Survey Data Collection and Processing - Sampling Spatial Units for Agricultural Surveys - page 218

(when the respondents are farms). To achieve this effectively, a questionnaire
should:

1. Provide clear instructions for respondents, enumerators, and processors.
2. Clearly define what is to be collected and recorded, maintain respondents '

cooperation and involvement, to avoid monotonous and confounding surveys.
3. Enable respondents to accurately complete the survey within a reasonable time.
4. Use a language that is understood by the respondents and enumerators, to avoid
biases in the wording of questions.
5. Make the job of the respondent and/or interviewer easy.
6. Provide suitable space for responses.
7. Allow easy processing by either manual or automated means.
8. Be in a form that is suitable for keeping as a hard copy record.

The survey forms must consider all the typical aspects of agricultural surveys
based on a list, areal or point frame. These include the method of data collection,
characteristics of the respondents, statistical burden, and complexity of the data. A
general questionnaire using a face-to-face technique includes information on land
use, livestock numbers, crops, rural development, management, and farm labor
(including the age, gender, and relationship to the holder of the agricultural
holding).
When the units are pieces of land (segments), a segment is divided into tracts of
land using field enumeration, each representing a unique land operating arrange-
ment. If possible, segment boundaries can be pre-printed onto the questionnaire.
However, any necessary changes result in complex operations. A questionnaire is
completed for all tracts within a segment, and contains screening questions that
determine whether each tract has agricultural activity. The screening applies to all
land (both inside and outside the segment). Through this process, any tract showing
agricultural activities is subsequently interviewed using the area version survey
form, which collects detailed agricultural information about the operator s land

'

(crops and land use).
When units are points, the geographical coordinates are generally pre-printed on
the form, and the enumerator adds the required information. In this case, the
questionnaire can be divided into three parts. The first contains information on
the conditions of observation (start and end times of the observation, type of
observation, distance between the enumerator and the point observed, diameter of
the possible circle and extended window around the point, etc.). The second part
concerns the land use and land cover (with two or more possible codes for each).
Other information may be requested, for example, height of trees, width of features,
information (land cover transitions and linear features) about the transect (straight
line of a certain length in one direction from the observation point), land manage-
ment (grazing), water management, and soil sample.
In Fig. 9.1 we show a simplified scheme of a questionnaire that considers a point
frame, and in Fig. 9.2 we consider an irregular polygon frame.