- Email: [email protected]
Internet ethnography: A review of methodological considerations for studying online illness blogs
International Journal of Nursing Studies 51 (2014) 1686–1692
Contents lists available at ScienceDirect
International Journal of Nursing Studies journal homepage: www.elsevier.com/ijns
Discussion
Internet ethnography: A review of methodological considerations for studying online illness blogs Jessica Keim-Malpass a,*, Richard H. Steeves b, Christine Kennedy c a
University of Virginia School of Nursing, Charlottesville, VA, USA Professor Emeritus, University of Virginia School of Nursing, Charlottesville, VA, USA c Madeline Higginbotham Sly Professor of Nursing, University of Virginia School of Nursing, Charlottesville, VA, USA b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 14 December 2013 Received in revised form 6 June 2014 Accepted 11 June 2014
Background: In recent history, the Internet has emerged as a wealth of archived, ongoing, interactive, and socially mediated data. Conducting Internet ethnography is a fairly new methodological approach, however, it has been previously described as a valid form of inquiry. Illness blogs, in particular, have great implications for nurse researchers, as they are able to study the experience of illness in a naturalistic and longitudinal manner, often with greater detail than data relying solely on participant recall. Participants are able to produce online illness blogs as a way to share their own illness narratives and connect with others going through similar processes. Objectives: The purpose of this paper is to discuss methodological considerations in studying online illness blogs through Internet ethnography. Methods: This article provides an overview of Internet ethnography as an emerging qualitative method and an introduction to research using illness blogs. Through use of this method in an exemplar study of young women with cancer, key decision points are highlighted along with the study team’s field experiences. Conclusion: Issues pertaining to method applicability, active vs. passive involvement as a researcher, ethical considerations, what constitutes data, sampling approach, procedural and analytic decisions, and thoughts regarding reflexivity and voice of the research participants’ are addressed. Strengths and limitations of the study of online illness blogs through Internet ethnography in nursing science are also discussed. ß 2014 Elsevier Ltd. All rights reserved.
Keywords: Qualitative Methods Internet Ethnography Illness blog Blog
What is already known about the topic? Conducting Internet ethnography is a fairly new methodological approach, however, it has been previously described as a valid form of naturalistic inquiry. Illness data captured on the Internet data privileges the current experiences of patients and contain detailed accounts of disease processes and psychosocial ramifications.
* Corresponding author at: P.O. Box 800826, Charlottesville, VA 22908, USA. Tel.: +1 6363466855. E-mail address: [email protected] (J. Keim-Malpass). http://dx.doi.org/10.1016/j.ijnurstu.2014.06.003 0020-7489/ß 2014 Elsevier Ltd. All rights reserved.
What this paper adds This paper explores methodological considerations of Internet ethnography in the context of nursing science using decision points from an exemplar study. This paper examines the need for fluidity, an iterative process, negotiation of study involvement as researcher, and early conceptualization of ethical considerations when using Internet ethnography. This paper describes the advantages in using this method to offer utility in studying experiences among populations that are traditionally difficult to study (e.g., end-of-life experiences, groups with stigma associated with their disease).
J. Keim-Malpass et al. / International Journal of Nursing Studies 51 (2014) 1686–1692
1. Introduction Over the past two decades, the Internet has become the most comprehensive electronic archive of written material and has significantly changed the way people communicate, access information, and live their lives. The daily influx of new information available online represents a wide range of current events, viewpoints, opinions, and virtual communities that allow ordinary citizens to have an active voice with an accessible medium in which to express it. Internet support groups, resources, social networking sites such as Twitter, Facebook, and Instagram, and illness blogs have recently emerged as popular media for patients, their caregivers and health care professionals (Thielst, 2007; Ziebland and Wyke, 2012). Online health resources play a supporting role in many people’s lives as they pursue better health and navigate the healthcare system (McWilliam, 2009). Illness blogs, or blogs that were initiated by users to document their experience with a specific disease, have many purposes and benefits. Blog writers documenting an illness are able to update family and friends through the course of treatment and recovery, become an advocate for and educate others going through the same experience, ask questions and get answers from their online community, and gain the emotional benefits of expressive writing (Pennebaker and Seagal, 1999; Thielst, 2007; Ziebland and Wyke, 2012). Illness blogs also grant a voice to patients through engagement in digital storytelling and socially mediated connection with others (Gubrium et al., 2013; Lagu et al., 2008; Ressler et al., 2012; Treadgold and Kuperberg, 2010). Additionally, because users may remain anonymous if they choose, there are vast implications for self-disclosure and subsequent support of potentially stigmatizing health topics (Rains, 2014; Saiki and Cloyes, 2014). It is estimated that 12% of adults with a chronic health condition maintain their own illness blog, and 28% read peers’ illness blogs (Fox and Purcell, 2010). While blogs are beneficial for the patients who write them (Stanton et al., 2013), they also offer a new modality for health care providers and researchers to understand lived experiences from digital storytelling and sharing of the illness narrative. The Internet as a social domain has extended the reach of human subjects research for social scientists and textual data found on these venues are a rich source for researchers interested in understanding the hermeneutics of patients’ lived experiences (Bassett and O’Riordan, 2002; Eysenbach and Till, 2010). Internet ethnography is a fairly new methodological approach, however, it has been previously described as a valid form of naturalistic inquiry (Markham and Baym, 2009; Miller and Slater, 2000). Previous work in this area has demonstrated that virtual technologies are best studied through a flexible and narrative approach that ethnography lends itself to (Markham and Baym, 2009). When compared to interview qualitative data, it has been noted that Internet data privileges the current experiences of patients and contain detailed accounts of disease processes and psychosocial ramifications (Keim-Malpass et al., 2013a,b). Unlike in-person interviews, researchers
1687
do not have to rely on the participants recall (Seale et al., 2010). Because of the ease with which large amounts of archived materials can be accessed and analyzed, this source of data has considerable potential for direct observation of illness experiences (Seale et al., 2010). This paper will discuss methodological considerations of Internet ethnography for use in nursing science by examining elements of an exemplar study involving Internet illness blogs of young women with cancer. Findings from this study have been previously reported (Keim-Malpass and Steeves, 2012; Keim-Malpass et al., 2013a,b) and methodological elements requiring decision were based on the study team’s experience and documented in field notes throughout the process. To the authors’ knowledge, this research was the first to use Internet illness blogs for hermeneutic interpretation immersed in an ethnographic context. In order to advance use of this unique and innovative methodology, key decision points are highlighted (Fig. 1), along with the study team’s experiences in the field and strengths and limitations of the use of this method (Keim-Malpass and Steeves, 2012; Keim-Malpass et al., 2013a,b). 2. Methodological considerations 2.1. Introduction to Internet ethnography As new forms of Internet ethnography begin to emerge, much of their epistemological underpinnings remain the same as traditional forms of ethnography. Ethnography
Fig. 1. Key decision points when using online ethnographic methods.
1688
J. Keim-Malpass et al. / International Journal of Nursing Studies 51 (2014) 1686–1692
focuses on understanding social stories and what is going on in the field, and this remains true both on and offline (Agar, 1986; Murthy, 2008). Traditionally, an ethnographer is thought of as embarking in field work in a community and seeking to understand the culture that she encounters by making sense of social actions from the point of view of the participants (Clifford, 1986). Within the past fifteen years, the Internet has changed the way qualitative data is both conceptualized and collected while also impacting how ethnographies can take place. Without a physical boundary defining place, ethnographers doing research online must still immerse themselves in the social situation being studied to learn the socially relevant things that are happening there (Hine, 2009). Logistically speaking, the notion that the Internet can erase all geographic boundaries and allow ethnographers to connect with people instantaneously has many advantages. It allows for increased access to various perspectives, including those of people who may not have participated in traditional face-to-face research interviews, or those who may be stigmatized due to their disease or circumstances (Schotanus-Dijkstra et al., 2013). It also provides a method by which international sampling and global perspectives can be easily accessed. 2.2. Is the method applicable: population identification, involvement of researcher, and ethical considerations Prior to embarking in Internet ethnography, researchers must first determine if the population of interest can be self-defined and ensure that the online community exists. In other words, is the population of interest readily accessible (through self-identification) in the online environment and do the members of this community interact with one another (Markham and Baym, 2009)? Traditional clinical research relies on having very strict inclusion and exclusion criteria, and admittedly, the inclusion/exclusion criteria may need to be modified for online ethnographic work. In the exemplar study involving women diagnosed with cancer from ages 20 to 39, the study team had to accept an age range of diagnosis as opposed to having very strict clinical knowledge regarding precise dates of diagnosis (Keim-Malpass and Steeves, 2012). From spending many months in the field, the lead author ascertained that women often did describe their age at diagnosis, but sometimes did not specify a particular age, instead using an age-range when diagnosed (i.e., diagnosed with ovarian cancer in my late twenties). Therefore, the inclusion criteria necessitated flexibility by virtue of not having access to specific clinical data elements (Keim-Malpass and Steeves, 2012). Additionally, the lead author spent initial time understanding the context in which young adults with cancer communicate with one another in the online environment. This step was critical in determining that the population of interest had an active presence online, could be selfidentified, and regularly used social media and their blogs to communicate with one another (Keim-Malpass and Steeves, 2012). The next step for ethnographers in virtual fields is to consider how active to be in relation to the participants
they are studying (Hine, 2009). Three different types of Internet-based research methodologies have been described: (1) passive analysis that involves understanding how information and dialogs are exchanged on discussion groups or blog sites without the researchers actually involving themselves, (2) active analysis in which the researcher actively is involved in participation and dialog exchange, and (3) self-identified active analysis where researchers identify themselves and their research interests a priori which may allow them to also gather information in a more in-depth interview or focus group setting or use the Internet to recruit participants for clinicbased studies (Markham and Baym, 2009). The difference in levels of active participation by the investigator has various methodological and ethical implications. In the exemplar study, the decision was made to be passive investigators and analyze illness blogs that were already in existence and located on public Internet sites (KeimMalpass and Steeves, 2012; Keim-Malpass et al., 2013a). This decision was based on the study team’s interest in a naturalistic inquiry to further understand the lived experience of young women with cancer without any preconceived interview questions to direct responses. This approach follows the key assumption that ethnographic methodology should use the entire experience of immersion to learn how life is lived, rather than coming in with particular a priori assumptions of the issues that are important to the group being studied (Hine, 2009). This form of inquiry was also critical in understanding the experience of cancer from a longitudinal perspective because we were able to follow the young women nonconcurrently, meaning that analysis began at the time of diagnosis and moved forward in time to the most recent blog posting. This method allowed us to follow as much of the disease process as possible (up to 4 years for some women), from the time participants were diagnosed through treatment, and transitioning into survivorship or death, noting experiences with recurrence and remission, and other setbacks and milestones along the way (Keim-Malpass et al., 2013b). After deciding on whether to conduct passive or active Internet ethnography, Institutional Review Board (IRB) approval is the necessary next step. The study team was careful to consider the ethical implications and ramifications of conducting an Internet ethnography using publically available Internet data prior to initiation of the study. These considerations tend to focus on the challenges of privacy, confidentiality and informed consent and are very dependent on where the data is being obtained, the level of privacy the author of the content assumes, and how active the researcher is in relation to the participants (Eastham, 2011; Heilferty, 2011). The exemplar study was submitted to the IRB and need for approval was exempted due to the public nature of the blog sites and because the study team was not directly interacting with participants to ask any specific questions (i.e., we were using passive analysis). The ethical implications of online data for social science research are evolving, and of note, IRBs have not yet shown standard methods in granting approval Internet-based observational studies. The study team then had to make a decision about permissions
J. Keim-Malpass et al. / International Journal of Nursing Studies 51 (2014) 1686–1692
needed if the blog included names and an email address. Some bloggers chose to use an alias, or a name not representing their real identity, and thus did not have email addresses present. Their privacy and anonymity was respected and no contact was made. However, if the blog did include a name and email address, the decision was made to contact the blog participant to ask for permission for their blog to be included in our study. Participants with accessible emails responded favorably and all granted permission for inclusion. Following IRB approval, the lead author became immersed in the Internet culture of young women with cancer (diagnosed between the ages of 20 and 39) for roughly six hours per week over a time period of four months before the actual blog analysis began. This process began at several key young adult cancer websites (stupidcancer.com, planetcancer.org, livestrong.org, youngsurival.org). The lead author followed the general content on the website, listened to weekly and special content Podcasts hosted by young adults with cancer, followed the live chats of survivors during the Podcasts, and followed postings on social networking sites such as Facebook and Twitter. Twitter searches were initiated, through ‘following followers’ of authors that were self-described as diagnosed with cancer during young adulthood. ‘Following followers’ is a method of connection on Twitter that allowed the user to see who follows certain individuals or young adult cancer organizations. Many of the Twitter users obtained through this method were either interested in young adult cancer survivor issues or self-identified as young adult cancer patients/survivors. This initial time of immersion was critical in understanding how young adults with cancer express themselves, describe issues that are important to them, communicate among each other, and represent themselves online. The study team was able to understand how the Internet allowed users to construct their social network around their experience (Keim-Malpass and Steeves, 2012). This time of immersion was central in developing and mapping online communities as a part of the ethnographic process (Markham, 2005). 2.3. What are data? Procedural considerations, sampling and analytic strategy During the time period of immersion, the next decision involved determining what type of data would be included in the analysis. Many of the illness blogs contained narrative text in diary form, access to photos, video blogs (videos of themselves created and often linked to YouTube), hyperlinks to websites, and other illness blogs the blogger followed, meaning that numerous items found on the blog could be considered data for analysis. All of these data elements are important and represent the digital story the young women were telling about themselves, however, the decision was made by the study team to limit the data for hermeneutic analysis to the narrative text found on the illness blogs (Cohen et al., 2000). Narrative text included actual blog entries, as well as comments posted on each blog entry site.
1689
Sampling considerations were also considered and further refined during the antecedent period of immersion. Potential blog sites were initially ascertained for inclusion using traditional inclusion/exclusion criteria but then had an additional sampling criteria applied including: three out of four randomly selected blog entries focused specifically on the cancer experience, the blog title or biography of the blog author highlighted the cancer focus, the blog author indicated cancer type and her age (or age range) when the cancer was diagnosed as well as current age (or age range) (Keim-Malpass et al., 2013a,b). The study used a snowball convenience sample to obtain the sample, which began with initial blogs that were highlighted on a young adult cancer website (stupidcancer.org). This purposive strategy was used to initially sample a small number of the young women who met inclusion criteria, then subsequently include women who were socially connected (followed each other Twitter) and actively engaged online (regularly commented on blog entries, conversed regularly on Twitter) with the original sample. Through the online medium, the snowball approach is achieved through social network mapping and identifying other members who are socially engaged with the initial participant. Sample ascertainment for the exemplar study is described in detail in previous publications (Keim-Malpass and Steeves, 2012). The final sample included sixteen illness blogs of women diagnosed with cancer who had maintained their blog on average for over two years (Keim-Malpass and Steeves, 2012). Central to the period of immersion, data collection, and analysis was the process of bracketing preconceived ideas and judgments about the research phenomenon in a field notebook, and this ritual was a daily process for the lead author (Cohen et al., 2000). Additionally, observations of the participants were made through data gathered from their blog and public Twitter handle (if applicable). Data from the blog sites were read and abstracted non-concurrently, meaning data originated from the time of diagnosis (or first blog posting related to cancer) and moved forward in time to present day. The timing was important to the investigators because we wanted to limit bias among those women who had been asked for permission to be included in the study by not analyzing data that was written after permission was ascertained. While reading each blog entry, the lead author archived and managed text data and supporting photographs and videos (for context) in a Word data file. All textual data (including blog entries and field notes) were reviewed multiple times, then analyzed to identify nodes, or individual units of meaning (Cohen et al., 2000). The nodes were then grouped according to categories based on similarities and then eventually into themes that expressed an overall meaning or interpretation (Cohen et al., 2000). Additionally, memos were maintained outlining how developed themes applied to previous findings in the literature and how the themes highlighted clinical, social and cultural aspects of young adult cancer survivorship. At this point, consensus was reached among investigators that saturation in themes had been achieved and no additional sampling would be needed.
1690
J. Keim-Malpass et al. / International Journal of Nursing Studies 51 (2014) 1686–1692
2.4. Online identity: voice and meaning From the beginning of the research analysis, the notion of online identity of the participants was considered. A blog site is a space in which to create an online identity that may or may not be consistent with the user’s offline identity. The users can choose how they wish to be represented, rather than allowing traditional demographic and socioeconomic factors such as race, gender, class, and body type to dictate how they are initially perceived by others, including researchers (Markham, 2005; Mo et al., 2009). Computer-mediated construction of self, other, and social structures are initiated by the user, whether those choices are conscious or not (Markham, 2005). In this study, text was central to the person’s construction, story, and negotiation of identity; however the design of the blog, the blog title, pictures associated with the blog, badges on the blog, lists of followers, and level of anonymity (real names/emails versus aliases) all contributed to the social construction of the online identity of the participant. There were numerous expressions of identity based on the design of the blogs. One blog was designed in all black with a picture of an atomic bomb explosion located on the profile. Another was reminiscent of a cabin in the woods. Some were scattered with personal photos. The narrative texts of the young women were equally as diverse, ranging from cynical language, to humor, to curse words, to words of comfort and contentment. Incorporating the construction of online identity into the ethnographic analysis became central in understanding the social and cultural context of illness blogs, and offers a unique strength to this methodology (Boellstorff et al., 2012). Created identities through an online medium represent an evolving approach to participant observation as a component of ethnographic research (Boellstorff et al., 2012). Social scientist and early Internet ethnographer, Markham (2005), suggests that researchers should continuously reflect on the interests of the research participants. She emphasizes the need to shift from re-representing existing concepts in the literature to focus on advocacy and understanding how this emerging body of literature can influence positive change by asking, ‘‘whose interests does this research serve and how can my writing and publishing give voice to those who might otherwise remain silent?’’ (Markham, 2005). Continually reflecting on these questions is central to online ethnographic research design and adherence to these principles hopefully achieves an ethically grounded and reflexive body of work (Lincoln and Guba, 1985). These concepts offer the reminder that the researchers’ voice is a privileged one and it remains paramount to give a voice to participants rather than unintentionally speaking for them (Fine et al., 2000). As noted previously, the lead author used a field notebook to continuously write down thoughts, perceptions and emotions during the initial months of immersion, during the time spent transcribing and reading blog narratives, as well as during analysis and dissemination phases. This helped in bracketing pre-conceived notions about ‘what was going on’, as well as providing ongoing iterative processing, enhancing trustworthiness, and limiting bias (Emerson et al., 1995; Lincoln and Guba, 1985).
Additionally, feminist undertones in the textual analysis were noted as they are necessary to relate the centrality of gender in the shaping of consciousness (Fine et al., 2000). Gender was a unique aspect to this study because of potential disruptions of roles in parenting young children through a cancer diagnosis and possible loss of fertility, which is often a more pronounced problem for young women cancer survivors (Dow and Kuhn, 2004). To incorporate this aspect of the study, gender was a central analytic lens that was often written about within the context of the field notes. Additionally, an unanticipated finding emerged during the process of initial analysis and dissemination. During the original study period, two of the young women died from their cancer diagnosis, yet continued to maintain their blog close to the time of their death. Their writings presented a unique perspective during the end-of-life time period that is often difficult for researchers to gain access to due to perceived burden to study participants. Sadly after initial analysis concluded, three additional young women died due to their disease, each representing widely varying end-of-life trajectories. To incorporate this emerging method in a way that enhances access to such populations during vulnerable stages of their disease trajectory (i.e., end-of-life), the lead author is currently in the process of conducting a post hoc analysis to gain insight from a time period. This lens is critical in giving voice to young women’s end-of-life experiences. 2.5. Strengths and limitations of method ‘‘The relationships that I formed online were probably the most true social interactions and relationships I have had. I was the most real with them. Perhaps it was the anonymity of the Internet that made it safe, perhaps it was the fact that they too had gone through something similar that made me feel connected. Whatever the reason, in my writings, I could open up to them about my thoughts and feelings. I have ‘‘met’’ many people whose lives have been touched by cancer. And with each introduction, discussion, comment, email, etc., I feel a commonality that I had not felt in my everyday life. I greatly appreciate this ridiculous world of the Internet’’ J., ovarian cancer survivor and blogger diagnosed at age 26 The use of Internet ethnography as an emerging qualitative methodology has numerous strengths (Table 1). Accounts documented in online illness blogs in this exemplar study were raw and honest and the blog authors were connected and immersed in each other’s cancer experience (Keim-Malpass and Steeves, 2012; Keim-Malpass et al., 2013a,b). Participants in the exemplar study were able to discuss many topics openly online that may have been difficult to discuss in traditional face-toface qualitative interviews (i.e., troubling symptoms such as anal fistulas, relationship difficulty, confrontation of mortality) (Keim-Malpass and Steeves, 2012; KeimMalpass et al., 2013a). Illness bloggers have the ability to choose when they want actively engage in emotional
J. Keim-Malpass et al. / International Journal of Nursing Studies 51 (2014) 1686–1692
1691
Table 1 Strengths and limitations of studying online illness blogs. Strengths
Limitations
No geographic boundaries Allows researchers to study those who may decline from participating in traditional qualitative study due to stigma Allows for the ultimate naturalistic inquiry while studying topics that are of importance to the participant No recall bias, participants write about the experience as it happens
Difficulty in having narrow inclusion/exclusion criteria Unable to confirm identity or validate diagnosis, treatment modalities, symptom-based experiences Digital divide exists – are only those with Internet/mobile access able to communicate online? Who is not represented online? Evolving ethical principles – no true consensus yet among IRBs, unable to generally provide informed consent when using completely public websites Findings may not be generalizable
Researchers can experience a disease in a longitudinal sense from diagnosis on while analyzing it non-concurrently Allows for low participant burden and unique perspectives (e.g., end-of-life experiences) Allows analysis of social support connection through online social networks
catharsis through expressive writing, and engage in social support mediated through blog comments and online communication on social media sites (e.g., Twitter, Facebook, Instagram). Markham (2005) suggests that study participants’ online persona can be fluid and we are able to bear witness to a range of emotion. Choosing levels of anonymity in online environments also gives control in the presentation of self (Markham, 2005). In doing so, participants also have the ability to disengage with Internet activity, lurk on other sites (read without actively participating) and/or re-engage when ready. The longitudinal, naturalistic and immersive approach in studying illness blogs through Internet ethnography is unmatched in terms of scope and range in understanding lived experiences. This method has very low participant burden, limited recall bias, there are no geographic boundaries limiting connection thereby increasing the ability of the researcher to incorporate diverse views in a global context. Further, it allows for analysis of longitudinal illness narratives and mapping of online social connectivity and social support derived and maintained through the Internet. When approaching this research using passive analysis and prolonged ethnographic immersion, the experiences and concerns of the participants can be privileged for an encompassing view of their lived experience. Despite this, limitations exist in using online modalities to answer socially and clinically relevant questions (Table 1). There may be difficulty in obtaining appropriate sample, findings are often not generalizable, demographic characteristics of the sample are very difficult to obtain and online identity cannot be validated. Due to the digital divide, only those with regular access to the Internet through computers or mobile phones and a conversant level of literacy are likely engaging in digital storytelling on illness blogs and online social networking. Two issues that are case-dependent are ethical issues and concerns over copyrighted data (Bassett and O’Riordan, 2002; Eastham, 2011; Grimes et al., 2009; Gubrium et al., 2013; Heilferty, 2011; Lipinski, 2009; Markham and Baym, 2009; Walther, 2002; Whiteman, 2007). Both of these considerations must be completely explored a priori prior to study initiation with a patient-centered approach and an iterative ongoing process. As Markham suggests, ‘‘putting the human subject
Unable to ascertain demographic characteristics of sample population (e.g., race, education level, etc.) Potential for issues with copyright depending on where the online data was sourced from
squarely in the center of the research both shifts the ethical considerations and allows for socially responsible research’’ (Markham, 2005, p. 815). 3. Conclusions The use of Internet ethnography in studying online illness blogs has numerous implications in the context of nursing science. The use of this method for our exemplar study allowed for a richness in understanding through iterative data analysis and interpretation. Our study focused on the use of digital storytelling to understand the lived experience following cancer in young women, however, an aspect of this method that can be expanded upon in future work is the study of patient-created Internet material as a modality for health information exchange and health communication between patients. The combination of Internet-derived research modalities and patient-centered approaches is an exciting area that nursing science is well-suited to pursue. References Agar, M.H., 1986. Speaking of Ethnography. Sage Publications, Newberrry Park, CA. Bassett, E., O’Riordan, K., 2002. Ethics of Internet research: contesting the human subjects research model. Ethics Inf. Technol. 4 (3), 233–247. Boellstorff, T., Nardi, B., Pearce, C., Taylor, T., 2012. Ethnography and Virtual Worlds. Princeton University Press, Princeton, NJ. Clifford, J., 1986. Introduction: partial truths. In: Clifford, J., Marcus, G. (Eds.), Writing Culture: The Poetics and Politics of Ethnography. University of California Press, Berkeley, CA, pp. 1–26. Cohen, M.Z., Kahn, D.L., Steeves, R.H., 2000. Hermeneutic Phenomenological Research. Sage, Thousand Oaks, CA. Dow, K.H., Kuhn, D., 2004. Fertility options in young breast cancer survivors: a review of the literature. Oncol. Nurs. Forum 31 (3), E46–E53. Eastham, L.A., 2011. Research using blogs for data: public documents or private musings? Res. Nurs. Health 34 (4), 353–361. Emerson, R., Fretz, R., Shaw, L., 1995. Writing Ethnographic Fieldnotes. The University of Chicago Press, Chicago. Eysenbach, G., Till, J., 2010. Ethical issues in qualitative research on Internet communities. Br. Med. J. 10 (323), 1103–1105. Fine, M., Weis, L., Weseen, S., Wong, L., 2000. For whom? Qualitative research, representations and social responsibilities. In: Denzin, N., Lincoln, Y. (Eds.), The Sage Handbook of Qualitative Research. 2nd ed. Sage Publications, Thousand Oaks, CA. Fox, S., Purcell, K., 2010. Chronic Disease and the Internet. Retrieved from: http://www.pewinternet.org//media/Files/Reports/2010/ PIP_Chronic_Disease.pdf (13.10.13).
1692
J. Keim-Malpass et al. / International Journal of Nursing Studies 51 (2014) 1686–1692
Grimes, J., Fleischman, K.R., Jaeger, P.T., 2009. Virtual Guinea Pigs: ethical implications of human subjects research in virtual worlds. Int. J. Internet Res. Ethics 2 (1), 1–19. Gubrium, A., Hill, A., Flicker, S., 2013. A situated practice of ethics for participatory visual and digital methods in public health research and practice: a focus on digital storytelling. Am. J. Public Health (August) . Heilferty, C.M., 2011. Ethical considerations in the study of online illness narratives: a qualitative review. J. Adv. Nurs. 67 (5), 945–953. Hine, C., 2009. How can qualitative Internet research define the boundaries of the project? In: Markham, A., Baym, N. (Eds.), Internet Inquiry, Sage Publications, Los Angeles. Keim-Malpass, J., Albrecht, T.A., Steeves, R.H., Danhauer, S.C., 2013a. Young women’s experiences with complementary therapies during cancer described through illness blogs. West. J. Nurs. Res. 35 (10), 1309–1324. Keim-Malpass, J., Baernholdt, M., Erickson, J.M., Ropka, M.E., Schroen, A.T., Steeves, R.H., 2013b. Blogging through cancer: young women’s persistent problems shared online. Cancer Nurs. 36 (2), 163–172. Keim-Malpass, J., Steeves, R.H., 2012. Talking with death at a diner: young women’s online narratives of cancer. Oncol. Nurs. Forum 39 (4), 373– 378, http://dx.doi.org/10.1188/12.ONF.373-378, 406. Lagu, T., Kaufman, E.J., Asch, D.A., Armstrong, K., 2008. Content of weblogs written by health professionals. J. Gen. Intern. Med. 23 (10), 1642–1646. Lincoln, Y., Guba, E., 1985. Naturalistic Inquiry. Sage Publications, Beverly Hills, CA. Lipinski, T.A., 2009. Emerging legal issues in the collection and dissemination of Internet-sourced research data: part II, tort law issues involving defamation. Int. J. Internet Res. Ethics 2 (1), 1–16. Markham, A., 2005. The methods politics and ethics of representation in online ethnography, 3rd ed. In: Denzin, N., Lincoln, Y. (Eds.), The Sage Handbook of Qualitative Research, 3rd ed., Sage Publications, Thousand Oaks, CA. Markham, A., Baym, N., 2009. Internet Inquiry. Sage Publications, Los Angeles. McWilliam, C., 2009. Patients, persons, or partners? Involving those with chronic disease in their care. Chronic Illn. 5, 277–292. Miller, D., Slater, D., 2000. The Internet: An Ethnographic Approach. Berg Publishing, New York, NY.
Mo, P.K.H., Malik, S.H., Coulson, N.S., 2009. Gender differences in computer-mediated communication: a systematic literature review of online health-related support groups. Patient Educ. Couns. 75 (1), 16–24. Murthy, D., 2008. Digital ethnography: an examination of the use of two technologies for social research. Sociology 42 (5), 837–855. Pennebaker, J., Seagal, J., 1999. Forming a story: the health benefits of narrative. J. Clin. Psychol. 55, 1243–1254. Rains, S., 2014. The implications of stigma and anonymity for self-disclosure in health blogs. Health Commun. 29 (1), 23–31. Ressler, P., Bradshaw, Y., Gualtieri, L., Ho Chui, K., 2012. Communicating the experience of chronic pain and illness through blogging. J. Med. Internet Res. 14 (5), e143. Saiki, L., Cloyes, K., 2014. Blog text about female incontinence: presentation of self, disclosure, and social risk assessment. Nurs. Res. 63 (2), 137–142. Schotanus-Dijkstra, M., Havinga, P., van Ballegooijen, W., Delfosse, L., Mokkenstorm, J., Boon, B., 2013. What do the bereaved by suicide communicate in online support groups? Crisis 1–9. Seale, C., Charteris-Black, J., Macfarlane, A., Mcpherson, A., 2010. Interviews and Internet forums: a comparison of two sources of qualitative data. Qual. Health Res. 20 (5), 595–606. Stanton, A.L., Thompson, E.H., Crespi, C.M., Link, J.S., Waisman, J.R., 2013. Project connect online: randomized trial of an Internet-based program to chronicle the cancer experience and facilitate communication. J. Clin. Oncol. 31 (27), 3411–3417. Thielst, C.B., 2007. Weblogs: a communication tool. J. Healthc. Manag. 52 (5), 287–289. Treadgold, C.L., Kuperberg, A., 2010. Been there, done that wrote the blog: the choices and challenges of supporting adolescents and young adults with cancer. J. Clin. Oncol. 28 (32), 4842–4849. Walther, J.B., 2002. Research ethics in Internet-enabled research: human subjects issues and methodological myopia. Ethics Inf. Technol. 4, 205–216. Whiteman, E., 2007. ‘‘Just Chatting’’: research ethics and cyberspace. Int. J. Qual. Methods 6 (2), 1–9. Ziebland, S., Wyke, S., 2012. Health and illness in a connected world: how might sharing experiences on the Internet affect people’s health? Milbank Q. 90 (2), 219–249.
Contents lists available at ScienceDirect
International Journal of Nursing Studies journal homepage: www.elsevier.com/ijns
Discussion
Internet ethnography: A review of methodological considerations for studying online illness blogs Jessica Keim-Malpass a,*, Richard H. Steeves b, Christine Kennedy c a
University of Virginia School of Nursing, Charlottesville, VA, USA Professor Emeritus, University of Virginia School of Nursing, Charlottesville, VA, USA c Madeline Higginbotham Sly Professor of Nursing, University of Virginia School of Nursing, Charlottesville, VA, USA b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 14 December 2013 Received in revised form 6 June 2014 Accepted 11 June 2014
Background: In recent history, the Internet has emerged as a wealth of archived, ongoing, interactive, and socially mediated data. Conducting Internet ethnography is a fairly new methodological approach, however, it has been previously described as a valid form of inquiry. Illness blogs, in particular, have great implications for nurse researchers, as they are able to study the experience of illness in a naturalistic and longitudinal manner, often with greater detail than data relying solely on participant recall. Participants are able to produce online illness blogs as a way to share their own illness narratives and connect with others going through similar processes. Objectives: The purpose of this paper is to discuss methodological considerations in studying online illness blogs through Internet ethnography. Methods: This article provides an overview of Internet ethnography as an emerging qualitative method and an introduction to research using illness blogs. Through use of this method in an exemplar study of young women with cancer, key decision points are highlighted along with the study team’s field experiences. Conclusion: Issues pertaining to method applicability, active vs. passive involvement as a researcher, ethical considerations, what constitutes data, sampling approach, procedural and analytic decisions, and thoughts regarding reflexivity and voice of the research participants’ are addressed. Strengths and limitations of the study of online illness blogs through Internet ethnography in nursing science are also discussed. ß 2014 Elsevier Ltd. All rights reserved.
Keywords: Qualitative Methods Internet Ethnography Illness blog Blog
What is already known about the topic? Conducting Internet ethnography is a fairly new methodological approach, however, it has been previously described as a valid form of naturalistic inquiry. Illness data captured on the Internet data privileges the current experiences of patients and contain detailed accounts of disease processes and psychosocial ramifications.
* Corresponding author at: P.O. Box 800826, Charlottesville, VA 22908, USA. Tel.: +1 6363466855. E-mail address: [email protected] (J. Keim-Malpass). http://dx.doi.org/10.1016/j.ijnurstu.2014.06.003 0020-7489/ß 2014 Elsevier Ltd. All rights reserved.
What this paper adds This paper explores methodological considerations of Internet ethnography in the context of nursing science using decision points from an exemplar study. This paper examines the need for fluidity, an iterative process, negotiation of study involvement as researcher, and early conceptualization of ethical considerations when using Internet ethnography. This paper describes the advantages in using this method to offer utility in studying experiences among populations that are traditionally difficult to study (e.g., end-of-life experiences, groups with stigma associated with their disease).
J. Keim-Malpass et al. / International Journal of Nursing Studies 51 (2014) 1686–1692
1. Introduction Over the past two decades, the Internet has become the most comprehensive electronic archive of written material and has significantly changed the way people communicate, access information, and live their lives. The daily influx of new information available online represents a wide range of current events, viewpoints, opinions, and virtual communities that allow ordinary citizens to have an active voice with an accessible medium in which to express it. Internet support groups, resources, social networking sites such as Twitter, Facebook, and Instagram, and illness blogs have recently emerged as popular media for patients, their caregivers and health care professionals (Thielst, 2007; Ziebland and Wyke, 2012). Online health resources play a supporting role in many people’s lives as they pursue better health and navigate the healthcare system (McWilliam, 2009). Illness blogs, or blogs that were initiated by users to document their experience with a specific disease, have many purposes and benefits. Blog writers documenting an illness are able to update family and friends through the course of treatment and recovery, become an advocate for and educate others going through the same experience, ask questions and get answers from their online community, and gain the emotional benefits of expressive writing (Pennebaker and Seagal, 1999; Thielst, 2007; Ziebland and Wyke, 2012). Illness blogs also grant a voice to patients through engagement in digital storytelling and socially mediated connection with others (Gubrium et al., 2013; Lagu et al., 2008; Ressler et al., 2012; Treadgold and Kuperberg, 2010). Additionally, because users may remain anonymous if they choose, there are vast implications for self-disclosure and subsequent support of potentially stigmatizing health topics (Rains, 2014; Saiki and Cloyes, 2014). It is estimated that 12% of adults with a chronic health condition maintain their own illness blog, and 28% read peers’ illness blogs (Fox and Purcell, 2010). While blogs are beneficial for the patients who write them (Stanton et al., 2013), they also offer a new modality for health care providers and researchers to understand lived experiences from digital storytelling and sharing of the illness narrative. The Internet as a social domain has extended the reach of human subjects research for social scientists and textual data found on these venues are a rich source for researchers interested in understanding the hermeneutics of patients’ lived experiences (Bassett and O’Riordan, 2002; Eysenbach and Till, 2010). Internet ethnography is a fairly new methodological approach, however, it has been previously described as a valid form of naturalistic inquiry (Markham and Baym, 2009; Miller and Slater, 2000). Previous work in this area has demonstrated that virtual technologies are best studied through a flexible and narrative approach that ethnography lends itself to (Markham and Baym, 2009). When compared to interview qualitative data, it has been noted that Internet data privileges the current experiences of patients and contain detailed accounts of disease processes and psychosocial ramifications (Keim-Malpass et al., 2013a,b). Unlike in-person interviews, researchers
1687
do not have to rely on the participants recall (Seale et al., 2010). Because of the ease with which large amounts of archived materials can be accessed and analyzed, this source of data has considerable potential for direct observation of illness experiences (Seale et al., 2010). This paper will discuss methodological considerations of Internet ethnography for use in nursing science by examining elements of an exemplar study involving Internet illness blogs of young women with cancer. Findings from this study have been previously reported (Keim-Malpass and Steeves, 2012; Keim-Malpass et al., 2013a,b) and methodological elements requiring decision were based on the study team’s experience and documented in field notes throughout the process. To the authors’ knowledge, this research was the first to use Internet illness blogs for hermeneutic interpretation immersed in an ethnographic context. In order to advance use of this unique and innovative methodology, key decision points are highlighted (Fig. 1), along with the study team’s experiences in the field and strengths and limitations of the use of this method (Keim-Malpass and Steeves, 2012; Keim-Malpass et al., 2013a,b). 2. Methodological considerations 2.1. Introduction to Internet ethnography As new forms of Internet ethnography begin to emerge, much of their epistemological underpinnings remain the same as traditional forms of ethnography. Ethnography
Fig. 1. Key decision points when using online ethnographic methods.
1688
J. Keim-Malpass et al. / International Journal of Nursing Studies 51 (2014) 1686–1692
focuses on understanding social stories and what is going on in the field, and this remains true both on and offline (Agar, 1986; Murthy, 2008). Traditionally, an ethnographer is thought of as embarking in field work in a community and seeking to understand the culture that she encounters by making sense of social actions from the point of view of the participants (Clifford, 1986). Within the past fifteen years, the Internet has changed the way qualitative data is both conceptualized and collected while also impacting how ethnographies can take place. Without a physical boundary defining place, ethnographers doing research online must still immerse themselves in the social situation being studied to learn the socially relevant things that are happening there (Hine, 2009). Logistically speaking, the notion that the Internet can erase all geographic boundaries and allow ethnographers to connect with people instantaneously has many advantages. It allows for increased access to various perspectives, including those of people who may not have participated in traditional face-to-face research interviews, or those who may be stigmatized due to their disease or circumstances (Schotanus-Dijkstra et al., 2013). It also provides a method by which international sampling and global perspectives can be easily accessed. 2.2. Is the method applicable: population identification, involvement of researcher, and ethical considerations Prior to embarking in Internet ethnography, researchers must first determine if the population of interest can be self-defined and ensure that the online community exists. In other words, is the population of interest readily accessible (through self-identification) in the online environment and do the members of this community interact with one another (Markham and Baym, 2009)? Traditional clinical research relies on having very strict inclusion and exclusion criteria, and admittedly, the inclusion/exclusion criteria may need to be modified for online ethnographic work. In the exemplar study involving women diagnosed with cancer from ages 20 to 39, the study team had to accept an age range of diagnosis as opposed to having very strict clinical knowledge regarding precise dates of diagnosis (Keim-Malpass and Steeves, 2012). From spending many months in the field, the lead author ascertained that women often did describe their age at diagnosis, but sometimes did not specify a particular age, instead using an age-range when diagnosed (i.e., diagnosed with ovarian cancer in my late twenties). Therefore, the inclusion criteria necessitated flexibility by virtue of not having access to specific clinical data elements (Keim-Malpass and Steeves, 2012). Additionally, the lead author spent initial time understanding the context in which young adults with cancer communicate with one another in the online environment. This step was critical in determining that the population of interest had an active presence online, could be selfidentified, and regularly used social media and their blogs to communicate with one another (Keim-Malpass and Steeves, 2012). The next step for ethnographers in virtual fields is to consider how active to be in relation to the participants
they are studying (Hine, 2009). Three different types of Internet-based research methodologies have been described: (1) passive analysis that involves understanding how information and dialogs are exchanged on discussion groups or blog sites without the researchers actually involving themselves, (2) active analysis in which the researcher actively is involved in participation and dialog exchange, and (3) self-identified active analysis where researchers identify themselves and their research interests a priori which may allow them to also gather information in a more in-depth interview or focus group setting or use the Internet to recruit participants for clinicbased studies (Markham and Baym, 2009). The difference in levels of active participation by the investigator has various methodological and ethical implications. In the exemplar study, the decision was made to be passive investigators and analyze illness blogs that were already in existence and located on public Internet sites (KeimMalpass and Steeves, 2012; Keim-Malpass et al., 2013a). This decision was based on the study team’s interest in a naturalistic inquiry to further understand the lived experience of young women with cancer without any preconceived interview questions to direct responses. This approach follows the key assumption that ethnographic methodology should use the entire experience of immersion to learn how life is lived, rather than coming in with particular a priori assumptions of the issues that are important to the group being studied (Hine, 2009). This form of inquiry was also critical in understanding the experience of cancer from a longitudinal perspective because we were able to follow the young women nonconcurrently, meaning that analysis began at the time of diagnosis and moved forward in time to the most recent blog posting. This method allowed us to follow as much of the disease process as possible (up to 4 years for some women), from the time participants were diagnosed through treatment, and transitioning into survivorship or death, noting experiences with recurrence and remission, and other setbacks and milestones along the way (Keim-Malpass et al., 2013b). After deciding on whether to conduct passive or active Internet ethnography, Institutional Review Board (IRB) approval is the necessary next step. The study team was careful to consider the ethical implications and ramifications of conducting an Internet ethnography using publically available Internet data prior to initiation of the study. These considerations tend to focus on the challenges of privacy, confidentiality and informed consent and are very dependent on where the data is being obtained, the level of privacy the author of the content assumes, and how active the researcher is in relation to the participants (Eastham, 2011; Heilferty, 2011). The exemplar study was submitted to the IRB and need for approval was exempted due to the public nature of the blog sites and because the study team was not directly interacting with participants to ask any specific questions (i.e., we were using passive analysis). The ethical implications of online data for social science research are evolving, and of note, IRBs have not yet shown standard methods in granting approval Internet-based observational studies. The study team then had to make a decision about permissions
J. Keim-Malpass et al. / International Journal of Nursing Studies 51 (2014) 1686–1692
needed if the blog included names and an email address. Some bloggers chose to use an alias, or a name not representing their real identity, and thus did not have email addresses present. Their privacy and anonymity was respected and no contact was made. However, if the blog did include a name and email address, the decision was made to contact the blog participant to ask for permission for their blog to be included in our study. Participants with accessible emails responded favorably and all granted permission for inclusion. Following IRB approval, the lead author became immersed in the Internet culture of young women with cancer (diagnosed between the ages of 20 and 39) for roughly six hours per week over a time period of four months before the actual blog analysis began. This process began at several key young adult cancer websites (stupidcancer.com, planetcancer.org, livestrong.org, youngsurival.org). The lead author followed the general content on the website, listened to weekly and special content Podcasts hosted by young adults with cancer, followed the live chats of survivors during the Podcasts, and followed postings on social networking sites such as Facebook and Twitter. Twitter searches were initiated, through ‘following followers’ of authors that were self-described as diagnosed with cancer during young adulthood. ‘Following followers’ is a method of connection on Twitter that allowed the user to see who follows certain individuals or young adult cancer organizations. Many of the Twitter users obtained through this method were either interested in young adult cancer survivor issues or self-identified as young adult cancer patients/survivors. This initial time of immersion was critical in understanding how young adults with cancer express themselves, describe issues that are important to them, communicate among each other, and represent themselves online. The study team was able to understand how the Internet allowed users to construct their social network around their experience (Keim-Malpass and Steeves, 2012). This time of immersion was central in developing and mapping online communities as a part of the ethnographic process (Markham, 2005). 2.3. What are data? Procedural considerations, sampling and analytic strategy During the time period of immersion, the next decision involved determining what type of data would be included in the analysis. Many of the illness blogs contained narrative text in diary form, access to photos, video blogs (videos of themselves created and often linked to YouTube), hyperlinks to websites, and other illness blogs the blogger followed, meaning that numerous items found on the blog could be considered data for analysis. All of these data elements are important and represent the digital story the young women were telling about themselves, however, the decision was made by the study team to limit the data for hermeneutic analysis to the narrative text found on the illness blogs (Cohen et al., 2000). Narrative text included actual blog entries, as well as comments posted on each blog entry site.
1689
Sampling considerations were also considered and further refined during the antecedent period of immersion. Potential blog sites were initially ascertained for inclusion using traditional inclusion/exclusion criteria but then had an additional sampling criteria applied including: three out of four randomly selected blog entries focused specifically on the cancer experience, the blog title or biography of the blog author highlighted the cancer focus, the blog author indicated cancer type and her age (or age range) when the cancer was diagnosed as well as current age (or age range) (Keim-Malpass et al., 2013a,b). The study used a snowball convenience sample to obtain the sample, which began with initial blogs that were highlighted on a young adult cancer website (stupidcancer.org). This purposive strategy was used to initially sample a small number of the young women who met inclusion criteria, then subsequently include women who were socially connected (followed each other Twitter) and actively engaged online (regularly commented on blog entries, conversed regularly on Twitter) with the original sample. Through the online medium, the snowball approach is achieved through social network mapping and identifying other members who are socially engaged with the initial participant. Sample ascertainment for the exemplar study is described in detail in previous publications (Keim-Malpass and Steeves, 2012). The final sample included sixteen illness blogs of women diagnosed with cancer who had maintained their blog on average for over two years (Keim-Malpass and Steeves, 2012). Central to the period of immersion, data collection, and analysis was the process of bracketing preconceived ideas and judgments about the research phenomenon in a field notebook, and this ritual was a daily process for the lead author (Cohen et al., 2000). Additionally, observations of the participants were made through data gathered from their blog and public Twitter handle (if applicable). Data from the blog sites were read and abstracted non-concurrently, meaning data originated from the time of diagnosis (or first blog posting related to cancer) and moved forward in time to present day. The timing was important to the investigators because we wanted to limit bias among those women who had been asked for permission to be included in the study by not analyzing data that was written after permission was ascertained. While reading each blog entry, the lead author archived and managed text data and supporting photographs and videos (for context) in a Word data file. All textual data (including blog entries and field notes) were reviewed multiple times, then analyzed to identify nodes, or individual units of meaning (Cohen et al., 2000). The nodes were then grouped according to categories based on similarities and then eventually into themes that expressed an overall meaning or interpretation (Cohen et al., 2000). Additionally, memos were maintained outlining how developed themes applied to previous findings in the literature and how the themes highlighted clinical, social and cultural aspects of young adult cancer survivorship. At this point, consensus was reached among investigators that saturation in themes had been achieved and no additional sampling would be needed.
1690
J. Keim-Malpass et al. / International Journal of Nursing Studies 51 (2014) 1686–1692
2.4. Online identity: voice and meaning From the beginning of the research analysis, the notion of online identity of the participants was considered. A blog site is a space in which to create an online identity that may or may not be consistent with the user’s offline identity. The users can choose how they wish to be represented, rather than allowing traditional demographic and socioeconomic factors such as race, gender, class, and body type to dictate how they are initially perceived by others, including researchers (Markham, 2005; Mo et al., 2009). Computer-mediated construction of self, other, and social structures are initiated by the user, whether those choices are conscious or not (Markham, 2005). In this study, text was central to the person’s construction, story, and negotiation of identity; however the design of the blog, the blog title, pictures associated with the blog, badges on the blog, lists of followers, and level of anonymity (real names/emails versus aliases) all contributed to the social construction of the online identity of the participant. There were numerous expressions of identity based on the design of the blogs. One blog was designed in all black with a picture of an atomic bomb explosion located on the profile. Another was reminiscent of a cabin in the woods. Some were scattered with personal photos. The narrative texts of the young women were equally as diverse, ranging from cynical language, to humor, to curse words, to words of comfort and contentment. Incorporating the construction of online identity into the ethnographic analysis became central in understanding the social and cultural context of illness blogs, and offers a unique strength to this methodology (Boellstorff et al., 2012). Created identities through an online medium represent an evolving approach to participant observation as a component of ethnographic research (Boellstorff et al., 2012). Social scientist and early Internet ethnographer, Markham (2005), suggests that researchers should continuously reflect on the interests of the research participants. She emphasizes the need to shift from re-representing existing concepts in the literature to focus on advocacy and understanding how this emerging body of literature can influence positive change by asking, ‘‘whose interests does this research serve and how can my writing and publishing give voice to those who might otherwise remain silent?’’ (Markham, 2005). Continually reflecting on these questions is central to online ethnographic research design and adherence to these principles hopefully achieves an ethically grounded and reflexive body of work (Lincoln and Guba, 1985). These concepts offer the reminder that the researchers’ voice is a privileged one and it remains paramount to give a voice to participants rather than unintentionally speaking for them (Fine et al., 2000). As noted previously, the lead author used a field notebook to continuously write down thoughts, perceptions and emotions during the initial months of immersion, during the time spent transcribing and reading blog narratives, as well as during analysis and dissemination phases. This helped in bracketing pre-conceived notions about ‘what was going on’, as well as providing ongoing iterative processing, enhancing trustworthiness, and limiting bias (Emerson et al., 1995; Lincoln and Guba, 1985).
Additionally, feminist undertones in the textual analysis were noted as they are necessary to relate the centrality of gender in the shaping of consciousness (Fine et al., 2000). Gender was a unique aspect to this study because of potential disruptions of roles in parenting young children through a cancer diagnosis and possible loss of fertility, which is often a more pronounced problem for young women cancer survivors (Dow and Kuhn, 2004). To incorporate this aspect of the study, gender was a central analytic lens that was often written about within the context of the field notes. Additionally, an unanticipated finding emerged during the process of initial analysis and dissemination. During the original study period, two of the young women died from their cancer diagnosis, yet continued to maintain their blog close to the time of their death. Their writings presented a unique perspective during the end-of-life time period that is often difficult for researchers to gain access to due to perceived burden to study participants. Sadly after initial analysis concluded, three additional young women died due to their disease, each representing widely varying end-of-life trajectories. To incorporate this emerging method in a way that enhances access to such populations during vulnerable stages of their disease trajectory (i.e., end-of-life), the lead author is currently in the process of conducting a post hoc analysis to gain insight from a time period. This lens is critical in giving voice to young women’s end-of-life experiences. 2.5. Strengths and limitations of method ‘‘The relationships that I formed online were probably the most true social interactions and relationships I have had. I was the most real with them. Perhaps it was the anonymity of the Internet that made it safe, perhaps it was the fact that they too had gone through something similar that made me feel connected. Whatever the reason, in my writings, I could open up to them about my thoughts and feelings. I have ‘‘met’’ many people whose lives have been touched by cancer. And with each introduction, discussion, comment, email, etc., I feel a commonality that I had not felt in my everyday life. I greatly appreciate this ridiculous world of the Internet’’ J., ovarian cancer survivor and blogger diagnosed at age 26 The use of Internet ethnography as an emerging qualitative methodology has numerous strengths (Table 1). Accounts documented in online illness blogs in this exemplar study were raw and honest and the blog authors were connected and immersed in each other’s cancer experience (Keim-Malpass and Steeves, 2012; Keim-Malpass et al., 2013a,b). Participants in the exemplar study were able to discuss many topics openly online that may have been difficult to discuss in traditional face-toface qualitative interviews (i.e., troubling symptoms such as anal fistulas, relationship difficulty, confrontation of mortality) (Keim-Malpass and Steeves, 2012; KeimMalpass et al., 2013a). Illness bloggers have the ability to choose when they want actively engage in emotional
J. Keim-Malpass et al. / International Journal of Nursing Studies 51 (2014) 1686–1692
1691
Table 1 Strengths and limitations of studying online illness blogs. Strengths
Limitations
No geographic boundaries Allows researchers to study those who may decline from participating in traditional qualitative study due to stigma Allows for the ultimate naturalistic inquiry while studying topics that are of importance to the participant No recall bias, participants write about the experience as it happens
Difficulty in having narrow inclusion/exclusion criteria Unable to confirm identity or validate diagnosis, treatment modalities, symptom-based experiences Digital divide exists – are only those with Internet/mobile access able to communicate online? Who is not represented online? Evolving ethical principles – no true consensus yet among IRBs, unable to generally provide informed consent when using completely public websites Findings may not be generalizable
Researchers can experience a disease in a longitudinal sense from diagnosis on while analyzing it non-concurrently Allows for low participant burden and unique perspectives (e.g., end-of-life experiences) Allows analysis of social support connection through online social networks
catharsis through expressive writing, and engage in social support mediated through blog comments and online communication on social media sites (e.g., Twitter, Facebook, Instagram). Markham (2005) suggests that study participants’ online persona can be fluid and we are able to bear witness to a range of emotion. Choosing levels of anonymity in online environments also gives control in the presentation of self (Markham, 2005). In doing so, participants also have the ability to disengage with Internet activity, lurk on other sites (read without actively participating) and/or re-engage when ready. The longitudinal, naturalistic and immersive approach in studying illness blogs through Internet ethnography is unmatched in terms of scope and range in understanding lived experiences. This method has very low participant burden, limited recall bias, there are no geographic boundaries limiting connection thereby increasing the ability of the researcher to incorporate diverse views in a global context. Further, it allows for analysis of longitudinal illness narratives and mapping of online social connectivity and social support derived and maintained through the Internet. When approaching this research using passive analysis and prolonged ethnographic immersion, the experiences and concerns of the participants can be privileged for an encompassing view of their lived experience. Despite this, limitations exist in using online modalities to answer socially and clinically relevant questions (Table 1). There may be difficulty in obtaining appropriate sample, findings are often not generalizable, demographic characteristics of the sample are very difficult to obtain and online identity cannot be validated. Due to the digital divide, only those with regular access to the Internet through computers or mobile phones and a conversant level of literacy are likely engaging in digital storytelling on illness blogs and online social networking. Two issues that are case-dependent are ethical issues and concerns over copyrighted data (Bassett and O’Riordan, 2002; Eastham, 2011; Grimes et al., 2009; Gubrium et al., 2013; Heilferty, 2011; Lipinski, 2009; Markham and Baym, 2009; Walther, 2002; Whiteman, 2007). Both of these considerations must be completely explored a priori prior to study initiation with a patient-centered approach and an iterative ongoing process. As Markham suggests, ‘‘putting the human subject
Unable to ascertain demographic characteristics of sample population (e.g., race, education level, etc.) Potential for issues with copyright depending on where the online data was sourced from
squarely in the center of the research both shifts the ethical considerations and allows for socially responsible research’’ (Markham, 2005, p. 815). 3. Conclusions The use of Internet ethnography in studying online illness blogs has numerous implications in the context of nursing science. The use of this method for our exemplar study allowed for a richness in understanding through iterative data analysis and interpretation. Our study focused on the use of digital storytelling to understand the lived experience following cancer in young women, however, an aspect of this method that can be expanded upon in future work is the study of patient-created Internet material as a modality for health information exchange and health communication between patients. The combination of Internet-derived research modalities and patient-centered approaches is an exciting area that nursing science is well-suited to pursue. References Agar, M.H., 1986. Speaking of Ethnography. Sage Publications, Newberrry Park, CA. Bassett, E., O’Riordan, K., 2002. Ethics of Internet research: contesting the human subjects research model. Ethics Inf. Technol. 4 (3), 233–247. Boellstorff, T., Nardi, B., Pearce, C., Taylor, T., 2012. Ethnography and Virtual Worlds. Princeton University Press, Princeton, NJ. Clifford, J., 1986. Introduction: partial truths. In: Clifford, J., Marcus, G. (Eds.), Writing Culture: The Poetics and Politics of Ethnography. University of California Press, Berkeley, CA, pp. 1–26. Cohen, M.Z., Kahn, D.L., Steeves, R.H., 2000. Hermeneutic Phenomenological Research. Sage, Thousand Oaks, CA. Dow, K.H., Kuhn, D., 2004. Fertility options in young breast cancer survivors: a review of the literature. Oncol. Nurs. Forum 31 (3), E46–E53. Eastham, L.A., 2011. Research using blogs for data: public documents or private musings? Res. Nurs. Health 34 (4), 353–361. Emerson, R., Fretz, R., Shaw, L., 1995. Writing Ethnographic Fieldnotes. The University of Chicago Press, Chicago. Eysenbach, G., Till, J., 2010. Ethical issues in qualitative research on Internet communities. Br. Med. J. 10 (323), 1103–1105. Fine, M., Weis, L., Weseen, S., Wong, L., 2000. For whom? Qualitative research, representations and social responsibilities. In: Denzin, N., Lincoln, Y. (Eds.), The Sage Handbook of Qualitative Research. 2nd ed. Sage Publications, Thousand Oaks, CA. Fox, S., Purcell, K., 2010. Chronic Disease and the Internet. Retrieved from: http://www.pewinternet.org//media/Files/Reports/2010/ PIP_Chronic_Disease.pdf (13.10.13).
1692
J. Keim-Malpass et al. / International Journal of Nursing Studies 51 (2014) 1686–1692
Grimes, J., Fleischman, K.R., Jaeger, P.T., 2009. Virtual Guinea Pigs: ethical implications of human subjects research in virtual worlds. Int. J. Internet Res. Ethics 2 (1), 1–19. Gubrium, A., Hill, A., Flicker, S., 2013. A situated practice of ethics for participatory visual and digital methods in public health research and practice: a focus on digital storytelling. Am. J. Public Health (August) . Heilferty, C.M., 2011. Ethical considerations in the study of online illness narratives: a qualitative review. J. Adv. Nurs. 67 (5), 945–953. Hine, C., 2009. How can qualitative Internet research define the boundaries of the project? In: Markham, A., Baym, N. (Eds.), Internet Inquiry, Sage Publications, Los Angeles. Keim-Malpass, J., Albrecht, T.A., Steeves, R.H., Danhauer, S.C., 2013a. Young women’s experiences with complementary therapies during cancer described through illness blogs. West. J. Nurs. Res. 35 (10), 1309–1324. Keim-Malpass, J., Baernholdt, M., Erickson, J.M., Ropka, M.E., Schroen, A.T., Steeves, R.H., 2013b. Blogging through cancer: young women’s persistent problems shared online. Cancer Nurs. 36 (2), 163–172. Keim-Malpass, J., Steeves, R.H., 2012. Talking with death at a diner: young women’s online narratives of cancer. Oncol. Nurs. Forum 39 (4), 373– 378, http://dx.doi.org/10.1188/12.ONF.373-378, 406. Lagu, T., Kaufman, E.J., Asch, D.A., Armstrong, K., 2008. Content of weblogs written by health professionals. J. Gen. Intern. Med. 23 (10), 1642–1646. Lincoln, Y., Guba, E., 1985. Naturalistic Inquiry. Sage Publications, Beverly Hills, CA. Lipinski, T.A., 2009. Emerging legal issues in the collection and dissemination of Internet-sourced research data: part II, tort law issues involving defamation. Int. J. Internet Res. Ethics 2 (1), 1–16. Markham, A., 2005. The methods politics and ethics of representation in online ethnography, 3rd ed. In: Denzin, N., Lincoln, Y. (Eds.), The Sage Handbook of Qualitative Research, 3rd ed., Sage Publications, Thousand Oaks, CA. Markham, A., Baym, N., 2009. Internet Inquiry. Sage Publications, Los Angeles. McWilliam, C., 2009. Patients, persons, or partners? Involving those with chronic disease in their care. Chronic Illn. 5, 277–292. Miller, D., Slater, D., 2000. The Internet: An Ethnographic Approach. Berg Publishing, New York, NY.
Mo, P.K.H., Malik, S.H., Coulson, N.S., 2009. Gender differences in computer-mediated communication: a systematic literature review of online health-related support groups. Patient Educ. Couns. 75 (1), 16–24. Murthy, D., 2008. Digital ethnography: an examination of the use of two technologies for social research. Sociology 42 (5), 837–855. Pennebaker, J., Seagal, J., 1999. Forming a story: the health benefits of narrative. J. Clin. Psychol. 55, 1243–1254. Rains, S., 2014. The implications of stigma and anonymity for self-disclosure in health blogs. Health Commun. 29 (1), 23–31. Ressler, P., Bradshaw, Y., Gualtieri, L., Ho Chui, K., 2012. Communicating the experience of chronic pain and illness through blogging. J. Med. Internet Res. 14 (5), e143. Saiki, L., Cloyes, K., 2014. Blog text about female incontinence: presentation of self, disclosure, and social risk assessment. Nurs. Res. 63 (2), 137–142. Schotanus-Dijkstra, M., Havinga, P., van Ballegooijen, W., Delfosse, L., Mokkenstorm, J., Boon, B., 2013. What do the bereaved by suicide communicate in online support groups? Crisis 1–9. Seale, C., Charteris-Black, J., Macfarlane, A., Mcpherson, A., 2010. Interviews and Internet forums: a comparison of two sources of qualitative data. Qual. Health Res. 20 (5), 595–606. Stanton, A.L., Thompson, E.H., Crespi, C.M., Link, J.S., Waisman, J.R., 2013. Project connect online: randomized trial of an Internet-based program to chronicle the cancer experience and facilitate communication. J. Clin. Oncol. 31 (27), 3411–3417. Thielst, C.B., 2007. Weblogs: a communication tool. J. Healthc. Manag. 52 (5), 287–289. Treadgold, C.L., Kuperberg, A., 2010. Been there, done that wrote the blog: the choices and challenges of supporting adolescents and young adults with cancer. J. Clin. Oncol. 28 (32), 4842–4849. Walther, J.B., 2002. Research ethics in Internet-enabled research: human subjects issues and methodological myopia. Ethics Inf. Technol. 4, 205–216. Whiteman, E., 2007. ‘‘Just Chatting’’: research ethics and cyberspace. Int. J. Qual. Methods 6 (2), 1–9. Ziebland, S., Wyke, S., 2012. Health and illness in a connected world: how might sharing experiences on the Internet affect people’s health? Milbank Q. 90 (2), 219–249.