Pilot testing

All survey procedures must be tested before survey implementation begins. Pilot testing at the end of the training helps to estimate the amount of time it takes to complete the survey in each cluster and to identify any potential remaining concerns with the survey questionnaire, forms and protocol.

A pilot test usually takes several days, however the time needed will vary depending on the complexity of the survey. Each field team should have the opportunity to conduct the pilot in a number of households from both an urban and a rural setting to provide a good range of experience. Urban and rural settings can be quite different and have different challenges.

The pilot test needs to be organized well in advance, and must not take place in enumerated areas selected for the main survey. Ideally, the location should be relatively close to the training venue to minimize travel and maximize implementation time.

The entire fieldwork process should be piloted, from organizing supplies and freezing gel packs to entering data and processing samples. After each day of the pilot there should be a debriefing session with the whole team to discuss their experiences and resolve any challenges faced. Trainers may need to provide some refresher training for any major problems identified. If the first day of the pilot raises major issues that require time to rectify, for example significant changes to the survey questionnaire or the specimen collection process, then the subsequent days of the pilot may need to be delayed. For electronic data collection, programmers should be involved and available to troubleshoot in as close to real time as possible.

Logistics

The sites for the pilot test should be identified one to two months in advance, and permission should be requested from the local administrative body and community leaders. Transport from the training venue to the pilot sites needs to be organized. The number of pilot sites required will depend on the number of field teams participating. To simplify logistics, several teams may work within one practice cluster if the cluster is large enough.

The practice cluster needs to be mapped. This may be done as practice during the training sessions for the mapping and household listing team, or it may form part of the pilot test if that is part of the field team’s role. Once mapped, team leaders should work with the Survey coordinator to decide which households in a particular cluster will be allocated to each team, according to the survey protocol.

Final field preparation

After the pilot test has been completed and systematic challenges have been identified and corrected, the teams are ready to start fieldwork. A short time period should be allotted between the end of the pilot test and the start of fieldwork. This will allow for any additional training or pilot testing, finalization of the survey questionnaire (paper-based or electronic) if needed, checking and cleaning electronic devices where used, preparation and inventory of all field supplies, and travel of field teams to their initial clusters.

Before the fieldwork begins, the final field teams should be selected and allocated to the various clusters. All trainees, even those not selected, may be provided with a certificate for completing the training course. A Generic Certificate of Participation is available in the online tool.

When possible, the initial fieldwork should begin in clusters nearest the training venue. This ensures that all the teams are working as near to each other and to the central coordinator as possible, and allows for close supervision and easier communication during the initial stages when concerns are more likely to arise.

Refresher training

Sometimes surveys take several months to complete, and it may be necessary to schedule a refresher training part way through the field work. This is particularly relevant if issues arise with data collection or if morale wanes during the field work. Sometimes refresher training may be scheduled around a break, such as a religious holiday, where field work is temporarily suspended. It can provide a great opportunity to gather everyone together for motivation and for discussion and troubleshooting any specific difficulties being faced in the field. In some countries, especially large countries, it may make sense to have several regional refresher training courses to save on travel time and cost.

Refresher training should be brief, one to two days at most, so that teams are able to get back to the field quickly and resume data collection.