Survey Data Collection and Processing - Sampling Spatial Units for Agricultural Surveys - page 220

9.3 Data Collection, Instruction Manual, Training
of Enumerators, and Field Work Management
In the data collection phase, the units selected for the survey are contacted if they
are respondents, or located if they are spatial units. The contact methods depend on
the type of statistical unit and on the data collection instrument. The chosen
methods have implications on the cost and quality of the data.
Regardless of the techniques used, data collection has three main objectives:
1. To identify the statistical unit (farm, polygon, or point), and to enable its
participation in the survey.
2. To collect the information without distorting or influencing the response or
measurement process.
3. To consider all stages of the survey and possibly its longitudinal structure to
facilitate any future contact (for points and polygons this means that the iden-
tification should be as permanent as possible so that it can be recovered in a
subsequent period, and for farmers it is important to leave a good impression).
To achieve these objectives, the data collection activities must be carefully
prepared. We must provide environmental conditions that make the tools and
procedures as simple as possible and efficient mechanisms that quickly detect
problems and recover information that would otherwise be lost.
The fundamental aspects that must be considered are:
1. Preparation of the questionnaire and supplementary material.
2. Contact or identification of the survey units and data collection management.
3. Timing and interaction between the bodies responsible for data collection.
4. Campaigns to raise awareness in the farmers (even if the units are portions
of land).
5. Training of the interviewers.
6. Supervision of operations and recovery of incomplete information.
This general list is not exhaustive. It focuses on the major issues that we need to
take into account during the data collection phase.
Some recommendations can be useful when planning and conducting a survey. It
is usually good practice to inform and alert the population using the press or
professional associations (such as cooperatives of farms). The selected units,
farmers, or land owners should be informed of the survey through letters that
highlight the purpose of the research, the benefits of the information, the confiden-
tiality of the data, and the random selection criterion (in sample surveys). To reduce
the burden on respondents, one should also provide a contact telephone number
(preferably free of charge), which can be called for further clarification requests,
comments, or suggestions.
The personnel involved in the operations should be adequately informed on the
entire process, not just on the segment that is their responsibility. In particular, if we
plan to use enumerators, they should be informed about the gravity of partial and