Survey Data Collection and Processing - Sampling Spatial Units for Agricultural Surveys - page 217

experienced and well-trained is critical to the success of the data collection stage.
The quality of data collection and the results of the survey depend on the consistency
and quality of these staff. Therefore, training and manuals are important, and a
proper investment in human capital formation is necessary for a successful survey.
After the units have been located or contacted, the observation, direct measure-
ment, or interviews can begin. This involves more than simply observing a crop or
asking questions to a farmer. It includes, for example, checking if the sample unit
(or part of it when dealing with polygons) belongs to private land, and contacting
the owner to ask for permission to perform all the necessary operations.
The respondent collaboration also needs to be considered, particularly when
farmers are involved in the interview, because it can minimize the nonresponse rate.
Failures may occur when recording the data, and we should always monitor the data
quality. Data collection can only be considered complete when all of these activities
have been finished. Then, we can proceed to estimate the parameters of the vari-
ables of interest.
This chapter is organized as follows. Section 9.2 outlines the basics of a
questionnaire design, and in Sect. 9.3 we describe the organizational and technical
aspects of data collection. In Sect. 9.4 , we briefly summarize the main features of
the data editing procedures together with some suggestions regarding the metadata
to be published in support of the estimates. We outline some data quality problems
in Sect. 9.5 , with a brief description of the characteristics of an additional survey
devoted to quality controls and an index that can serve as supporting metadata. Our
conclusions are presented in the last section.
9.2 Questionnaire Design
Questionnaires (or survey forms) play a central role in the data collection process
(Madans et al. 2011 ). A poorly designed survey form can be the biggest direct or
indirect source of non-sampling error. The questionnaire can influence the response
rate and the quality of responses, which then further influences the conclusions
drawn from the survey results.
A questionnaire design should consider the purposes of the survey and the
required output. A list of questions should then be planned to accurately record
this information. Careful consideration should be given to a number of factors,
including the types of questions, the questionnaire wording, the structure and design
of the survey form, and questionnaire tests to ensure quality. Testing the draft
questionnaire allows problems to be identified and corrected prior to conducting the
full survey.
The main functions of questionnaires are to collect, record, and process accurate
and relevant information. This is true for both spatial surveys and list frame surveys