Research Design:
Exploring Space and Place With Walking Interviews
This article explores the use of walking interviews as a research method. In spite of a wave of interest in methods which take interviewing out of the “safe,” stationary environment, there has been limited work critically examining the techniques for undertaking such work. Curiously for a method which takes an explicitly spatial approach, few projects have attempted to rigorously connect what participants say with where they say it. The article reviews three case studies where the authors have used different techniques, including GPS, for locating the interview in space. The article concludes by arguing that researchers considering using walking interviews need to think carefully about what kinds of data they wish to generate when deciding which approach to adopt.
Keywords: walking interview; go-along; mobile method; GPS
Suggested Citation: Jones, P., Bunce, G., Evans, J., Gibbs, H., & Ricketts Hein, J. (2008). Exploring space and place with walking interviews. Journal of Research Practice, 4(2), Article D2. Retrieved [date of access], from http://jrp.icaap.org/index.php/jrp/article/view/150/161
The process of generating data through interviewing research subjects is unremarkable to most social scientists and yet when deciding to use this technique the researcher is presented with a whole series of questions. What form of interview: structured, semi-structured, or unstructured? Should an audio recording be made? How significant is the location in which the interview occurs? Should transcripts be produced? If so, should the interviewee’s words be tidied up (removing “ums” and “ers”), or should elaborate transcription systems be deployed to reflect intonation, hesitation, repetition, and so forth? What, if any system of coding should be deployed: strips of paper, cut-and-paste in a word processor, ATLAS.ti, NVivo? The list goes on.
All these questions and more have been pored over in the methodological literature. Those of us teaching in higher education generally say to the students that the best way to learn is to get out there and do it (and probably make a mess of it the first few times). Over the last few years a small but growing number of social scientists have been using walking interviews within research projects. Like students, we have been going out there and doing it (and occasionally making a mess of it), but the advantages and disadvantages of the different ways one can employ the “walk and talk” method have received less critical attention. Walking interviews are an ideal technique for exploring issues around people’s relationship with space, yet there has also been a somewhat curious lack of work attempting to more directly connect what people say with where they say it. This article therefore reviews some of the variations on walking interviews where making the connection between the words and the space is critical, focusing on three case study projects undertaken by the authors.
Mobility and methods used by researchers in motion have attracted a fair degree of academic attention within the social sciences in recent years. Sheller and Urry (2006) have talked about the formation of a new “mobilities paradigm” and there has been a proliferation of research using “mobile methods” (Ricketts Hein, Evans, & Jones, 2008). Whether considering movement by the participant or the researcher, mobility takes the research process out of fixed (safe, controlled) environments and introduces a range of new issues to consider. Of course the idea of studying life in motion is nothing new. Anthropological fieldwork and techniques such as participant observation often ask the researcher to study the subject in motion, rather than taking a participant out of their everyday context to ask them questions about their life. Movement in this context can occur at a series of spatial scales, from examining hunting rituals across a landscape, to microscale activities such as classic time-and-motion studies looking at the body movements of factory workers. In the latter case it is movement itself which is under scrutiny, although it is more common to use movement as a tool for exploring the activities those movements permit and the spaces in which these take place.
In any interview situation, the power relations between interviewer and interviewee can have a significant effect on the kinds of data that are generated. When interviewing socially marginalised individuals, for example, the fact that researchers are generally better educated and wealthier creates an uneven relationship bringing the risk that the interviewees might be too intimidated by the researcher to articulate their ideas for fear of “sounding stupid.” Movement brings an additional dimension to these messy issues relating to power. Questions around power/movement can in fact be present in stationary modes of research, but are not always conceptualised in such terms. Consider the issues that arise when thinking about the location in which an interview takes place. Even before one gets into the room, the whole act of travelling to an interview sets up these power issues. This might, for example, be the researcher transformed into a supplicant, heading off to a corporate office and fitting around a free time slot in the schedule of a high powered informant (Schoenberger, 1991). On the other hand, asking informants to come to a focus group held, say, on a university campus, changes the group dynamics before the research act begins. Participants have to go to, and pass through, a space which can be somewhat exclusionary and intimidating (particularly to those with no experience of higher education), regardless of the atmosphere in the meeting room where the data collection itself takes place. Furthermore, for cultural or safety reasons, researchers may not be comfortable interviewing in certain kinds of spaces, such as bars or people’s homes. These same kinds of issues are also present in more mobile forms of interviewing, with the researcher perhaps not wanting to walk with their participants in certain neighbourhoods or down secluded alleyways.
Safety is one of a number of issues highlighted by Carpiano (2009), using “go-alongs” to explore the relationship between place and health in two relatively deprived areas of Milwaukee, USA. The go-along is essentially a hybrid of interviewing and participant observation, with the researcher accompanying informants as they go about their daily routines and asking them questions along the way (Kusenbach, 2003). There are interesting parallels between the go-along and the technique of “shadowing” used in organisational research, where researchers follow their subjects around, sometimes for days at a time (McDonald, 2005). Carpiano notes some distinctive advantages of the go-along technique, particularly in terms of empowering and building rapport with participants. The technique also means that the environment acts as a prompt to discussions, without the researcher having to rely on photos and objects brought into the interview room (Anderson, 2004).
One of the key purposes of the go-along and walking interviews more generally is to examine a participant’s relationship with the environment, yet, spatial location is often dealt with rather crudely. In Anderson’s (2004) “bimbles” (aimless walks) with environmental protestors, the emphasis was on being away in the countryside, briefly escaping from the site of conflict. Arguably, specific location is less important in this kind of strategy--it does not really matter where the participant walks, just that they are no longer in the protest camp. Even here, however, some of Anderson’s interviewees talked about specific locations, that the developers would soon want “this bit” of meadow or forest (p. 259). Being able to place these discussions more accurately could have added another layer of interpretation to the stories recounted.
A number of projects making use of walking interviews have made little or no attempt to map the data (Clark & Emmel, 2008; Hall, Lashua, & Coffey, 2006; Ingold & Lee, 2008; Kusenbach, 2003; Moles, 2008). In some cases, interview material has been spatially contextualised by inserting prompts into the audio recording, for instance by mentioning the name of the street being walked down (Carpiano, 2009). One possibility for providing spatial context is the use of video and there has been some interesting work undertaken by Sarah Pink (2007) examining the practice of walking and filming. Video can, however, be somewhat disruptive and, particularly when the interviewer is trying to film, walk, and talk at the same time, the output can sometimes be unwatchably disorientating.
Clearly the ability to link words and location unlocks the potential of walking interviews for tackling more explicitly spatial issues. The authors have been working on a number of projects exploring some of the technical challenges that this poses. This article now examines three case studies which highlight different approaches to providing the spatial context for interview data gathered while walking.
This project aimed to critically analyse a planning discourse of promoting European style public spaces and “café culture” within Birmingham, a large post-industrial city in the English midlands. Exchange students from continental Europe were asked to walk around the city centre and discuss questions of urban design in recently regenerated areas. Where Kusenbach’s “go-along” empowers the participant to choose the route walked to reflect their everyday behaviours in space, the focus on particular locations in this project drove a decision to require all interviewees to follow a standard route determined by the researcher. A key advantage of doing this was in generating a cross section of responses to the same spaces and, indeed, taking some respondents through areas they had not previously visited to record their first impressions. The route was also designed to provide locations for sheltering from adverse weather, while still being able to look directly at the environment being discussed.
Clearly adopting this fixed route approach makes it easy to determine from the audio recording the general location of participants at any given point in the transcript. A nice example of this was in one participant, Nicholas, commenting, “I have never seen at the time of day Lyon so empty next to a beautiful place . . .” Taken away from its spatial context this kind of quote lacks significance, but knowing that Nicholas was standing in Centenary Square at this point gives an interesting cross-cultural insight into contrasting uses of public open space in Lyon and Birmingham. Similarly while walking through part of Birmingham’s International Convention Centre (ICC), Eva recalled the first time she went there:
When we were entering it we had no idea if we can enter it; we thought it was for the complex . . . and then it was a kind of a hotel in between and then it was the Symphony Hall or something; it was really weird.
The route through the ICC, which contains Symphony Hall and other facilities, is a classic piece of enclosed semi-public space, which forms a pedestrian link to developments around the canals on the western side of the city core. One can hypothesise that this type of privatised public space, more common in North America, would be less disorientating to individuals visiting from the United States, while it remains somewhat unfamiliar to continental Europeans. In this second example, the spatial clue is in the transcript--the mention of Symphony Hall--but this is simply a chance mention by the interviewee. Without this descriptive cue, it would not be clear that the reason why the participant struggled to understand what constraints there were on the use of this space was the fact that it has semi-public qualities.
In the previous example, the distinctive nature of the sites being discussed and the fixed route taken made it easy to work out from the transcript what was being said where. In the next project, the study site was a large single redevelopment built in the last decade on former dockland at the edge of the central business district of Bristol, in south west England. Unlike the highly distinctive sites explored in Birmingham, it would be more difficult to work out from a transcript alone which parts of the site were being discussed at any given time. This difficulty was compounded by the researcher’s decision to allow participants to choose their own route, restricted only by the general boundaries of the redeveloped docks.
In a project exploring urban design and distinctive types of buildings and spaces, it was crucial to know precisely which areas were being discussed by the interviewees. The decision was taken to use the Global Positioning System (GPS) to record the routes walked by participants. GPS uses a network of satellites to allow users to calculate their position on the earth, with an accuracy of around 5-10 metres. Any uncertainty about the location of a quote from the transcript could be resolved by matching up the time count on the audio recording with that of the GPS log of the user’s position. These logs also allowed aggregate maps to be created in a GIS mapping program, showing the routes walked by the participants, which helped to indicate the areas more commonly chosen to walk around. This in turn gave some insights into which spaces are relatively easy for users to understand whether they can pass through them or not and how they are supposed to be used (qualities sometimes described as the “legibility” of a space, see Lynch, 1960).
Creating vistas is an integral part of contemporary urban design. The ability to link locations to particular comments becomes a tool for exploring the way people respond to the views unfolding as they pass through spaces. An example of this came as Phil was walking next to the Floating Harbour and commented, “What a splendid sight in front of us. There are masts with flags, there is a sense of celebration there . . .” Knowing the location where the comment was made is critical to understanding the participant’s response to the theatrical side of urban design. Along similar lines, Geoff’s comment, “These buildings look like they have been designed with a ruler, there is no sympathy to the human experience. It’s very crisp; it’s a bit jarring . . .” becomes more meaningful when the specific location can be worked out from the GPS tracklog at a later date. In this case, Geoff was talking about the buildings on Cathedral Walk, making a general point about the extent to which specific pieces of architecture impact upon the feel and usability of spaces within the development.
The final example comes from another project based in Birmingham, this time trying to capture local understandings and memories of the Digbeth and Deritend districts of the city. Large parts of these areas are slated for a major regeneration project branded “Eastside,” which will significantly transform the physical environment. The research project sought to capture some of the social memories that animate these spaces before they disappear. Walking interviews were an obvious methodological strategy for undertaking this work, although again it was crucial to know what spaces were stimulating which stories.
GPS was again used to record the routes walked, although in this case using a more sophisticated approach. The audio recording was transcribed in 10-second pieces, the same interval as the GPS record, and then the two combined within a geographic information system (GIS). The result was that each point on the map of the route contained 10 seconds’ worth of text, creating a kind of spatialised transcript--examples of these can be found on the project Web site (www.rescuegeography.org.uk). Interviewees were asked to choose their own routes to generate stories about “their” spaces, although, as with the Bristol example, participants were asked to stay within some broadly defined boundaries.
One interviewee, Blair, stood outside the site of what had been his father’s factory in the area--now just another derelict building--and commented:
The business, Banks and Davies, was formed way back in the 1800s by Mr Henry Davies, and went through the war, the’39-’45 war, successfully because it made handles for pots and pans and buckets and things like that. And every time the British Army retreated, they took all their weapons with them, but they left all of those ancillaries behind [laughs] so they made a lot of money . . . during the war and promptly sold out afterwards.
These kinds of stories add a richness to spaces which, to the casual observer, have no meanings. It is often the spaces themselves that prompt these personal histories, meaning that it will be the stories as well as the spaces which are lost as regeneration strategies change these areas beyond all recognition. Matching a GPS record to a location can also give insights into what prompts interviewees to make particular comments, which are not necessarily attached to a specific place. Standing outside the new Bullring shopping centre, Clare commented:
It was quite a long bitter period, and people in Birmingham started feeling anti-Irish because of the pub bombings [in the 1970s] and the Irish feeling got at because they knew, or there was a growing awareness, the wrong people had been convicted; but if you said so you were accused of being an IRA supporter and so on.
This is not a story about the Bullring, but the new shopping centre is located very close to the site of one of the pub bombings and Clare was talking at a point where there is a vista across the city looking down towards the area which the council have labelled the Irish Quarter.
Clare subsequently went on to give her opinions about the rebirth of the St Patrick’s Day parade in the city, a tradition which had died out following the pub bombings. Thus being in a location can stimulate a socio-political narrative which is not related to the actual built form itself. With a GPS record, one can start to understand the effect of location on story telling. Indeed, by overlaying multiple spatialised transcripts within a GIS, one can start to identify locations which have a particularly potent effect in terms of stimulating recollections, associations, and opinions--something which may be useful in terms of planning future redevelopment in the area. Without the use of GPS it would become much harder to identify these kinds of spatial prompts recurring across a group of separate interviews.
The three projects explored above used walking interviews as a technique because they were particularly concerned with spatial experiences and place. Walks with fixed routes have the distinct advantage that even with a relatively small sample, one starts to get recurring data about particular locations. Fixed routes do, however, lose the empowering element that Carpiano notes with the participant-led “go-along.” The use of GPS gives the flexibility of a participant-determined route, but allows location to be tracked with a fair degree of accuracy. GPS is a problematic technology, however, and not just because the accuracy of location recorded can fall significantly in highly built up areas as the signals are scattered by tall structures. There is an uncomfortable “Big Brother is watching you” total surveillance quality to GPS as the researcher tracks the participant’s movements and this again raises questions about the power relations between interviewer and interviewee (Propen, 2006). Certain participants would doubtless be very uncomfortable having their words and location recorded through the use of technologies developed for military use--one wonders how many of Anderson’s (2004) environmental protestors would have consented to this. There are also significant practical problems. Put simply, combining GPS with interviewing is complex and somewhat difficult to master. One is required to set up a piece of technology before starting the interview, extract the data from it afterwards, and then make use of this in the data analysis, with the potential for this to go wrong at any stage. One has to think very carefully before undertaking the research about the degree to which it is important to have a fairly precise record of what was said where, or whether it would be sufficient to simply make a comment about location where this is significant, sacrificing some of the interview’s natural flow.
Various kinds of mobile conversation (“go-along,” “bimble,” and walking interview) are increasingly being deployed by social scientists as a research method. These different techniques have advantages and disadvantages, and before choosing to include these in a research project, it is worth considering just what kinds of data are needed. For many projects it may be that inserting verbal prompts into the transcript would be sufficient to note the occasional location where an interviewee’s comments make limited sense without the spatial context. Where the role of space itself is a key object of study, use of GPS or video might be worth considering, despite the technological and practical issues they raise. Alternatively, sticking to fixed routes and more prescriptive question sets can quickly generate information about key sites from a range of participants. Mobile interviewing is still somewhat at the experimental stage, though as the projects described above indicate, it has great potential to shed light on how participants use and understand different spaces. These techniques also provide a means to take the interviewing process out of the “safe” confines of the interview room and allow the environment and the act of walking itself to move the collection of interview data in productive and sometimes entirely unexpected directions.
The sections on Europeanness in Birmingham and Bristol Harbourside are based on dissertation research undertaken by Hannah and Griff respectively. The Rescue Geography project was funded by a grant from the Economic and Social Research Council, RES-000-22-2375.
Anderson, J. (2004). Talking whilst walking: A geographical archaeology of knowledge. Area, 36(3), 254-261.
Carpiano, R. M. (2009). Come take a walk with me: The “Go-Along” interview as a novel method for studying the implications of place for health and well-being. Health & Place, 15(1), 263-272.
Clark, A., & Emmel, N. (2008, March). Walking interviews: More than walking and talking. Paper presented at Peripatetic practices: A workshop on walking, London.
Hall, T., Lashua, B., & Coffey, A. (2006). Stories as sorties. Qualitative Researcher, 3(Summer), 2-4. Retrieved November 7, 2008, from http://www.cardiff.ac.uk/socsi/qualiti/QualitativeResearcher/QR_Issue3_06.pdf
Ingold, T., & Lee, J. (2008). Ways of walking: Ethnography and practice on foot. London: Ashgate.
Kusenbach, M. (2003). Street phenomenology: The go-along as ethnographic research tool. Ethnography, 4(3), 455-485.
Lynch, K. (1960). The image of the city. London: Technology Press.
McDonald, S. (2005). Studying actions in context: A qualitative shadowing method for organisational research. Qualitative Research, 5(4), 455-473.
Moles, K. (2008). A walk in thirdspace: Place, methods and walking. Sociological Research Online, 13(4).
Pink, S. (2007). Walking with video. Visual Studies, 22(3), 240-252.
Propen, A. (2006). Critical GPS: Towards a new politics of location. Acme, 4(1), 131-144.
Ricketts Hein, J., Evans, J., & Jones, P. (2008). Mobile methodologies: Theory, technology and practice. Geography Compass, 2(5), 1266-1285.
Schoenberger, E. (1991). The corporate interview as a research method in economic geography. Professional Geographer, 43(2), 180-189.
Sheller, M., & Urry, J. (2006). The new mobilities paradigm. Environment and Planning A, 38(2), 207-226.
Received 14 October 2008 | Accepted 7 November 2008 | Published 14 November 2008
Copyright © 2008 Journal of Research Practice and the authors