Great relief all around. The surveying of 250+ households in the informal settlement of Mshenguville has been completed. Highlights included a puncture, having a kitchen knife waved at me and learning how to recognize “white man” in three different languages. The knife was very friendly fire. White men or women on foot are very rarely seen in the township. A concerned young mother saw me walking around on my own and left her food prep to come check I was not lost or demented or both. “Make sure they come talk to me” she said once I’d explained that I was helping the volunteers survey caregivers of children. I was invited into some of the shacks and was struck by how orderly they were arranged. In one there were salt and pepper grinders and a glossy magazine laid out on a small table next to the small paraffin stove. The magazine was opened at an article titled “How to remain sexually confident – no matter who you’re loving”. Another shack owner had arranged two armchairs outside the front door so as to best catch the light.
The picture shows some of the volunteers celebrating with an awkward umlungu after the last form had been collected. To the front is Thuli, Project Coordinator for Project Hope, and in the background is fellow GSK volunteer Donna who very kindly helped out throughout the two weeks.
Taking a car on the rough tracks that run through Mshenguville wasn’t that wise but the car became an essential tool in the survey. How could we eliminate the long evenings where we all headed back to the main office to make sure the days challenges were captured? The supervisors pointed out that standing around in the market area having a chat about shack 398 might be seen as a breach of confidentiality. Solution, I should sit in the back seat of the car and each team would enter in turn to relay their challenges. Now I am certain most of the town think I’m a spy.
The Project Hope staff and the two leaders of the volunteer group met with me the next week and complied some recommendations to help make any future surveying go better. I’ve posted them here, in case anyone is fortunate enough to have to take on a similar exercise.
A lot of background tasks need to be completed to carry out a survey, including:
- Printing and collating forms in advance of each days survey
Planning routes for the next day of each survey
Registering each team (pair of two interviewers) in and out each day
Briefing teams on the day’s route, meeting place, any matters leftover from the last day
Documenting challenges from the day for follow-up
Checking forms from each team for completeness and accuracy
Supporting teams when interviewing
The leaders can greatly assist in these tasks, but ideally they should be selected in advance so they are well prepared.
Candidates should be made aware and discuss the commitment required at the first training session. Ideally include volunteers from all areas to be surveyed. Decide ideal number of teams in advance, more teams in theory means more homes covered, but:
a. More work to effectively train, supervise and organise
b. Harder to be sure if each team really understands their tasks
c. Harder to ensure each team has their own area to survey on a given day
Should be done as close to the time of the surveying as possible. Attendance is compulsory for all agreeing to volunteer after a first session where: explain purpose of the survey, introduce/describe survey forms and survey methods, discuss the commitment required from interviewers and leaders, draw up contracts with those willing to assist (contract for interviewers/leaders) and let the volunteers discuss the contract with their families at home if need be (e.g. if days will be long)
Later training should review the content/purpose of the survey form, discuss challenges of interviewing people about their personal affairs, and discuss how to follow-up when people request a different date or when challenges are revealed when visiting homes that need referring to appropriate bodies. Practicing interviewing should be the main focus of the training. If the interviewers do not know each other that well or are not used to working together then a team building exercise is worth considering before going to the field.
Essential. Use it to test the form as well as the interviewers. The data collected is unlikely to be used (team still learning) but do take challenges seriously and report and act on them. Limit the number of interviews per team (as data unlikely to be used) but ensure each interview is as expected for full study. Discuss any problems as a group after interviewing. Data collected should be reviewed carefully by the leaders and the problems identified should be discussed with the teams before the full survey
Itinerary to target the time of day when the people of interest are likely to be available for interview but it also needs to consider the volunteers too (they have their own families). De-brief and collect challenges at the end of each day of surveying. Set aside one of the first couple of days to be office based so that all teams review their data with the leaders. This way errors in data collection can be sorted early on and not repeated. If some challenges can be predicted (e.g. where can I get help to get my child registered at school?) prepare a directory of actions/advice in advance so that interviewers can give the advice as they are interviewing. Emphasise that these challenges need to be captured (perhaps in a note book) along with any advice given.
Some people will ask to be interviewed on a different date or time. Design a form for each team so that they can keep a note of and complete re-visits
After the survey
Review what went well and what could be improved. Try to ensure data-entry is completed shortly after the survey. Make sure personal data is stored separately from other data. Keep and store forms securely. Invite the team to see the main results of the survey once compiled