning English, with more than 72,000 threads, and gazeta.pl - the biggest forum with more than 7,400 different forums on it and more than 102 million written posts. These became the key streams, and the starting point for streaming was a specific virtual space - i.e. the forum, from which networks of relationship and interactions were traced.
Within these key streams it was then necessary to identify sub-streams, which were the relevant message boards and threads, found by analysing headings and, where possible, by using the sites' internal search engines. At the level of sub-streams, essential sampling criteria were increasingly the relevance of posts to the research objectives rather than just the traffic. Examples of topics included 'job seeking' and 'applying for positions', 'hotel work' and 'other jobs in hospitality'. The relevance of some streams was obvious, for example: 'Working in a hotel? Discuss it!' or 'Why are educated Poles washing pots in the UK??' on gazeta.pl or 'Manchester-waiter-a few questions
' on mojawyspa.pl. However, other titles were less obvious, for example 'Are you planning to return to Poland one day?' on gazeta.pl and others discussing migrants' future plans, but these made repeated references to hospitality work. Approximately 100 existing threads, launched between 162004 and 2008, were downloaded, printed and analysed in this research. Some continued for only a few days while others until 2010.
Key emerging questions that drove the processes of streaming, as noted in the previous paragraph, concerned the thread topics and subsequent interactions. Examining the threads led to other relevant issues and helped answer such questions as who uses the space and for what purpose. For example, a thread launched on gazeta.pl, entitled 'How is it getting a job after returning to Poland?', had an intensive beginning, with 95 posts within 3 days (11-14.04.2006). Tracing this stream helped to understand the development of migrant careers over extended periods, including the experience of return and the possibility to utilise the human capital developed during work in the UK. Other streams showed how migrants interacted, building their social networks. For example, a thread 'Anyone going to Bournemouth?' on ang.pl was launched in 2005 and the last post was written in January 2010.
Indicative emerging themes identified through streaming Streams of inquiry helped to understand specific aspects of instrumental and emotional support directly related to hospitality employment. For example, one reply to a posting entitled 'Interview - waitress - on Friday help me!!!' (ang.pl, 27.05.2008) provided twenty examples of possible interview questions such as 'What do you like about waitressing?' and 'Are you good at multi-tasking?' Other aspects of support revealed through tracing key substreams incorporated reflections on hospitality work. Examples included streams entitled 'Why are qualified Poles washing pots in the UK?' 17
(02.08.2006), which discussed the experience of having to take jobs for which they were overqualified. Others were less philosophical and more prescriptive, such as the stream entitled 'Don't come! No work here!', where posts revealed the bleak situation of Polish migrants: 'Only those who have worked in the UK before and those who speak English well will find a job even such as 'on the sink' [i.e. low status kitchen porter positions] - this is what it looks like now!' (gazeta.pl, 30.01.2007).
Not all the postings were so negative; some sub-streams also revealed information about migrants' positive work-related learning experiences. For example, a thread entitled: 'A waiter in England' (gazeta.pl, 25.11.2004), started by a user seeking advice on learning English while working in foodservice, included a number of constructive postings by other users:
It is hard work, especially when it gets busy and you need to deal with a number of things at the same time. But it gets better, for example when they ask you where you come from and they don't want to believe that you come from a non-English speaking country and they appreciate your English
Another reply, posted on the same day, also revealed themes of emotional support from fellow migrants and positive work-based encounters. It also helped identify further streams of inquiry concerning the role of formal education in migrants' experiences:
18
Don't give up! I don't know why [the other site users] are scaring you! I work behind a bar, maybe it's not the same as being a waiter but it is not simple either! (sometimes) I have been learning English for 4 years and I don't have a FCE [First Certificate in English] (yet) but I don't have problems in communicating. So don't put him off! People are really understanding!
Significance, applications and contributions to knowledge Streaming, using particular online forums as initial points of focus, was important in meeting the study's objectives because these were spaces of interaction dedicated to a particular social group. However, these spaces were used by individuals from a wide range of geographical areas and it would have been very difficult to reach such a disparate network of individuals through conventional methods. In some cases, sites, and threads within them, concentrated on specific geographical areas or activities e.g. hospitality work or language learning, which helped to focus in on specific subgroups, based on location or activity. Finally, streaming was useful because it helped to follow particular lines of enquiry extending from these key streams.
By conducting internet-based research on this and related topics, through streaming, hospitality academics can contribute to knowledge outside their subject areas, thus helping to raise the profile and influence of the research community. More specifically, migration has become a sensitive political 19
topic, especially in the current economic climate, so understanding their experiences, for example relating to adjustment, learning, access to work and social inclusion/exclusion has clear implications for policy makers regarding the type of migration that should be encouraged and how migrants can be supported to make positive contributions to society. Furthermore, understanding how migrants perceive their work, understanding their career aspirations, appreciating their processes of learning, particularly through hospitality work has clear relevance to practitioners who devise human resource strategies concerning recruitment, training and development as well as retention.
5.2 Streaming in research on consumer experiences
Background and purpose Recent years have seen the development of 'foodie' culture and the rise of enthusiastic consumers using the internet to communicate their meal experiences and their passion for food (cf. Watson et al., 2008). Blogs describing restaurant visits are narratives that weave together a complex set of events into a coherent story and thus help to understand the consumer experiences restaurants offer. Such forms of consumption and their representation through blogging can be considered a form of serious leisure (Stebbins, 1992), which is used by people to construct their identities and build communities of consumption (Watson et al., 2008). Many food bloggers are examples of Stebbins' (1992) skilled amateurs: they become self-aware participants in their chosen activity and develop the ability to 20
articulate their emotions and express their opinions. The second case illustrating streaming is a study of blogging, with particular reference to meal and restaurant experiences. Given the central role of identity construction and communities of consumption within food blogging, a key objective of the research was to understand how internet users articulate their identities and engage with networks to mediate their food related experiences.
Data collection and streaming This study also adopted a passive or observational form of online research, although permission was sought from the blogger to study the site. Streaming and the identification of key streams began with a broad search of the internet using 'food blogs' as the core search term. From the millions of returns, irrelevant ones, for example those focusing primarily on personal cooking, were rejected. As in the previous case, the sampling for key and sub-streams was purposive, and a specific blog was selected because it was substantial, relevant and active. The key stream, Grab Your Fork (grabyourfork.blogspot.com/, referred to hereafter as GYF) was established in 2004. The blogger, Helen Yee, posts regularly and responds to each comment individually. GYF won of the 'Best Restaurant Review Blog' award in 2006, and was a finalist in 2004, 2005, 2007 and 2008 in various categories for food blog awards. In addition, it was listed fourth in a worldwide top 10 food blogs by The Times in 2008. Yee is based in Sydney, Australia but comments are made by people from all around the world. The study focused on posts made over 24 months during 2006 and 2007. In total this covers 460 posts plus 2774 comments by other visitors, but 15 key sub21
streams discussing 'memorable meals' were identified within this process of streaming. One example of these key sub-streams is her description of a visit to Sydney's famous Tetsuya's restaurant in April 2005, which received 51 comments, the latest as recent as September 2009.
In this study the starting point for streaming was the individual, Helen Yee, and the study traced her experiences, interactions and relationships extending from her blog. Key themes and emerging issues driving, and emerging from, streaming included presentations of self, particularly as they are articulated through her descriptions of food and eating. Presentations of self were also important issues in other users' posts. Additional important themes were her extending networks, the reciprocal transactions involved, and the shared codes and meanings in interactions.
Indicative emerging themes identified through streaming The construction of identity and articulations of self were central features of food blogging and Yee was candid about her passion for food and why she blogs: 'An unbridled commitment to share, to comment and to espouse the joys of food are the main reasons why many of us blog' (Helen Yee, GYF, 27. 01.2010). Her commentaries also revealed other aspects of her identity, for example, her use of English points to her constructed class status: 'The roast suckling pig embodies all three in glorious gormandizing hedonistic glory' (Helen Yee, GYF, 02.02.2006). Moreover, her descriptions of eating offal and other animal parts reveal information about her ethnicity and how it defines food habits, for example, 'if there are three particular idiosyncrasies 22
about Asian palates it's our love of bones, skin and fat' (Helen Yee, GYF, 02.02.2006).
Beyond Yee's commentaries, it was useful to trace her interactions through her blog to better understand how communities of consumption were articulated through online interaction. There are numerous visitors to Yee's blog, some simply 'lurking' (passively observing rather than participating) or making transient visits, but she has developed relationships with several posters. Between them they have established interactional repertoires and codes. For example, they appear to be familiar with technical culinary terms, including 'galette', 'confit' and 'quenelle', which are used without explanation. Moreover, Yee and followers of the blog regularly perpetuate interactional routines with comments and descriptions of their meal experiences or of reading other people's meal experiences:
Sorry I meant "yee sang" in my earlier comment. It's a bit hard to type and drool at the same time as you know. (Oslofoodie, GYF, 02.02.2006)
Tracing the interactions, which occur in physical and virtual settings, also reveals the potential role of Yee and her extending network as informed critics who influence purchasing decision making:
[...] next time try the wat tan hor and fried radish cake which are both very good and if having the ice kacang make sure 23
you ask for them to put enough rose water. you'll see they put corn in there's [sic]. some pics of the dishes i've had are available to look at [link attached] in my food slideshow. (SimonLeong, GYF, 16.10.2006)
Comments such as this, from 'SimonLeong', not only identify subtle details of the meal experience, they also draw upon, and reveal, aspects of their own identity. References to heritage and personal experience are thus entangled with commentary on various aspects of the service operation:
I am Burmese and new to Sydney. I have been to several other Burmese restaurants in Asia, UK and US, and I must stay I was rather impressed with the food here. It is authentic indeed and the service from Victor and his staff were [sic] impeccable. (Cho, GYF, 16.10.2006)
Furthermore, as the quotes from 'SimonLeong' above and 'fooDcrazEE' below highlight, interactions between Yee and other blog followers provide links to further sites and discussion threads, which helps the researcher illustrate key points, identify other users and trace further streams.
What a feast! The YEE SANG is different from Malaysia though. Take a look here for a post of Malaysian Yee Sang. [Link attached] (fooDcrazEE, GYF, 02.02.2006)
24
Significance, applications and contributions to knowledge A streaming approach, in this case taking the individual blogger as the starting point from which to extend investigations, was central to meeting the study's objectives for several reasons. Firstly, focusing on a committed skilled amateur, such as Yee, provides a rich starting point for examining how individuals express notions of identity through serious leisure. Yee is an extreme case because of her nuanced appreciation of foodways and her commitment to expressing her engagement. Following Yee thus helps to generate a wide set of conceptual themes that are useful when examining the behaviour of other skilled amateurs. Secondly, Yee is a common point of reference, which other similar enthusiasts use when articulating their own identities and expertise. Linked to this, Yee's blog is an expressive space where performances of self are mediated and others can develop interactional routines and connect to a broader social network of likeminded individuals.
Applying this technique to such social activities also has a number of implications for academia and practice. An analysis of food blogs, again, provides opportunities for hospitality academics to make contributions to knowledge in other fields, for example in leisure studies about emerging forms of serious leisure, or to social scientific debates about identity, networked sociality and its relationship to food. Researching food blogging also has implications for hospitality marketing and operations management. Key individuals in networks disseminate information, shape tastes and influence purchasing decisions. Therefore, identifying influential individuals, 25
like Helen Yee, understanding their motivations, modus operandi and relationships with other consumers provides clear marketing opportunities. Moreover, their critical commentary on decor, food and service gives managers insights into consumer perceptions of their commercial propositions.
5.3 Streaming in research on emerging hospitality trends and phenomena
Background and purpose The final illustration of streaming takes as its focus an emerging trend or phenomenon in commercial hospitality. This is the most ambiguous of all three types of streaming because trends may actually lack clear definition, and exploratory research offers a way to clarify them and their dynamics. The use of streaming to conceptualise the research process is particularly appropriate in these cases because, although it may be informed by an ethnographic sensibility and desire to gain emic insights, it is problematic to call it netnography. Emerging phenomena are amorphous in nature and there may not be a clear community to which they can be ascribed.
Some hospitality phenomena such as 'homestay' (Lynch et al., 2009) have evolved to a stage where their manifestations in different international contexts are easy to identify. However, there are many other emerging ones that are only now being investigated and understood. One such phenomenon are 'rom' and 'kert' venues, which developed in Budapest, Hungary in the 26
beginning of the 21st century. The term 'rom', meaning 'ruin' in Hungarian, refers to hospitality venues operating in dilapidated buildings in Budapest. 'Kert' means 'garden' and was used in reference to outdoor venues. The two terms were frequently used in combination as rom were located in the open courtyard of venues, but the two should be treated as separate categories of hospitality operation. No research had been conducted on these types of operation so a key objective of this study was to understand what the 'propositions of hospitality' were in these venues and how this genre of hospitality emerged.
Data collection and streaming Tracing the evolution of the rom/kert venues helps to illustrate streaming that may, at various stages, involve focus on individuals and specific places, but ultimately focuses on the phenomenon. Initial broad investigations began with general internet searches on commentary that included references to one venue, the Szimplakert, which was considered by many commentators to be pioneering. Moreover, it was important to examine what other venues were discussed in relation to or in reference to Szimplakert. Identifying similar venues and descriptions of them helped to understand these venues, not as individual operations, but as points of reference in a constellation of venues.
An important aspect of streaming was identifying key streams, which included cultural commentators and virtual outlets where commentary was located. Three key streams were identified, again, though the focused use of search engines. Index.hu is an established media site, which has a number of writers 27
producing regular features on culture, politics and urban affairs. One commentator, András Földes, wrote several articles for Index.hu on the emergence and growth of the rom/kert phenomenon. He also collaborated on a guide, distinguishing venues according to customers' 'neediness'/'needlessness', and discussed elsewhere the struggles venues faced in the 2005 summer season. Szimpla also maintained a press centre on their website, which stored scanned copies of articles about the Szimplakert that usually discussed it alongside other, related venues. Index.hu, Földes and the Szimpla repository, cross-referenced each other and became the key streams used to trace the evolution of the rom/kert phenomenon.
Key sub-streams were subsequently identified. These were used to pursue the emerging themes in the research including the venues' defining operational characteristics, variations on the rom/kert model, and the collaborations and conflicts between different stakeholders influencing venues' survival. Importantly, the information gained from Földes, Index and Szimpla's archived material helped to understand not only the venues as individual commercial entities but how they existed in relation to their social, cultural, physical and political environments.
Additional important streams were traced through other sites and individual blogs that made reference to venues and to the rom/kert phenomenon. For example, professional reviews and lay commentaries made repeated reference to particular design features and cultural events hosted in the 28
venues. Some websites contained a single article, followed by comments from different users giving their own accounts of venues. The small number of active users, who engaged in dialogue, made it problematic to treat such idiosyncratic commentaries as authentic representations. Therefore, as Hine (2000) suggests, it was necessary to consider individual commentaries and representations in relation to others, which may not directly acknowledge, engage with or respond to them. Streaming, using the phenomenon as the focal point from which to extend inquiry was therefore necessary in constructing a rich story of the rom/kert phenomenon. Interpreting different representations in relation to others also became a form of triangulation. Several attempts were made to contact some of the bloggers and writers who provided insightful accounts, but none were successful. Nevertheless, the study was augmented by visits to venues, the collection of photographic data, and recorded interviews with operators.
Indicative emerging themes identified through streaming The emergence of this hospitality genre can be attributed to other venues operating in Budapest in the 90s, but commentators agree that a pioneering example of a romkert was the Szimplakert, which opened in 2002. The first Szimplakert was housed in the courtyard of a building awaiting demolition. This venue, including the use of courtyards of empty residential buildings, became the template which was reproduced, albeit slightly differently, in subsequent years. In 2003, two similar venues opened in the district; and by 2004 numerous operators had created their own version of the rom/kert. More recently, numerous other venues have opened in surrounding city districts 29
and several have opened on rooftops rather than in courtyards (see Lugosi et al., 2010).
Tracing different representations of venues and their histories also helped to understand the hospitality proposition of the venues and how these propositions were shaped by the social, economic, political and physical environment. For example, the decaying nature of the urban fabric, aging population, rising 'rent gap' between the potential value of the land and the income actually generated from rents led to the emptying of numerous buildings. Many of the venues were established because buildings awaiting demolishment could be occupied by operators, and urban decay, therefore, gave these venues distinct identities. Some of the districts were going through a process of social and economic regeneration. Several operators exploited pauses in the regeneration, for example, while developers and local authorities clashed over development plans. Other operators actively contributed to an area's regeneration by maintaining a 'cultural scene' (see Lugosi et al., 2010).
Tracing multiple representations of the venues revealed how distressed, mismatched tables and chairs, table football, lampions and artistic decorations were fundamental to the propositions of hospitality in these spaces. Closely linked to this were the various cultural events hosted by the venues, including book launches, talks and debates, fashion shows and art exhibitions. Moreover, tracing consumer representations of the venues,
30
through streaming, helped to understand how consumers experienced the venues. For example:
[...] last time I was at Szimpla, a huge norwegian contingent was there - love norwegians dont' [sic] get me wrong, but not the kind who feel the need to paint their flags on their faces and ladies loo was filled with Aussies. Actually, foreigners don't bother me, as long as they are the right kind (a bit boho, can string an intelligent sentence together etc) but those trendy fake-nailed, hair-straightened, designer clothes wearing idiots that would be better at the White Party or whatever its [sic] called. Leave the romkocsmak to the rest of us who probably wont' [sic] be spending 3 hours to get ready to go there. (pestcentric.com, 08.05.2007 [emphasis added])
These types of commentaries provide critical, though fragmented, information about consumer perceptions of the venues and their customer profiles, including how both have changed. Moreover, such reflections on the scene, offered insights into commentators' (sub)cultural values and their sense of identities.
Significance, applications and contributions to knowledge The use of streaming was particularly important in understanding the evolution of this genre and the propositions involved. Because of its fragmented, evolving nature, identifying key and sub-streams helped to 31
create a composite picture of rom/kert hospitality. Beyond the transactions of food and drink, multiple cultural activities were entangled in the genre. Therefore, only by tracing multiple representations, which included those of cultural commentators, critics, journalists, operators and customers, could the complex dynamics of its development be understood.
The examination of the rom/kert and similar trends and phenomena has a number of implications. Firstly, understanding this interaction between venues and urban transformation contributes to debates among academics working in areas of geography and social policy, and, potentially, policy makers, concerning the multiple contributions hospitality can make in urban regeneration. Secondly, for commercial operators, investigating the dynamics of this genre of hospitality can inform their business strategies in occupying unorthodox physical spaces and utilising internet-based consumer-toconsumer interaction to understand consumer tastes and create demand. Furthermore, investigating this genre can inform operational policies concerning music, decoration and cultural activities used to define their commercial propositions. Pursuing these lines of inquiry, through streaming, again, provides hospitality researchers opportunities to contribute to wider academic, cultural, commercial and policy agendas.
6. Conclusion
This paper has discussed netnography and introduced the notion of streaming to conceptualise a related investigative process on the internet. As 32
we argued, online research might be informed by the principles of netnography, and, more broadly, ethnography, but it may be problematic to call all such investigative internet research 'netnography'. In short, all netnography involves streaming, but not all IRI involving streaming is netnography. We introduced three studies to demonstrate different aspects of streaming, including the identification and tracing of key and sub-streams. We suggested that the researcher can begin tracing streams of inquiry from three initial points of focus: 1) a specific virtual space, 2) an individual, and 3) an emerging trend or phenomenon. Table 1 summarises potential themes and issues to be considered in the different focuses of streaming, which were highlighted in the three cases.
Table1. Initial focus and potential investigative themes in streaming Initial focus S pa ce Individual Potential • Topics of discussion?• Performances of selves themes and • Uses of space • Extending networks issues in • Users of space • Codes and shared streaming meanings in • Interactions, interactions including collaborations and • Transactions and conflicts reciprocity
• • • •
Phenomenon Key actors/agencies involved in creation of phenomenon Phenomenon's characteristics Evolution of phenomenon Conflicts associated with phenomenon development
Such investigative techniques can be used within exploratory studies, which have individual value (cf. Stebbins, 2010); but, they may also be deployed within broader data mining exercises. IRI can be used to identify both key 'concepts' (e.g. types of behaviour, relationships, venues, (sub)cultural codes etc.) and 'contexts' (i.e. specific sites or other online sources of data) (Lau et
33
al., 2010). These can then be analysed quantitatively to discover new patterns and test hypotheses (ibid).
It is important to acknowledge the challenges and limitations associated with the investigative approaches discussed here. Engaging in unobtrusive research limits the data to that which occurs naturally. It also raises a number of ethical questions, concerning covert research, invasion of privacy and the absence of informed consent (cf. Lugosi, 2006). Even if greater interaction is sought, engaging with active users has fundamental sampling biases, and although online investigative research can provide access to many sections of the population that may otherwise remain invisible, it still cannot account for how 'passive' users receive or react to online material. Finally, as with many other types of qualitative research, online investigations may be criticised for their lack of generalisability, although multiple investigative studies can be 'concatenated' (Stebbins, 2010), and, as noted above, IRI can become the first stage in a confirmatory, quantitative exercise.
Regardless of whether it is used within an exploratory study, or as part of a broader data mining exercise, we have argued that engaging in this type of online research has broad applications in examining the experiences of employees, consumers and in developing human resource, marketing and operational strategies. Moreover, as the cases highlight, such research offers hospitality academics opportunities to contribute to broader disciplinary and policy debates. The ubiquity of the internet and the access it provides to data will undoubtedly lead to its increasing use in the future among academics. 34
However, a remaining challenge, certainly for the readership of this journal, is to translate these investigative techniques so commercial operators can also explore and exploit the internet to develop sustainable business practices.
7. References
Beaven, Z. and Laws, C. (2007), "'Never Let Me Down Again': Loyal customer attitudes towards ticket distribution channels for live music events: A netnographic exploration of the US leg of the Depeche Mode 2005-2006 World Tour", Managing Leisure, Vol. 12, pp. 120-142.
Dholakia, N. and Zhang, D. (2004), "Online qualitative research in the age of e-commerce: Data sources and approaches", Forum: Qualitative Social Research, Vol. 5 No. 2, Art. 29, available at:http://nbnresolving.de/urn:nbn:de:0114-fqs0402299 (accessed 10 August 2010).
Falzon, M. A. (2009) (Ed.), Multi-Sited Ethnography, Ashgate, Abingdon.
Fielding, N. G., Lee, R. M. and Blank, G. (2008) (Eds.), The Handbook of Online Research Methods, Sage, London.
Grossnickle, J. and Raskin, O. (2001), Handbook of Online Marketing Research, McGraw-Hill, Blacklick, OH.
35
Guimarães, M. J. L. (2005), "Doing anthropology in cyberspace: Fieldwork boundaries and social environments", in Hine, C. (Ed.), Virtual Methods: Issues in Social Research on the Internet, Berg, Oxford, pp. 141-156.
Hannam, K., Sheller, M. and Urry, J. (2006), "Editorial: Mobilities, immobilities and moorings", Mobilities, Vol. 1, pp. 1-22.
Hewson, C., Yule, P., Laurent, D. and Vogel, C. (2003), Internet Research Methods: A Practical Guide for the Social and Behavioural Sciences, Sage, London.
Hine, C. (2000), Virtual Ethnography, Sage, London.
Hine, C. (2005) (Ed.), Virtual Methods: Issues in Social Research on the Internet, Berg, Oxford.
Janta, H., Ladkin, A., Brown, L. and Lugosi, P. (2011), "Employment experiences of Polish migrant workers in the UK hospitality sector", Tourism Management, Vol. 32, pp. 1006-1019.
Jones, S. G. (1998), "Information, internet, and community: Notes toward an understanding of community in the information age", in Jones, S.G. (Ed.), Cybersociety 2.0: Revisiting Computer-Mediated Communication and Community, Sage, London, pp.1-35.
36
Kozinets, R.V. (2002), "The field behind the screen: Using netnography for marketing research in online communities", Journal of Marketing Research, Vol. 39, pp.61-72.
Kozinets, R. V. (2010), Netnography: Doing Ethnographic Research Online, Sage, London.
Langer, R. and Beckman, S. C. (2005), "Sensitive research topics: Netnography revisited", Qualitative Market Research: An International Journal, Vol. 8, pp. 189-203.
Lau, K.N., Lee, K.H. and Ho, Y. (2010), "Text mining for the hotel industry", in Enz, C.A. (Ed.), The Cornell School of Hotel Administration Handbook of Applied Hospitality Strategy, Sage, Thousand Oaks, CA, pp. 66-84.
Leander, K. M. and McKim, K. K. (2003), "Tracing the everyday 'sitings' of adolescents on the internet: A strategic adaptation of ethnography across online and offline spaces", Education, Communication and Information, Vol. 3, pp. 211-240.
Lofland, J. and Lofland, L. H. (1995), Analysing Social Settings: A Guide to Qualitative Observation and Analysis, 3rd ed., Wadsworth, Belmont, CA.
37
Lugosi, P. (2006), "Between overt and covert research: Concealment and revelation in an ethnographic study of commercial hospitality", Qualitative Inquiry, Vol. 12, pp. 541-561.
Lugosi, P. (2009), "Ethnography, ethnographers and hospitality research: Communities, tensions and affiliations", Tourism and Hospitality: Planning and Development, Vol. 6, pp. 95-107.
Lugosi, P., Bell, D. and Lugosi, K. (2010), "Hospitality, culture and regeneration: Urban decay, entrepreneurship and the 'ruin' bars of Budapest", Urban Studies, Vol. 47, pp. 3079-3101.
Lynch, P., McIntosh, A. and Tucker, H. (2009) (Eds), The Commercial Home: An International Perspective, Routledge, London.
Markham, A. N. and Baym, N. K. (2009) (Eds), Internet Inquiry: Conversations about Method, Sage, London.
Miller, T. W. (2005), Data and Text Mining: A Business Applications Approach, Pearson, Upper Saddle River, NJ.
Patton, M. (2002), Qualitative Research and Evaluation Methods, 3rd ed., Sage, Thousand Oaks, CA.
Pink, S. (2007), Doing Visual Ethnography, 2nd ed., Sage, London. 38
Stebbins, R.A. (1992), Amateurs, Professionals and Serious Leisure, McGillQueens's University Press, Montreal.
Stebbins, R. A. (2010), "The Internet as a scientific tool for studying leisure activities: exploratory Internet data collection", Leisure Studies, Vol. 29, pp. 469- 475.
Strauss, A. and Corbin, J. (1998) The Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory, 2nd ed., Sage, Thousand Oak, CA.
Urry, J. (2007), Mobilities, Polity Press, Cambridge.
Watson, P. Morgan, M. and Hemmington, N. (2008), "Online communities and the sharing of extraordinary restaurant experiences", Journal of Foodservice, Vol. 19, pp. 289-302.
Wittel, A. (2000), "Ethnography on the move: From field to net to internet", Forum: Qualitative Social Research, Vol. 1 No. 1, Art. 21,http://nbnresolving.de/urn:nbn:de:0114-fqs0001213 (accessed 10 August 2010).
39
doc_599677782.docx