Collecting stories in a poor urban community
by Jonathan Carter, janniecarter at gmail.com
Editor's note: In this case study Jonathan refers to anecdote circles, which use methods covered in the deciding on methods and supporting storytelling sections of Working with Stories.
Framing the project
What was the impetus for your project? What led to it taking place? Why were you doing it?
Policy documents by the South African government express an interest in social networks and suggest that interventions should be improve the utility of these networks. These documents define terms like social networks, social capital and social cohesion hazily and tend to merge them. I had a particular interest in social networks and set out to understand the dynamics of social networks and groups in poor communities that could be tapped into by government to assist in government service delivery.
What were the project's goals?
My 'research objective' was to assess how existing government approaches to service delivery 'fit' with social networks. I was searching for a set of unknown unknowns. I didn't like the social network analysis approach to networks and when I started the research decided to use techniques like event maps, open discussions and repeated sessions with groups. I hoped these repeated visits would establish trust and insight into the dynamics of the community.
My research started in a predominantly Black African community with a scoping exercise to test the feasibility of the techniques I wanted to use and by what we saw, it looked like we would get what we wanted. Soon after the research started, the colleague I was working with decided to take a job elsewhere. This was a major problem as she spoke the local language, which I couldn't. As a result, I asked someone else to assist me and I gave her the opportunity to decide which community to work with. As I was subject to government funding cycles, I was under time pressure to get the work done quickly and therefore did not have the luxury to scan the new community. What we didn't know was that the new community held a set of worst case examples of managing and working with social networks. The history of the community is plagued with stories of corruption, being ignored by government, state land being handed over to an un-democratic entity to run, poverty and the transient nature of life that comes with poverty in many parts of the country.
When we first arrived, we heard one set of stories through the first few interviews we held. We then broke into another circle of people and we could not believe the stories we were told. The most interesting aspect of this was that I still feel no person lied to me. Some people failed to mention some facts that they could have, but I have no reason to think that anyone lied. After about two months of getting to know the community I was still not sure what 'the vibe' was. Too many gaps were missing.
At that time, I was fortunate enough to attend the Cognitive Edge course and learn about anecdote circles and then planned to hold some anecdote circles with what I determined were a representative sample of the community.
So in a nutshell, I ran the anecdote circles because after two months of conversations and discussions with a wide range of people I still could not validate anything I had found to that point. I saw anecdote circles as the way to really get into the head of the community. My plan was to gain a deeper understanding of the community.
The story of the project
How did the project get started? What happened first?
Getting people into the anecdote circles was really easy as we had been visiting the community for a while. So a few key people knew us and were happy to be part of the process. I had given the community a feedback report based on the research up to the time of the anecdote circles, which also built our trust.
What sorts of stories did you collect? How were they collected? Who collected them?
I agreed with my research assistant that she would arrange the groups, the venue and the time groups would meet as part of her fee. We discussed how to arrange the groups and identified key people to ask to pull the groups together. These were as follows:
- A priest we knew and had interviewed arranged a meeting with the religious leaders in the community. We managed to get seven priests; unfortunately, no Islamic leaders from the community were able to attend.
- A lady who worked with people living with HIV/AIDS was asked to bring a group of people living with HIV/AIDS together. Four ladies attended this anecdote circle.
- A group of women we had met a few times were asked to participate in a group and four of them came.
- We asked an NGO in the community to arrange some youth and got a group of four men and two women between the ages of about 20 and 22, all unemployed. This was a cracking success.
- An elderly lady who is an active civic leader in the community was asked to arrange a group of African ladies. This was the first group, seven attended and it was a complete failure due to my inexperience.
- We asked a youth civic leader to arrange a group of youth that would be very mixed racially. We got a group of five youth together. He asked to arrange a second group, which flopped because we could not find enough people.
- We asked a principal of a pre-school to organize a group with other principals and five attended.
All circles were held in a church hall. All the groups were seven or less, but none were less than four. I think four, plus myself and the research assistant, is an ideal number. From the second anecdote circle on, I sat at the opposite side of the circle to my assistant.
All groups were preceded by an explanation of the ethical issues concerned and signing of consent forms. This was an absolute pain, but unavoidable. If handled properly, it can be used to create the right atmosphere, although that atmosphere can be achieved a lot quicker than the time it takes to explain these forms.
Once the forms were signed, I explained the three rules:
- no interrupting,
- if you disagree tell the experience how you remember it, and
- only share experiences.
I also showed them the voice recorder and warned, quite sternly, that everything I said would be recorded.
I had read over the guide to anecdote circles from Anecdote a few times and wanted to experiment and find the ultimate question. Experimenting was a bad idea as I found myself asking very long winded questions about past friends and happy and sad experiences that took long to ask and were not understood. I stopped experimenting once we found that keeping it as simple as possible was the most effective route and started them by asking: Let's say I am looking for a place to live and I am considering this place, what experiences about living here would you share with me to help me decide to live here or not?
I think this question can be asked in any community. It is ambiguous enough, but simple enough to be understood. Ambiguity is critical to ensure that people do not game and are forced into raising issues that they decide are important to them.
Occasionally we had to remind the participants that opinions were not of value and that we just wanted to hear experiences, but most of my time was spent as an engaged listener.
In some groups, they would decide to go around the group giving each person a chance to answer the question. Once the first round was finished, a pause of silence would follow before the random outpouring of experiences began. It was best to let the group decide how best to start and simply not force anything. Out of all the circles, we had to probe two or three people to get them to talk, otherwise most people contributed without any encouragement. Generally, the people who needed probing would open up after the probing. The probing involved nothing more than just asking: what about you, do you have any experiences?
There was only one group that did not lead to the sharing of many experiences, which was the first. I had planned to take a research assistant that could be a translator but she couldn't make it. We had also not prepared the venue before picking up the group, which didn't set the right tone. In the rush, I didn't think about how best to arrange the group and as result we had seven women staring at the two of us. One of the people in the group acted as a translator, which did not work. I think it created a power imbalance that we could not manage or influence. It was an awkward experience from beginning to end and being the first one didn't help my confidence much.
On average, the introductory part lasted thirty minutes and the main discussions lasted an hour. Most of them fizzled out fairly rapidly and it was usually quite clear when to end the session.
How did the project end? Were conclusions drawn, and by whom?
I have written up the findings elsewhere in terms of data that emerged. [Editor: If you are interested in those findings please contact Jonathan.]
Evaluation
What turned out the same as you expected? What was worse than expected? What was better?
Part of the deep insight that was shared was stories about important events in the history of the community. These were especially important to me as I was aware of some events, e.g. a housing project, in the past that had created tensions, but was told different and contradictory versions of the events. During the anecdote circles, the detail shared was incredibly helpful in clearing up confusion. I did not get the full history of the community, but did get some of the events that are marked in individual and collective memories as turning points in the community. Knowing these turning points provides a base from which one can explore further.
The extremely rich detail in the stories and experiences shared gave a sense of where, when, why, what and how. Myths and stories of how myths were formed also emerged, adding to these rich insights. For instance, a number of people commented about people that spread rumours about HIV/AIDS in drunken conversations in a bar. From these simply told stories it became clear that there is stigma towards HIV/AIDS in the community and that the bars are one place where consciousness about HIV can be dealt with. This is one example of many insights I never asked for or planned to ask about, but I got it.
Can you share one conclusion of your project that you don't think you could have arrived at in any other way than by asking for and looking at stories?
Kurtz and Snowden (2007) argue that the naturalizing sense-making approach collects a sufficiency of information, rather than trying to collect all that can be known. When I finished the anecdote circles, I felt I had a sufficient sense of the ecology of the community to stimulate evolution (as Kurtz and Snowden suggest). A wide range of issues were brought forward, but I don't have enough to validate or verify any of them. As an example, I was told stories about children on drugs a sufficient number of times to know it was a problem that must be dealt with. In a general sense I understand the dynamics of the problem well enough to start breaking it (the key issue is being addressed in that community), but also understand exactly how complex an issue this is to address that a few constantly monitored and coordinated efforts are required. My understanding of these coordinated interventions, and associated risks, was gained only from what I heard during the anecdote circles. However, I have no valid measure of the number of young children addicted to drugs; how many of them finance their habits by selling scrap at an illegal scrap yard, or; how many take drugs with their parents. But I know all of these are problems that require a coordinated effort to deal with.
Many of my colleagues love their quantitative methods and want to show how valid and reliable their findings are. I am not sure how to respond to validity. I heard the above issue enough times to know I collected a sufficiency of information to know that the above issue is real. I did it very quickly. I don't need time to conceptualize how to repeat this in another community, nor do I need many resources.
What sorts of reactions did you get to the project during and after it?
I have used the results in articles I have written and to inform other research I am doing. But I am very disappointed, and ashamed, that I cannot use the findings of the research to effect change in the community. The only consolations I have are the people who participated enjoyed the discussions and were given quite personal advice from an experienced social worker. I also gave the community the report I mentioned above. The major recommendations I made in that report have not changed. I have sought advice from a range of role players, one of whom have extensive experience with communities similar to this community and have personal experience with individuals from the community. He advised me sternly to not publicize my recommendations unless I am able to spend the time that will be necessary to defuse tensions as a result of my findings. I will not be able to manage these and keeping away from the community in hand happens to be the most responsible way forward.
I was not asked by the community to do the research and was asked during the research to be responsible. I have taken the 'cop out' way, which happens to be the most responsible way out too.
High and low points
Do you remember any pleasant surprises during the project?
I found the participants to be incredibly open about their experiences. We thought that the participants probably never get opportunities to speak openly like we gave them and reveled in it. Besides the trauma that is shared, it also means that you are able to gain a deep and rich insight into the psyche of the community in an incredibly short period of time. The short time spent creating the right atmosphere is all that is needed to create the trust and start peeling open a deep can of worms. I learnt more about the community from completely ordinary people in the first three anecdote circles (i.e. 2 days) than I had in the previous two months of direct questioning.
How about unpleasant surprises?
I found more than I wanted to know. The experiences shared to me during the anecdote circles depressed me at the time I hosted them and they depressed me again when I read the transcripts. The community was an extremely poor and psychologically damaged community so the issues were extreme, but I am sure similar stuff will emerge in similarly poor communities elsewhere in the world, including Northern hemisphere countries. I think above all, preparing yourself to hear traumatic experiences is critical and do not try ignore the effects they have on you. I have read similar comments by Dave Snowden.
My assistant is a mature and experienced social worker and I knew she would be able to counsel anyone were a very traumatic experience shared. Researchers wanting to use the same technique, especially in poor urban communities anywhere in the world, should have a clear plan for counseling someone after sharing an experience of rape or sexual abuse. I hate to put it that way, but want to get the message across.
Do you recall any "aha" moments when you realized or learned something critical?
There were so many surprises and aha moments during the research as a result of dots being joined and complete hair raising issues being discussed (e.g. a priest visited a family to find a two year old smoking a joint with his parents) that it is difficult to pin-point one worth mentioning. However, I was very surprised by (a) how easy it was to get the deep insight into the way people behave and think in the community and (b) that the simpler the question I asked, the deeper were the insights and the richer were the stories. We all have stories and you don't have to read and write to be able to tell them!
Advice
What do you wish you had known before your project that you know now?
I think most people like to talk and discuss their experiences. I have only ever participated in one focus group. I was badly placed and felt that throughout the process I was being judged. In anecdote circles, participants are not being judged about their perceptions. They are being asked to share their experiences and the dynamic is completely different and this creates openness and plenty of sharing. Be prepared for the volume that comes out.
I shared the details of the failed circle because I think other researchers who want to use this technique should plan for failed anecdote circles. Occasionally they are going to happen and you must accept it. In a focus group, you can ask direct questions. This does not work in an anecdote circle and a lot of what determines success is out of your hands and you should not be put off by the odd failure. You are working with the natural side of humans; failure is a natural part of success.
What do you think you'll do the same and differently when you do your next project?
I think that were I to approach this research from the start now, I would plan that my first visit to the community is a Monday or Tuesday morning in the middle of the month. This timing is important so that the effects of end-of-month and weekend socializing and bingeing are avoided. I would arrive with a research assistant (of appropriate ethnic origin), a cartoonist, a voice recorder and some refreshments and find a private place, like a church that I could use and then ask a few people to join in an anecdote circle in return for some refreshments. I would literally ask people on the side of the road to join and would host two or three anecdote circles on that first day. Each anecdote circle would have no more than seven people. I would start by explaining to them that I have come to do some research in the community to understand the community's strengths and ask them to share experiences with me about living in this community to help me decide whether to live here or not. Every word said will be recorded and I would let the cartoonist run wild as he draws up cartoons of the various characters in the anecdotes.
At the end of this first day, I would get the recordings transcribed and the cartoons drawn up neatly and then decide on my next steps. I feel confident enough in this technique that I could follow the above steps tomorrow in any community in South Africa where levels of unemployment are high. I cannot think of any research interventions that would benefit from starting this way. However, had I started like this in the community I worked in I would have probably run like mad and never set foot in it again.
An example where I could have applied the above approach is to understand aspects of xenophobia. This could be achieved by hosting anecdote circles with each of the ethnic groups living in the area. Start each group with the question I mentioned above and once it is going, ask them what experiences they would share if you were of other ethnic origins in the community (e.g.: if I was Somali, If I was Zulu). However, if xenophobia is a problem, it will come out naturally in a group that is targeted by xenophobic hatred without having to fish for it as long as you have not mixed ethnic groups in the same group. The above tips are provided for extreme cases: if something is an issue, it is likely to emerge without prompting.
What advice would you give to a person who wants to do a similar project?
I have touched on some process issues that others should be aware of, but based on what I experienced, I think the following are critical:
- Keep the starting question (and any others) short and simple.
- Although not always possible, two facilitators create an atmosphere that participants are part of a discussion rather than talking to a facilitator. When using two facilitators they must sit at opposite sides of the circle.
- I feel four in a group is a minimum, even if there is only one facilitator, and this is a very good number. My guess is that eight participants is the maximum feasible limit.
- If you use a voice recorder, advise the participants that it will capture everything said. Dave Snowden comments that if you force people to tell the truth, they lie; if you let them lie, they tell the truth. I advised participants to hold back and they did the opposite.
Anecdote circles are best used when trying to capture unknown unknowns. It is not a technique that can be used for all purposes, but it can complement other techniques, especially quantitative techniques, very well. It is possible to find out some specific issues, but do not rely on the technique to achieve this. In their guide, Shaun Callahan and others from Anecdote use the analogy of a ship setting sail. I think this is appropriate as you can't force the wind to blow or even blow in a certain direction, but once you get going you can tack within the limits of the wind strength and direction. The same applies in an anecdote circle. Get the group going by asking a very simple and ambiguous question, then steer them by asking the starting question from a different angle. For instance, if you want to know HIV/AIDS related issues you could ask some way into the discussion what experiences they would share if you had HIV/AIDS. However, do this intelligently. For example, asking this question to a group of people whose HIV status you do not know may backfire.
Be conscious of the bonds of coherence. I hoped bringing all the priests together would increase the variety of issues discussed as well as reveal the full ambit of issues relevant to the churches. I had vague inclinations about a problem all the priests had with one of the priests in the community, but didn't fully understand it. Nor did I appreciate the lack of communication between the priests. We asked one priest to invite the rest so assumed if there were unworkable tensions the invitations would reflect this. The disliked priest came, but could only stay for the first few minutes and once he left, the mood changed and they started sharing experiences. But not all the priests knew each other and therefore the anecdote circle was hard work. It did not produce as much volume as I hoped. If I was more conscious about the lack of communication between the churches, I would have held separate anecdote circles with parishioners from each church i.e. individual anecdote circles with three churches would have produced far deeper insight than one anecdote circle represented by 8 churches together.
But we a very successful anecdote circle with some pre-school principals who were all from different organisations. The major difference between the pre-school principals and the churches is the pre-schools help and support each other, whereas the churches compete with each other. Remember those bonds of coherence!
References cited
Kurtz, C. and Snowden, D. (2007). Bramble bushes in a Thicket. In Gibbert, M and Duran, T (eds). Strategic Networks. Learning to Compete. Blackwell Publishing.
Snowden, D. (2005). From atomism to networks in social systems. The Learning Organization; 2005; 12, 6; ABI/INFORM Global pg. 552.