Location of collecting samples and data
The survey management team should determine where the interviews, the collection of biologic specimens and food samples and biological laboratory processing will take place. These may be at the same or different locations. With more complex surveys, especially those in settings where households are very far apart or in difficult terrain, it may be best to set up a mobile laboratory in a central location. Sampling could also be done in households or through a clinic or school.
The main advantages of setting up a field laboratory for collecting and processing data include:
- A minimum number of sets of equipment may be required per team, thus reducing costs.
- Data collection in a single place eliminates the time and effort needed to move between households carrying equipment and supplies, including anthropometry equipment, that would need to be set up in each household. It also reduces data collection time, is more convenient for the survey team members, prolongs equipment life and can control the conditions where precautions against contamination are required, such as with blood collection for zinc analysis.
However, it is not always possible to find a convenient central site or it might be inconvenient for participants to travel to the central location.
Conducting interviews at the household rather than in a central location can improve participant comfort and privacy. In addition, household-level data collection may lead to more complete information about programme processes, such as being able to access child health cards and observe the labeling on food products. A household location also helps avoid potential biases in the collection of food samples, for example, a respondent may buy or obtain some of the requested food on their way to the central site.
It may be useful to use both locations, for example asking individual questions and collecting food samples at the household, and collecting biological specimens and anthropometry data at a central location. No matter the setting, care must be taken to match the unique ID labels for questionnaires with associated samples and specimens so that they can be correctly linked during data management and analysis.
All field procedures must be pilot tested. The survey management team needs to be able to change plans based on the most efficient and effective method for high-quality data collection and high response rates. The selected method should be the same for each cluster.