Participant-generated, self-made videos engender powerful, often highly emotional, reactions from viewers who experience a stronger connection and identification with participants and their experiences than we have ever achieved with researcher-shot footage. Reactions have ranged from shock, discomfort, and offers of Freudian psychological analyses to laughter, immediate recognition and discovery. Through several video examples from recent fieldwork we explore the reasons for this heightened reaction, and raise questions related to representation, authenticity, intimacy and the role of the ethnographer in the age of YouTube, social networking sites, and reality TV. What is the ethnographer’s role when participants share their lives in videos we request that are stylistically similar to online user-generated content? What is that ethnographer’s ‘Do’, and what role does she play in editing, framing and presenting these videos? How do participants conceptualize what they are creating?
Ever since we got a peek inside Nanook’s igloo, walked down a long corridor toward a hall full of screaming supporters with John F. Kennedy, witnessed the Loud parents break-up in the middle of a fight, and peered inside the bedroom of the first Real World, viewers have been fascinated by intimate glimpses into the private lives of others. (Flaherty 1922; Drew Associates 1960; Raymond & Raymond 1973, MTV 1992) In corporate ethnography, stories from the field, photographs, and video have long been used to bring the world of research participants to life. Their habits, practices, joys and frustrations come alive, especially through video, and help us communicate insights to colleagues and customers. As ethnographers at Intel Corporation, the content of our videos has generally consisted of interviews with research participants and observational footage of their daily practices in the home. In the last few years a new, transmissive part of our research methodology has driven our video documentation practice in a different direction as we supplement our traditional Ethnographer-Made Video (EMV) with Participant-Made Video (PMV). In 2008 we began taking advantage of new, inexpensive video technologies that allow our research participants to share their everyday lives by filming themselves. Equipping research participants with cameras, both still and video, is not a new practice either in anthropological research (Collier & Collier 1967; Worth & Adair 1970, 1972; Frota 1995; Pink 2007) or in other disciplines invested in ethnographic research methods, such as user-centered design (Brun-Cottan and Wall 1995; Ylirisku & Buur 2007; Raijmakers, Gaver Bishay 2006; Bean 2008), and HCI (Taylor, Wiche, Kaye 2008; Reponen, Lehikoinen, Impiö 2007). Additionally, education researchers and educators (including anthropology professors!) have embraced low-cost video cameras as part of educational training with students in and beyond the classroom (Rowell 2009; Durrington 2009). Recognizing the multiple ways videos made with Flip video brand cameras are used to enhance education experiences (“From video book reviews and school news reports to teacher observations for professional development,”) Pure Digital, makers of Flip Video cameras, offers discounted prices to US based educators so they can use video as a “hands-on way for students and teachers to engage more deeply”.1
This marketing rhetoric rings true for how PMVs have enabled us as ethnographers to engage more deeply with our stakeholders. The videos our participants make have a sense of immediacy and intimacy, and elicit emotional responses and curiosity to learn more on the part of our stakeholders. Unlike our EMVs, the videos our research participants make using video cameras are not talking head interviews or footage of participants pointing and explaining how they do a particular activity; instead they offer a glimpse of participants doing activities they normally just talk about when we are there. We increasingly find that we use more PMV in our research presentations than traditional EMV primarily because they engender powerful and often highly emotional reactions from viewers and transfix our stakeholders who experience a stronger connection and identification with participants and their experiences than we have ever achieved with footage shot by researchers. We use these attention-grabbing videos to engage engineers, sales representatives, and key decision-makers internal to Intel Corporation, and drive key research insights and recommendations more effectively than we have been able to through slides and photographs alone.
Stakeholders react to PMVs as less mediated, and more real, than the footage we have traditionally shot, though these highly reflexive videos are as consciously staged as a slice of reality as our EMV. The politics and problematics of self-presentation, have been well explored, and clearly PMVs are another manifestation of identity work, and performativity (Butler 1990, 1997). Research participants often model their performances in these videos on widely available user-generated content (UGC) they are familiar with on the Internet. Indeed, the popularity of online user-generated videos including video diaries, and new ways to distribute and share those videos on YouTube and Facebook serve as inspiration or, at the very least, an example for research participants when they are faced with the task of creating their own self-reflexive videos. Reality TV shows and their individual, direct-to-camera confessions are another illustration for participants. Our participants follow several documentary video conventions consistent with UGC and reality TV, and in the absence of visual and audio cues of a middleman – the guiding voices and presence of the ethnographers –PMVs can initially appear indistinguishable from the types of videos stakeholders are familiar with from content sharing sites like YouTube.
While the EMVs we produce of our in-home conversations make clear the nature of the interaction and power dynamics between the ethnographers and participants, (rendered visible through our presence and guidance of the conversation) in the PMVs the nature of the ethnographer/participant relationship is masked; we are not present, and the relationship primarily unfolds off-camera. The heightened intimacy and engagement our stakeholders experience with these videos poses new challenges for how we, as ethnographers, consistently and assertively guard the integrity of these research materials and our relationships with our research participants.
Asking research participants to shoot their own videos is not new; and these videos are not more or less real, complete, or genuine than other ethnographic representations. What is significant and revelatory about participant-made videos is the privileged view they offer of intimate moments in our research participants’ lives – moments we are not privy to when we are physically present in their homes and have previously not been able to share with stakeholders in such a direct and visually rich way. During a home interview, we may see how people lie on their living room floor to watch TV; in a PMV we may see them late at night in the bedroom, half asleep watching TV. No matter how much these videos build on shared cultural templates for how to properly share intimate moments with the world, (or at least large, unknown audiences such as YouTube and Reality TV viewers) these videos are the product of our relationships with our research participants, produced for very specific and clearly defined audiences that are always directly tied to the ethnographers’ presence and presentation. While we, as ethnographers, are less present in the actual videos than in our EMVs, our presence is arguably more urgently required when viewing these videos than with other representations of participants’ lives that we produce in our ethnographic practice. Because our research participants entrust us with personal, private views of their home life that are intimate and engaging for our stakeholders, we take great care in how we frame, interpret and share them.
SOLICITING PARTICIPANT-MADE VIDEOS
We first added PMVs to our field methods in 2008 as a variation on a fairly standardized three-part engagement with households. In a given project, we visit each household twice during the course of a roughly two to three week period while we are in field locations that can range from Phoenix to Jogjakarta. The first visit generally consists of an open-ended ethnographic interview, home tour and tour of other relevant locations, during which we video tape and take still photos. We end the first visit with a request that the participants complete a research exercise before we return in approximately 7-10 days. These exercises have included photo diaries, mapping exercises, and video questionnaires using Flip Cameras. Many of the between-visit research exercises we ask participants to complete result in artifacts we use only with them and rarely share with stakeholders. We use exercises to help us engage with our participants – to start a conversation, to probe on topics and practices that come up during our first interviews that we feel we do not fully understand. We discuss with participants how we will use these materials, and participants sign written release forms that clearly detail how the materials will be used. They are also given copies of these release forms for their records. The release forms detail that the materials can be used for Intel internal presentations as well as external presentations such as business meetings, conferences and research publications, but that digital (or analog) copies of the materials will not be distributed beyond the researchers, and that materials will not be posted on the Internet or used in any marketing campaigns. Participants understand that we will show these materials to various audiences in the course of our work, but that we will not further distribute or publish the materials. During the second interview, we watch the videos together with the participants and discuss the content – both the scenes and activities they recorded, as well as further questions these videos prompt us to ask. Occasionally, participants have recorded a video they then decide at the second interview they do not want us to have. We then immediately erase the video from the camera and our laptop so no copies exist.
With the raw PMV footage we elicit from research participants we edit and create video artifacts that we use in a number of ways. While we have not created artifacts with complex editing or a re-worked narrative structure like the Design Documentaries described by Raijmakers, Gaver, Bishay (2006) or the portraits described by Yliriski and Buur (2007), we always edit videos we present to stakeholders — from tightening-up the pace on a single clip, to audio narration, subtitle overlays, and montages created from multiple videos of one or several participants.
The most common use of these video artifacts is to describe our practice and the user experience definition process as illustrations of the type, and breadth, of ethnographic work related to television practices we have conducted in the last five years, and to illustrate the type of ethnographic work that is the starting point for our team’s user-experience focused innovation process. We also use these videos with product development teams to illustrate the physical, social, cultural, and technological contexts in which the product they are developing will be used. The videos are short, (1-3 minutes, occasionally up to 20 minutes long,) and are designed to prompt participants’ actions and reactions to specific tasks or questions rather than systematically capture ongoing interactions with technology that can be analyzed frame-by-frame and used in a product design process.
HEIGHTENED ENGAGEMENT WITH SELF-MADE VIDEOS
We value the videos our participants have made for us, and our stakeholders are captivated by them, because they depict participants doing things they normally only describe to us during in-home visits. They are consistently more entertaining and engaging to watch than the footage we shoot during interviews. For example, over the course of a three-hour interview with the Sudah family outside of Tokyo we learned of several ways members watch television throughout a typical day – in the morning, while doing household chores, while driving to work, while cooking and eating dinner, and in bed before falling asleep at night. While useful information, none of this (besides perhaps watching in the car) seemed particularly interesting, and none of the footage of the Sudahs describing these activities is visually gripping. However, the PMVs the Sudahs subsequently made for us doing all the types of TV viewing they described during our interview arrest our viewers every time we show them. In research report-outs, in customer meetings, and in internal corporate events promoting user experience research, we have used a version of these videos edited into one longer clip showing how the Sudahs watch TV during a typical day. We have also used selected video segments to illustrate how ordinary, personal, and extremely intimate television viewing can be. All of the Sudah family’s videos provoke strong reactions from viewers. The images resonate with viewers’ own lives – cooking with TV; folding laundry with TV, and eating dinner with TV. They also bring forth a more complicated mix of desire, amusement and disapproval incited by footage of the Sudahs watching TV while driving to work, and a scene of Mr. Sudah watching TV before bed while his adolescent daughter gives him a foot massage. In the US, we are routinely asked if we have intervened to stop Mrs. Sudah from watching TV while driving, and have been advised about Freudian theory and the psychological damage the Sudah daughter is incurring by massaging her father’s feet.
As the reactions to the Sudahs’ PMVs illustrate, one of the most important reasons our PMVs are useful and valuable is the heightened engagement they provoke with viewers. Viewers find the videos more compelling than our EMVs, as participants are depicted engaging in daily activities or actively engaged in illustrating in situ what they usually do, rather than responding to interview questions. The videos generate emotional reactions and practical questions such as, “Is it legal to drive a car and watch TV at the same time in Japan?” These questions help us engage with stakeholders and make them more receptive to our research-based business recommendations. In the end, the PMVs become as much prompts for starting in-depth discussions with stakeholders as they are probes we use with participants.
We also value PMVs because participants experience a more heightened engagement in sharing their experiences with us than they do when we film, as they take an active role in constructing how they will be portrayed. They are conscious that they are performing and that this performance has an audience beyond the two ethnographers currently in their home. Anticipating these audiences, participants are doing much more with these exercises than the actions and words captured on video. They are creating intentional representations of who they want others to imagine them to be. When asked about creating PMVs, participants tell us that they are more self-conscious when they film themselves than when we film them during interviews. Brett, a 28-year old self-employed event planner in London explains it this way:
Brett: When you’re here, I just forgot that you’re pointing a camera at me. But when you’re doing it yourself, you are kind of conscious of it all the time. Aren’t you? And trying to get it right, so it’s sort of a video, something that might be interesting…if anything, it’s easier when you’re here because I don’t have to think about it (the videotaping.)
Another Londoner, 18-year old student Ray, takes it a step further.
Researcher: How did (making videos of yourself) feel for you?
Ray: Knowing that I was being watched, I tried to act as if I wasn’t being watched, having to do it myself. I did it all myself as well, so I set the camera up, walked over to what I was doing before and then after a minute or so walked back around it and turned it off. It was fun, awkward at times. I think there was two I deleted and re-did just because I looked a bit silly. I didn’t want to look too silly.
Researcher: How did it compare to being filmed by us?
Ray: I dunno. I guess I’m not in control, so it’s easier. I can be myself, whereas I felt as if I was almost acting when I was doing it myself. I dunno. I just didn’t feel at ease, if you know what I mean.
Even though we use a much larger video camera and a sizeable shotgun microphone when we interview people in their homes, neither of these men felt the self-made videos were more real or genuine, but they were more invested in the image produced when filming themselves than they were when we filmed them. They, and other participants, also captured scenes and moments that we were not invited to witness first-hand, and that any outsider would be unlikely to see. The videos are simultaneously intimate and mundane: time in living rooms, bedrooms and cars (and bathrooms!); family dinners; small crises, and boring Sunday afternoons. The often shaky hand, poor lighting, and questionable audio of these videos only furthers their authenticity as glimpses into other people’s lives.
PARTICIPANT CONTROL AND ETHNOGRAPHIC CURATION
With the participants in control of the recording, they can carefully orchestrate how they want to represent themselves while simultaneously sharing moments that are much more intimate and mundane than our presence as researchers in their homes generally allows. Stylistically, participant-made videos fall along a continuum between two extremes. At one end are narrated videos in which the participant explains what he or she is doing, and at the other are ‘fly-on-the-wall’ videos in which the participant consciously acts as if he is not filming himself. In narrated videos both the viewers and the creators experience an immediacy and collapse of distance. The participant addresses the camera directly and acknowledges the viewer with statements like, “You’ll be interested in this,” or “I really want you to see how I do that.” In the case of Cédric, a French teenager, this means starting a video by repeating one of our questions to him. He says, “What a good TV moment means to me is,” and proceeds to act out, and narrate, all the components that go into a fun TV-watching experience – putting his feet on the table, drinking ice tea right out of the bottle, choosing mindless music videos his mother would despise. In fly-on-the-wall, ‘cinéma verité’ videos the viewer has a sense of the action unfolding in such a natural, unplanned way that it seems it would have happened whether the camera was on or not. For instance, a mother in Hong Kong cooks and serves dinner to her children and they eat while watching TV. In some cases the fly-on-the-wall feeling can seem voyeuristic and the viewer feels privy to a moment she wasn’t supposed to see, such as Mia Sudah giving her father a foot massage.
Four PMVs that have elicited strong viewer reactions illustrate both the diversity in participants’ self-representation styles, and the different types of work we, as ethnographers and curators of these intimate videos, take when presenting them to varied audiences. To make them more than YouTube style clips offering arbitrary, intimate glimpses into people’s daily lives, and into insightful research data for our stakeholders requires that we position and contextualize the videos in our broader knowledge of the research participants and of their cultural practices.
One of our most powerful, thought-provoking and frequently viewed videos is from the Sudah family in Japan. The image of Mia, the teenage daughter, giving her father a foot massage as he lies on a futon before going to sleep epitomizes a fly-on-the-wall video. It is late at night; the girl is wearing a robe and a towel around her wet hair. It is a privileged view of a warm family moment that shows us one of the myriad things people do while watching TV. When we show this video to stakeholders we explain the foot massage is another ordinary event in the course of the Sudahs’ TV-watching day. Its hold on viewers lies in the intimacy of the scene, the otherness of the activity, and the feeling that we are seeing a nightly routine that has never been shared with anyone before. In addition to shock, and occasionally offense, some viewers react with envy to this footage and express a desire for their own foot massage. While the foot massage draws in viewers, there is a larger point we are making when we show the video. Because the scene is arresting it is a useful example to start a conversation about the many postures people adopt while watching TV, and the diversity of activities people engage in while watching. It vividly communicates for stakeholders the tension between how people really watch TV, (at least at certain times of the day,) and the interaction intensive technology of Internet connected-TV.
If the foot massage comes across almost as voyeuristic, the clip created by Cédric, the French teenager who narrates a very funny video about his ideal TV watching moment, is a good example of the other end of the spectrum. He is putting on a very entertaining show that includes a mini-advertisement for ice tea (he apologizes for “le pub”) and clever production credits at the end. The clip grabs the attention of stakeholders because it’s witty, smart and well shot. Viewers laugh at Cédric’s jokes, and his description of his ‘magnificent couch,’ as well as nod in recognition that his actions are similar to how they watch TV. We explain to them that it was filmed solo by Cédric, and that the clip shows a side of him we researchers never got to see when we were with him and his parents. Making the video was a creative act of joyful teenage rebellion that wouldn’t exist if we (or his parents) had been there while he filmed it.
Another video that elicits a strong, but very different type of response shows Brigitte, a Parisian grandmother, who has just returned late from babysitting and is frantically trying to watch her daily soap opera on a malfunctioning TV. She pushes buttons and whacks at the set before scurrying into her bedroom to watch the show on a smaller TV that is not connected to her faulty IPTV service. The clip provokes laughter, piques interest, and reminds stakeholders of similar frustrations in their own lives. We tell them it is a moment we wouldn’t have been there to witness, and Brigitte’s great fluster and panic are emotional states we very rarely see in our role as researchers. It’s a fly-on-the-wall moment that underscores how flawed, buggy technology results in real emotional distress, and stakeholders are captivated by her aggravation. We use the clip to start a conversation about what’s at stake for consumers when their favorite technologies don’t work.
Somewhere in the middle of this continuum is a style of participant-generated video that feels fly-on-the-wall, but is carefully narrated and explained. American college student Sherry apologizes for “doing nothing” on a Sunday afternoon. Three housemates sit in the living room relaxing in front of the TV. Sherry is searching the Internet, updating Facebook, playing a computer game, and doing homework. Her boyfriend is hanging-out, and her brother is “about to start” texting friends on his phone. A moment this mundane rarely occurs when researchers are in a home, and it is a powerful reminder to stakeholders that technology is seldom used in isolation. Stakeholders discover something that deeply affects the way they think about designing technology —“doing nothing” means multi-tasking with multiple screens, services and applications.
In an age when the ability to make and share videos of home life and reality TV makes docu-soaps like Keeping Up With The Kardashians, and Jon and Kate Plus 8 regular viewing, PMVs create an interesting tension in our practice as ethnographers. The styles our participants use in their PMVs make them look a lot like YouTube videos or snippets from reality TV shows – the types of content that are entertaining and can go viral on the Internet. While we would never publish these videos on the Internet, we do want the intimate moments, and the insights they convey, to go viral internally with stakeholders. We want stakeholders to talk about, remember and absorb our research so its lessons permeate their work, but we have an ethical obligation to our participants to be worthy of their trust in sharing the mundane moments of their daily lives that are not usually subject to public scrutiny, and to not use the video to make them ‘look silly’ as Ray feared.
What separates PMVs from YouTube is our skill as ethnographers. Participants do create more interesting videos than the ones we produce, and it’s not hard to find people to shoot footage of themselves, (e.g. most of the You Tube library.) We have experimented with sending video cameras to research participants we have never met, and find that while we might get back video that is compelling and powerful, we are much more limited in our ability to interpret the footage and find the “so what?” kernel at its core. But, it takes a trained ethnographer who has spent time getting to know the subjects in situ to guide the video-making process, make sense of the resulting footage, glean insights that bring design or business-related value, and can frame and contextualize the video artifacts for diverse audiences and interests.
With PMVs, participants have taken an active role in the research, and when they are energized by the process, like Cédric, the Sudahs, Brigitte and Sherry, they produce video artifacts that go way beyond most ethnographer-made videos in their power to captivate, compel and entertain. As ethnographers we give up a lot of control when we hand cameras to research participants. Some of what we get back is unwatchable, unusable or painfully uninteresting. But, the good ones, the videos that transport us to places, activities, and moments-in-time that we would never have the opportunity to witness first-hand, are extraordinarily valuable to our ethnographic practice as a calling-card to explain what we do in vivid, visual terms, and as a powerful and direct way to communicate insights to stakeholders.
Susan Faulkner is a research scientist and video ethnographer in Intel Labs’ Interactions and Experience Research lab at Intel Corporation. She holds an MA in documentary film and video production from Stanford University and a BA in comparative literature from Brown University. She began working as a video ethnographer at Interval Research Corporation in 1993.
Alexandra Zafiroglu is a senior researcher in Intel Labs’ Interactions and Experience Research lab at Intel Corporation. She earned her Ph.D. in Cultural Anthropology from Brown University, where she was a Fulbright Scholar.
2008 Beyond Walking with Video: co-creating representation, Ethnographic Praxis in Industry Conference Proceedings, American Anthropological Association, p. 104-115
Brun-Cottan, Françoise and Patricia Wall
1995 Using video to re-present the user Communications of the ACM 38(5): 61-71.
1997 Excitable Speech: A Politics of the Performative London: Routledge
1990 Gender Trouble: Feminism and the Subversion of Identity London: Routledge
Collier, John A. and Malcolm Collier
1967 Visual Anthropology: Photography as a Research Method Albuquerque: University of New Mexico Press
1960 Primary Robert Drew, director Drew Associates
2009 OBAltimore blog (http://obaltimore.wordpress.com/) accessed July 20 2010
Flaherty, Robert J., dir
1922 Nanook of the North
1996 Taking Aim: the video technology of cultural resistance in Resolutions: Contemporary Video Practices edited Michael Renov and Erika Suderburg Minneapolis: University of Minnesota Press, p. 258-282
1992 The Real World
2007 Doing Visual Anthropology: images, media and representation in research London: Sage
Raijmakers, B., Gaver, W. W., and Bishay, J.
2006 Design documentaries: inspiring design research through documentary film. Proceedings of the 6th Conference on Designing interactive Systems ACM, New York, NY, 229-238.
Raymond, Alan and Susan Raymond, Dir.
1973 An American Family PBS
Reponen, E., Lehikoinen, J., and Impiö, J.
2007 Mobile phone video camera in social context Proceedings of the HCI International, vol. 2, Springer, Berlin.
2009 How Tiny Camcorders are Changing Education in eLearning Magazine October 8 2009 http://www.elearnmag.org/subpage.cfm?searchterm=Flip+video&article=98-1§ion=articles) accessed 8 August 2010
Worth, S., and J. Adair
1970 Navajo Filmmakers. American Anthropologist72: 9-34.
Worth, S., and J. Adair
1972 Through Navajo Eyes Bloomington, IN: Indiana University Press.
Ylirisku, Salu and Jacob Buur
2007 Designing with Video: Focusing the User-Centered Design Process London: Springer