In my (Sonja) last entry, I mentioned a recent project in one of the large gold mining concerns in South Africa. In this entry, I’d like share some of the experiences from this project.
I have been part of multiple pre-hypothesis narrative enquiry projects over the last few years; it’s probably the Cognitive Edge process I am most familiar with. However this project, a Culture Audit in the supply chain, provided a number of new and interesting challenges, the most notable being language and literacy. Because this process depends heavily on the ability of the participants to capture and cluster items, solving these problems was critical to the success of the project.
One of the problems was that 90% of the participants do not have English as a first language, many of them weren’t proficient in English at all, and could barely understand our instructions, never mind relate a story in English. The fact that South Africa has 11 official languages, added another layer of complexity, as at any given time we could be dealing with multiple local languages in a single group. There was also some sensitivity around the participants translating for each other which necessitated the use of external translators. Luckily we had one facilitator who was able to speak one or two of the languages and could double up as a translator, but in some of the sessions we had to use translators provided by the mine, who had no understanding of the process. This proved extremely difficult, as we had to continually stress with them to translate word for word as far as possible and not give their own interpretation of what was being said. A key learning that we took from this process was to source people with the required language proficiency and train them to be facilitators.
Different levels of literacy also proved a major problem, as many of the participants were unable to read or write, and therefore couldn’t cluster without assistance. As we only had a limited time each day to finish all the workshop tasks, we had to find a better and faster way to facilitate these tasks. Our answer came in the form of Rob Hooper our favourite cartoonist. Our workaround essentially worked as follows: Instead of capturing the characters in written form on hexies, Rob sat in the anecdote circles and drew rough sketches of each of the stories. We had a rule that he would always draw the main character on the right side of the page and the secondary characters on the left. We usually got a minimum of 13 of these stories out of a circle of approximately an hour. We then put these drawings on a wall and asked the participants to work in pairs (where at least one was literate) and assign a range of positive and negative attributes to the drawn characters. These were then clustered as a final step, with the more literate participants taking a lead role in the clustering.
We typically did not get as many archetypes out of these sessions as when we followed the standard process, but there was no difference in the richness of the outputs. We also found that it was easier for the group to brief Rob in drawing the cartoon representing the archetype i.e. describing what the archetype would look like, than to think of a name. Once the picture was complete, they could look at it and come up with a name.
We have learnt a lot during the course of this project. I must say it has only strengthened my conviction that these methods work and that they are appropriate for any level of employee. None of the other survey based tools can be used by illiterate people, and I’d question the validity of the outcome when completed by people with a limited grasp of the English language. We put a range of employees (from Senior Managers to illiterate store Workers) through essentially the same process (with minor deviations a as described above) and got very interesting and comparable outputs from all of them, making the culture audit results much more usable than others I’ve seen!