A change of view: Using visual methods to explore experience in qualitative research
A change of view: Using visual methods to explore experience in qualitative research
By Lorna Rouse
Background – the use of visual based methods in the social sciences
Whether it’s as a tool of assessment, a stimulus to trigger responses or a means of displaying cognitive models and presenting results, images have many uses in psychology (Bagnoli, 2009; Frith, Riley, Archer & Gleeson, 2005). Traditionally, visual methods have involved coding or counting responses or behaviours. More recently, interest has grown in performing qualitative research which focuses on the visual images themselves in order to explore participants’ experiences and meaning making (Frith et al., 2005).
The use of images is not a new methodology but is rooted in disciplines such as anthropology and sociology (Harper, 2002). Collier (1957, as cited in Harper, 2002) highlighted the value of photographs as a visual tool in social research after using a photographic survey to help researchers to categorise the quality of housing in an examination of the environmental basis of stress (Harper, 2002). Later, whilst exploring how families adapted to living in new environments, the research team used photographs of old and new surroundings to assist their interviews.
They found that the use of photos improved the quality of the interviews in many ways. It prompted memory, reduced misunderstandings and elicited higher quality and more comprehensive interviews (Harper, 2002). Collier (1957, as cited in, Harper, 2002) named this method ‘photo elicitation’ and it has since been used to research many topics in the social sciences, including experience of illness and homelessness (Radley & Taylor, 2003; Radley, Hodgetts & Cullen, 2005).
A valid approach?
Described by Harper (2002) as ‘a waif on the margins’ in disciplines such as psychology, photo elicitation and other visual methods have encountered some scepticism amongst social scientists and have only recently entered the mainstream (Frith et al, 2005). Such wariness has been attributed to doubt over the validity of images, which are ambiguous and open to multiple and subjective interpretations. This did not fit with psychology’s desire for objective data representing an accessible reality (Frith et al., 2005; Guillemin, 2004). Following the ‘crisis of representation’ in social science there was a shift in thinking towards exploring meaning making and the construction of multiple realities. From this perspective the ambiguous nature of visual data is not a difficulty because, like all data, it should not be interpreted as reflecting a stable reality that the researcher can access. Instead it is a way of constructing multiple realities influenced by social and cultural factors and situated in a particular time and space (Frith et al., 2005; Guillemin, 2004). This has led to an increasing acceptance that visual methods can provide valuable and validdata about issues of concern to the social sciences and there have been calls for further application of these methods to psychological questions (Frith et al., 2005).
What can we learn from visual data?
One advantage of visual methods is that not all participants are able to express themselves verbally, for example, young children. Similarly, not all experience is best expressed through words (e.g. pain) and some people have a preference for visual expression (Guillemin, 2004). Indeed, Bagnoli (2009) describes insight conveyed by images with a condensing quality, such as participants depicting themselves at a crossroads with different potential futures stretching ahead.
Changing the interview process
Furthermore, the discussion or production of images can aid the interview process by breaking the ice, prompting memory, improving the flow and content of the interview and helping establish rapport and shared understanding (Harper, 2002; Bagnoli, 2009).
The use of images can place control of the interview process with the participant, bringing out issues that are meaningful to them (Frith et al., 2005). This also elicits details that might otherwise be difficult to talk about leading to the disclosure of more sensitive issues and emotional details (Bagnoli, 2009). Oliffe and& Bottorff (2007) highlighted this in a study using photo elicitation to study men’s experiences of prostate cancer. The authors felt that describing photos put participants in the role of expert, helping them to relate difficult details at their own pace.
A fresh perspective
The process of producing a visual image allows participants time to reflect on the topic being explored, which may not only produce rich and insightful images but inform a more detailed interview. Use of a novel medium also provides participants with the opportunity to reflect on experience in a different way. This has been described as ‘breaking the frame’ of experience (Harper, 2002). For example, in photo elicitation studies participants are engaging with their environment and experience, of homelessness or as a hospital patient in a way that would not usually occur (Radley & Taylor, 2003; Radley et al. 2005).
How do researchers go about collecting visual data?
There are many ways of collecting visual data and the choice of method depends on the aims and theoretical perspective of the researcher. This may include the following categories of visual data:
maps, diagrams and matrices (e.g. Copeland & Agosto, 2012)
photographs and video footage (e.g. Radley, Hodgetts & Cullen, 2005; Ross, Renold, Holland & Hillman, 2009)
collage and drawings (e.g. Bagnoli, 2004; Guillemin, 2004).
Visual methods can be used alone, in combination with verbal data or as one of a number of multisensory methods. For example, Bagnoli (2004, /2009) describes the use of multi-method biographies to holistically explore young people’s identities. Methods included verbal interviews, written diaries and visual methods such as self-portraits, video diaries, relational maps and time-lines. The aim is not only to use visual methods as a tool to assist with interviews as a kind of ‘add-on’ but as an important method of eliciting and understanding experience in its own right (Bagnoli, 2009).
Some researchers have warned against using visual methods in isolation because this removes the image from the context in which it was produced (Frith et al., 2005). This was illustrated by Radley and Taylor (2003) who asked inpatients to photograph their hospital experience. One participant’s photos of a bathroom only became meaningful in the context of her explanation. The bathroom looking inwards represented the traumatic experience of pain, struggling and waiting for help following surgery. The view looking outwards signified the end of that experience. Photographs are described as being ‘articulated into significance’ (Radley & Taylor, 2003).
The collection of visual data varies according to:
who produces them – the researcher, participant alone or with input from the researcher
whether they pre-exist the research, are created for the research, are enduring or temporary (Gibson & Riley, 2010)
instruction given to participants - structured (e.g. Copeland & Agosto, 2012) or open (e.g. Bagnoli, 2009)
when the image is produced - before, after, or during the interview
Whether the interview focuses on the topic of interest (e.g. Gullemin, 2004) or is centred on the image (e.g. Radley & Taylor, 2003).
These choices have different implications for data produced and analysis that can be conducted.
How are visual data analysed?
Approaches differ but analysis often involves identification of themes which construct experience. One framework recommends asking questions about the production of the image, the image itself and the relationship between the image and its audiences (Rose, 2001, as cited in Guillemin, 2004). Bagnoli (2009) uses a narrative analysis looking for the story told with multi-media coding, linking data collected through different mediums.
Examining the process through which an image was created is often viewed as a vital part of the analysis (e.g. Radley & Taylor, 2003 ). Attention is paid to what is made visible and what remains hidden. This is important to understanding the construction of a particular reality (Frith et al., 2005). For example, Radley & Taylor (2003) found the world of hospital inpatients was constricted by their mobility.
Other important considerations include: acknowledgement of the researcher’s part in the production process, identification of contradictions within the data and analysis of images within the context of data collected through other means (Bagnoli, 2009; Radley, 2010).
Challenges of visual methods
Researchers have reported that a minority of participants are uncomfortable with using visual methods and may occasionally refuse to engage with activities such as drawing (Bagnoli, 2009). Interestingly, Guillemin (2004) incorporated refusal to take part into the data analysis, speculating that those who felt unable to draw were still coming to terms with and making sense of on-going medical conditions.
Activities may impose a particular way of thinking about the world. Bagnoli (2009) gives the example of relational maps which assume participants see themselves at the centre of a relational world, allowing no opportunity for people to express an experience of aloneness. There may also be problems of inclusivity. Bagnoli (2009) found that people with intellectual disability were unable to take part in producing timelines due to difficulty with the concept of time.
Studies involving photographs and videos are often limited by organisational rules. This has occasionally proven unexpectedly useful, for example, although hospital authorities would not allow patients to take photographs alone (in case they included other patients) Radley &and Taylor (2003) found that being present whilst patients chose and set up
pictures led to field notes that were valuable during analysis. Other practical considerations include ensuring those involved know how to use technology such as cameras and the risk of losing data through accidents (Gibson & Riley, 2010).
The need for confidentiality and informed consent may mean that images have to be shared in an altered form e.g. in videos and photos faces may be blurred and voices disguised (Frith et al., 2005). Even then participants could be identified by those who know them e.g. by location or build (Gibson & Riley, 2010). If the participant who produced an image cannot be contacted to give consent it may not be possible to publish the image (Rouse, 2013). Other ethical issues include questions around copyright ownership and potential negative interpretations following publication that are beyond the control of both researcher and participant (Frith et al., 2005).
A multisensory approach
There are clearly many issues to consider when using visual methods, but they open up a fascinating opportunity to elicit and understand experience. Images are used for other purposes in qualitative research e.g. conversation analysts examine interactions using video footage (Gibson & Riley, 2010). Although verbal methods have been privileged our experience of the world is multisensory and researchers are acknowledging this through the use of increasingly multisensory approaches (e.g. Taylor & Coffey, 2009), contributing to a rich understanding of people and the social world from multiple perspectives.
Originally published in JEPS Bulletin, Journal of European Psychology Students. http://jeps.efpsa.org/blog/2013/05/15/a-change-of-view-using-visual-methods-to-explore-experience-in-qualitative-research/
Bagnoli, A. (2004). Researching identities with multi-method autobiographies. Sociological Research Online, 9(2). Retrieved from: http://www.socresonline.org.uk/9/2/bagnoli.html.
Bagnoli, A. (2009). Beyond the standard interview: The use of graphic elicitation and arts-based methods. Qualitative Research, 9(5), 547–570. doi: 10.1177/1468794109343625.
Copeland, A. J. & Agosto, D. E. (2012). Diagrams and relational maps: The use of graphic elicitation techniques with interviewing for data collection, analysis, and display. International Journal of Qualitative Methods, 11(5), 513-533. Retrieved from: ejournals.library.ualberta.ca/index.php/IJQM/article/view/10259.
Frith, H., Riley, S., Archer, L. & Gleeson, K. (2005). Imag(in)ing visual methodologies [Editorial], Qualitative Research in
Psychology, 2(3), 187-198. Retrieved from: http://dx.doi.org/10.1191/1478088705qp037ed.
Gibson, S. & Riley, S. (2010). Approaches to data collection in qualitative research. In Forrester, M.A. (Ed.), Doing Qualitative Research in Psychology: A Practical Guide (pp. 59-76). London: Sage.
Guillemin, M. (2004). Understanding Illness: Using Drawings as a Research Method. Qual Health Res, 14(2), 272-289. doi: 10.1177/1049732303260445.
Harper, D. (2002). Talking about pictures: a case for photo elicitation. Visual Studies, 17(1), 13-26. doi: 10.1080/14725860220137345.
Oliffe, J. L. &. Bottorff, J. L. (2007). Further than the eye can see? Photo elicitation and research with men. Qual Health Res. 17(6), 850-858. doi: 10.1177/1049732306298756.
Radley, A. & Taylor, D. (2003). Remembering one's stay in hospital: A study in photography, recovery and forgetting. Health: An Interdisciplinary Journal for the Social Study of Health, Illness and Medicine, 7(2), 129-159. doi: 10.1177/1363459303007002872.
Radley, A., Hodgetts, D. & Cullen, A. (2005). Visualizing homelessness: A study in photography
and estrangement. Journal of Community & Applied Social Psychology, 15, 273–295. doi: 10.1002/casp.825.
Radley, A. (2010). What people do with pictures. Visual Studies, 25(3), 268 – 279. doi: 10.1080/1472586X.2010.523279.
Ross, N.J., Renold, E., Holland, S. & Hillman, A. (2009). Moving stories: using mobile methods to explore the everyday lives of young people in public care. Qualitative Research, 9(5) 605-623. doi: 10.1177/1468794109343629.
Rouse, L. (2013, June). Visual and arts-based methods in social sciences: an interview with Dr Anna Bagnoli. News & Views, The Open University Psychological Society.
Taylor, C. & Coffey, A. (Eds.). (2009). Qualitative Research and Methodological Innovation [Special issue]. Qualitative Research, 9(5).
This article originally appeared in News & Views March 2014.