Setting Up the Frame - Sampling Spatial Units for Agricultural Surveys - page 91

Chapter 5
Setting Up the Frame

5.1 Introduction

One of the aspects that have the greatest impact on the quality of the results of a
survey is the type of frame used to select the units from the population. Neverthe-
less this aspect is very often underestimated in survey methods. The sampling frame
has significant implications on the cost and the quality of any agricultural or
non-agricultural survey. In agricultural surveys, defective sampling frames are a
common source of nonsampling error, particularly under-coverage of important
population sub-groups. There is, therefore, a need to elaborate best practices in
frame construction and usage taking into account various stages of sampling.
The sampling frame is any material or device used to obtain observational access
to the finite population of interest (S ¨ rndal et al. 1992 ), and refers to the list of units
in the survey population. In other words, the frame is the means by which the
statistical units to be enumerated in the survey are identified (FAO 2005 ). It must be
possible with the aid of the frame to identify and to select a sample in a way that
respects a given probability sampling design and to establish contact with selected
elements. A sampling frame could be a list of units (such as households or
holdings), geographical areas, or any other materials (such as maps), and may
also include information about each unit, such as their size, to help with the sample
selection or survey estimation (FAO 1989 , pp. 32-41; United Nations 1986 ).
Some properties that a frame should have are the following (Colledge 2004 ):

• The units of the frame are identified through a unique identifying code.
• All units that are selected in a sample should be contactable.
• The frame contains a set of additional information for each unit.
• Each element in the population of interest is present in the frame (completeness),
and only once (accuracy).
• There is no element in the frame that is out of the target population.
• Frame data are up-to-date.