Data Analysis

The participant’s VIA visual narrative tapes are collected and dubbed to make master tapes as well as VHS copies for the participant. In making these copies, each original tape is assigned a dedicated number, indicating the participant and tape number, which, along with video time code, generates a unique location code for each frame of VIA tape so researchers can locate and direct each other to specific audiovisual data for analysis.

Logging is the first step of VIA visual narrative analysis. Analogous to a transcription of an audio tape, a VIA log is a guide to the audiovisual data, text documentation of what is observable on each tape. It provides a simplified, indexable guide with which to find certain scenes, situations, or concepts without having to watch and listen to the entire set of tapes each time. As we develop better data management tools and techniques, the VIA analysis process is constantly evolving. Although pilot VIA projects were logged by hand onto logsheets, we now log directly on computers. Differentiated by distinct styles of text, the location code, who is controlling the camera, and descriptions of the objective and subjective content of each scene are recorded in the log.

The distinction between objective and subjective content is of the highest importance. Objective content is that which is concretely visible or audible, emotionally neutral, and reproducible by any observer. This would include direct transcriptions of dialogue and simple descriptions of scenes. However, one of the unique and powerful features of VIA is its ability to capture the subjective components of the participant’s life experience. The emotional tone of the scene and the participant’s perspective on it are important data that bear directly on the participant’s attitudes, beliefs, and resulting behaviors. Unlike the neutrality of objective observations, however, subjective observations may vary between observers. The complexity of human experience captured in VIA visual narratives is best transcribed into what anthropologists call “thick description,” a rich record of the subjective elements of what VIA has captured on tape.

Each logger works with the visual narrative of one participant and watches the tapes all the way through before starting to log. This “movie watching” gives the logger a chance to get acquainted with the participant and the people and places of the participant’s life. The second viewing of the tapes is a logging pass. Logging frequently takes much longer than simply watching the tapes because the logger often pauses and rewinds scenes in order to clarify observations by hearing or seeing the scene several times. All tapes are screened “real time,” regardless of how uninteresting the content may appear. Loggers are asked never to fast-forward scan the tapes because they may miss crucial audio or video data from on- or off-screen.

Loggers meet with the full research team on a regular basis to discuss what is being observed on the videos and to compare findings. Techniques derived from ethnography, participant observation, and visual anthropology are used to analyze the content of the visual narratives. Grounded theory is used to identify and develop emerging themes that unify the participants’ illness experiences. In discussions with the interdisciplinary research team, themes are compared and contrasted between participants’ visual narratives. Loggers are encouraged to re-view the video data to clarify or modify themes as the research team develops them. Literature searches are conducted as needed to generate a library of background material on developing themes and findings.

As of late 2005, we are exploring use of Transana, an open-source software that allows direct analysis of video. Once the logs have been completed, they are exported from Transana and imported into NVivo, a qualitative analysis software package that manages, analyzes, and displays the logged VIA data.

NVivo allows VIA logs to be analyzed with a maximum of flexibility while retaining the richness and depth of the original data. As logged data are coded, NVivo can link analyzed log data to the original video data and to any other documents that augment or clarify the data. This is particularly useful as we publish research papers in new formats and communications media.

Once the visual narrative data are coded and structured using NVivo, members of the multidisciplinary research team examine them through the theoretical frameworks of medicine, public health, psychology, clinical social work, and anthropology. Different disciplinary perspectives on the material allow us to triangulate on the data. The theoretical framework of each discipline illuminates and expands the analyses of the others, as we work to characterize and understand the patient's illness experience.

>> Learn how the visual narratives are created

Related Publications

view Patashnick, J. L. and M. Rich (2005). "Researching human experience: Video Intervention/Prevention Assessment (VIA)." Australasian Journal of Information Systems.
view Rich, M. and J. L. Patashnick (2002). " Narrative research with audiovisual data: Video Intervention/Prevention Assessment (VIA) and NVivo." International Journal of Social Research Methodology5(3): 245-261.
view Rich, M., S. Lamola, et al. (2000). "Asthma in life context: Video Intervention/Prevention Assessment (VIA)." Pediatrics105(3 Pt 1): 469 - 477.

Center on Media and Child HealthPrivacy Policy & Terms of UseChildren's Hospital Boston