Researching secretive Internet phenomena: Netnographical methods and employees blogs
James Richards
Heriot-Watt University, Edinburgh
Contact:
ABSTRACT
This paper looks at conducting research with employee bloggers - employees typically suspicious of outsiders and not recruitable via conventional gate keepers, such as HR managers. The paper provides an insider take on the methodological problems and dilemmas faced when conducting a qualitative study of employee blogging practices.
INTRODUCTION
This paperconcerns a discussion of the methodological details of a research project focused onemployee bloggers, or employees who write online diaries about their work (Schoneboom, 2007). A further key focus of the paper is the application of conventional methods, such as questionnaires, adapted to conduct a netnographical study of online communities and cultures (Janta and Ladkin, 2013) - in this instance communities of employee bloggers. The main aim of the paper, however, is to provide a personal or "behind the scenes" account of a novel and innovative research project, representing, in effect, the trials and tribulations of researchers that rarely makes it into journal articles.
The emergence of employee blogging
Wide interest and awareness of employee bloggingemerged circa 2004 on the back of a range of international news stories reporting employees disciplined for posting views about their jobs, employers and colleagues, to new and emergent social networking platforms, such as blogs. In this instance, however, employee blogging quickly became branded by the popular business press as a form of deviant or anti-business behaviour. Examples of such one sided views include employee attempts to cause harm to employer reputations (Spencer, 2005), create inconsistency with business mission statements (Joyce, 2005) and exposing employer brands in a way not possible just a short time ago (Phillips, 2008).
However, it quickly became evident that employees blogging about their jobs was something quite different from employee misuse of the Internet at work,a quickly growing sub-field of HRM scholarly research. Indeed, simple observations of employee blogging reveal a distinct difference between acts such as "cyberloafing" (e.g. see Lim and Teo, 2005; Blanchard and Henle, 2008) and "cyberslacking" (e.g. see Block, 2001; Garrett and Danziger, 2008) on work time and on work premises, and acts involving employees keeping an online diary in their own time, using their own ICT equipment, and deploying a range of strategies, such as writing under pseudonyms, in order to avoid the distinct possibility of disciplinary action by employers (Richards, 2008).
As aconsequence of these events and realities, it became apparent that HRM academics wishing to research employee bloggers faced a range of challenges. These challenges included designing methodologies that put to one side popular and emergent views of employee bloggers, but more importantly, designing methodologies to gain access to employees suspicious of outsiders and not accessible through conventional research gate keepers, such as, HR managers and other senior employers of work organisations.
Structure of paper
This paper, as such, detailsthe realities of an unfunded, yet innovative research project involving the use ofqualitative research methods in order to gain insights into the hidden and secretive world of the employee blogger. However, it is the fuller details, details usually omitted, overly compressed or edited out of final versions of journal articles, that bring out the innovative side to this particular HRM research project. To do this the paper is structured as follows. The first section more clearly delineates the methodological problems and dilemmas faced by a researcher attempting to research employee blogging practices. The second section discusses details of the eventual research strategy used to research employee bloggers. The third section highlights the realities of researching hidden and secretive Internet phenomena. The fourth section sums up all the key points in a discussion section.
METHODOLOGICAL PROBLEMS AND DILEMMAS ASSOCIATED WITH RESEARCHING EMPLOYEE BLOGGERS
The research project in question began late 2004, at the tail end of a PhD ethnographical research project on traditional forms of workplace employee misbehaviour and resistance.The impetus for the research project in question was the emergence of a new form of employee misbehaviour that had begun to attract the attention of the popular business press. As such, a decision was made to launch an unfunded, side-research project on employee blogs. The onset of the research project, however, came with a range of key methodological problems and dilemmas. These problems and dilemmas are critical to unpicking the innovative methodology of the research project on employee blogs, and as such are detailed below.
Scale of the employee blogging phenomenon
At the time there was no definition of what an employee blog was. There was also no research to indicate how employee bloggers could be found, how many employee blogs existed, or how influential they may be in terms of blogging communities and possibly beyond.
Data collection methods
With no pre-existing literature on employee blogging, what would be the most efficient method or methods to collectdata on employee blogging practices, especially as a research project of this kind required the gathering of a significant amount of qualitative data, from a high number of employee bloggers, to tease out the many possible angles on employee blogging?
Research ethics
What would need to be considered in terms of research ethics? Without too much prior thought it seemed that attention would almost certainly be needed in terms of participant confidentiality and potential harm to research participants and the blogging communities they belong to. In other words, developing trust between researcher and research participants could be difficult given how employee blogs had been portrayed by the popular media.
Recruitment of research participants
How would it be possible to recruit employee bloggers to a research project necessitating the divulgence of potentially identifying details? Another problem was that the vast majority of employee bloggers went to great lengths to protect their identity. A third particular recruitment problem was that employee bloggers could not be recruited through conventional means, such as employers or HRM practitioner networks.
DESIGNING A STRATEGY TO RESEARCH EMPLOYEE BLOGGING
In this section how the methodological problems and dilemmas identified in the previous section were tackled are detailed and discussed.
Scoping out the employee blogging blogosphere
Deciding what was to be considered an “employee blog” was an important first consideration, as no definition existed at the time. A test for inclusion, however, was quickly developed in terms of being a blog that contains strong reference to matters of work, particular in terms of providing accounts of working or customer relationships. Blogs of a highly technical or specialised nature were not considered for the study, neither were blogs where work was a passing or fleeting matter.
While the main priority of the research project was to establish employee views of their blogging practices, it was also necessary, to a point, to establish the scale of such activities. Given the limited time and resources available to conduct the research project,a priority, however, was finding sufficient quantities of employee blogs to make a study of employee practices worthwhile.
At first employee blogs were identified through following up media news stories using Internet search engines. However, such an approach helped identify a very limited number of employee blogs. What allowed further and many more employee blogs to be identified was the use of the blogroll (a list of blogs, with hyperlinks, found on a blog and typically representing blogs that the keeper recommends or is connected to in some way) of each employee blogger. It quickly became apparentthat employee bloggerstypically forged links with other employee bloggers and in particular employee bloggers from similar jobs, occupations or professions. Blogrolls, as such, significantly eased the challenge to find many more employee blogs. Further, it seemed sensible and practical after a while to start a research blog (research blog referred to in this paper can be located at and to create an extensive blogroll of identified employee blogs.
By early2007 a database of approximately 1000 employee blogs had been created, with the 1000 blogs categorised by broad occupational themes, with various health professionals, police officers and educators making up the vast majority of employee blogs found. The database also denoted the origin of the blog - most blogs collated were written by USA employees, but the database also accounted for approximately 200 blogs written by employees from the UK, Australia, Canada and New Zealand.
Data collection and researching employee bloggers
A wide-range of data collection methods were considered in order to explore the employee views of employee blogging practices. For example, ideal methods associated with ethnography, such as, participant observation, were ruled out because they were a combination of impossible to do or would be far too time consuming for a small, unfunded research project that was already proving to be time consuming in terms of scoping out the employee blogging blogosphere. Interviews with employee bloggers were also considered, yet rejected as a method because of unsuitability in terms of the number of participants required of the project, the time consuming nature of arranging and conducting interviews and the costs and practicalities associated with conducting interviews on a global scale (note: this was a time before Skype). Interviews were also not considered as they were judged to be too intrusive at this stage, although they were eventually used for follow up data collection in late 2007/early 2008 (see Ellis and Richards, 2008; Richards and Kosmala, 2013). The eventual decision was to adopt what was then an emergent data collection tool - an electronic questionnaire.
In order to acquire the necessary qualitative data to widely explore employee blogging habits the eventual electronic questionnaire was formulated on a range of self-reporting and open-ended questions covering the subject of personal employee blogging habits, such as: what motivated you to start blogging about your work and what is the purpose of your blog?However, to avoid being overly intrusive and almost certainly significantly reduce the return rate of the electronic questionnaire, personal details were not requested from research participants. A key issue in mind at the time was that electronic questionnaires do not offer the same degree of flexibility as interviews or observations, such as the opportunity to rephrase or reorder questions. As such, it was decided to pilot electronic questionnaires on a small number of employee bloggers. In this instance piloting was not expected to be conducted in a conventional fashion, such as requiring feedback from participants on the questions set. Instead, it was decided that evaluating the quality of data received from the first batch of electronic questionnaires would be the deciding factor in any changes to the electronic questionnaire.
A key advantage of electronic questionnaires is that electronic questionnaires come with a permalink that makes distribution efficient in Internet-based scenarios. A further key advantage of using the electronic questionnaire is that it complimented the nature of what was being researched, that of employees highly engaged with emergent forms of Internet-based communication technologies.
The ethics of researching employee bloggers
Between attempts to consider the most efficient and pragmatic research methods to gather qualitative data on employee blogging practices, and attempts to recruit employee bloggers to the study, the ethical implications of the study were further considered. For instance, at the time debates were emerging about the ethical implications of conducting research in domains that may well be technically accessible, yet few out of the many hundreds of millions of Internet users were likely to stumble upon the vast majority of employee blogs. Further, it was necessary to factor into the study design the negative impact that the popular business press had already had on employee blogging communities, such as the heightened fear of employee bloggers facing exposure to the wider world and the possibility of employer-led disciplinary action due to publishing unofficial accounts of work organisations to the Internet. Extra care and consideration was also needed, if anything, to go beyond the consideration of not causing any further harm and instead create researcher-researched relations based on trust and mutuality, almost certainly necessary in order to present a rich account of employee blogging practices.
What this entailed is summarised as follows. At the most general level it involved joining to a point a range of employee blogging communities. Having set up a blog in order to keep track of employee blogs it seemed a sensible strategy to become an active blogger and use the research blog as a means to develop a rapport with employee bloggers. Central to ethics strategy was to establish trusting relations between researcher and employee bloggers, or at the very least present a means by which the researcher could establish his credentials should a potential research participant question the status and motives of the researcher. This part of the research strategy quickly evolved over the first year of the life of the research blog (April 2005 to March 2006), but continued for many more years afterwards. The strategy involved the following. All contact and research interest details of the researcher were placed on the blog, with details of the research project (in lay terms) presented there too. Over time draft papers on employee blogging were posted on the research blog. Over time the researcher also blogged about how employees were making use of a wide-range of social media, especially blogs, and more lately, for example, Twitter, Facebook and LinkedIn, in order to express, pursue and defend their interests as employees. The researcher, as such, became a hub of informal activity surrounding employee blogs, with the research blog widely linked through the blogrolls of employee bloggers. The blog itself created a wide-range of interest from the popular media leading to the researcher being interviewed several times by national newspapers, from several countries, on the subject of employee blogging, thus further strengthening the credentials and reputation of the researcher in relation to employee blogging communities.
Recruiting employee bloggers
Attempts to recruit for the study in question began before it was decided to no longer continue adding to the blog-based database of employee blogs. As such, the recruitment process began in April 2005 when approximately half of the eventual total of employee blogs had been identified and categorised. As such, the data gathering from the self-reporting questionnaire began in April 2005 and ended in October 2005. During that time 520 questionnaires were distributed to employee bloggers, with 207 or 40 per cent of distributed questionnaires returned during the seven months of data collection. However, details of how many eventually took part in the study are less significant than considering how employee bloggers were made aware of the study and the accompanying questionnaire.
Based on ethical principles outline in the previous sub-section employee bloggers were recruited in two ways. Firstly, observations of employee blogs often revealed email contact details of the employee blogger, although the email address in nearly all cases rarely revealed any identifying features of the blogger. In this instance afriendly and respectful email was sent to the employee blogger outlining brief details of the study and an invitation, including a web-link to the electronic questionnaire, to take part in the research project. The email also contained a web-link to the research blog in order to verify the credentials of the researcher and the research project, as well details of how to contact the researcher in terms of questions and queries related to the research project. Secondly, where an email address was not available on an employee's blog, a message similar to the email described above was left as a comment on the latest post/diary entry made by the employee blogger. Employee bloggers were only contacted once, however, even if there was an option to send an email and leave a message on a blog comment.
THE REALITIES OF RESEARCHING EMPLOYEE BLOGGERS
The execution of the research with employee bloggers was, however, not without its problems, although the purpose of this section of the paper is not to simply comment on the downsides of such a research project. Before conducting a broader discussion of the key points to take from this paper, it is important to present and discuss a range of importantlessons to arise from the study.
Time taken to conduct a netnographical study
As the study was unfunded and an experimental side research project the time taken to complete the research project on employee bloggers was neither anticipated nor recorded as it unfolded. However, a retrospective estimate suggests the research design and data collection stage accounted for ten or possibly more hours per week for nearly a year. In other words, the equivalent of what is typical of the fieldwork associated with a PhD, or at least a modestly funded research project. In this instance, however, the research project in question was mostly conducted outwith regular office hours, often in the evening and at weekends. Having said that, the research project was exploratory, focused on a range of phenomena that little was known about at the time. Subsequently, more time than an average amount of time was required to conduct research on employee blogging practices.