Page 248 - Special Topic Session (STS) - Volume 4

STS583 Michael W. et al.
informativeness. Surveys based on an inadequate sampling frame may deliver
imprecise or biased estimates. Since any information related to sampling
errors is based on this frame, the errors related to inadequate sampling frames
usually remain undiscovered.
The problem of inadequate sampling frame is quite common, even in High
Income Countries (HIC). But where the latter commonly has an abundance of
administrative data to address these problems, LIC and MIC countries most
likely don’t have this fallback option, since the quality of their administrative
data is not sufficient so far. For this reason, the latter group of countries very
often relies on the information collected once every 10 years.
To overcome this considerable drawback, we propose the use of remote
sensing data as auxiliary information in the sampling frame. To further address
the issue of non-available sampling frames, we will also us this type of data as
a substitute for the census data.

2. Methodology
Simulation and Frame
To compare the efficiency of the different sampling frames and designs,
we will apply an empirical sampling simulation. In this type of (Monte-Carlo
style) simulation, either a true or synthetic population is used as the target
population. By applying a specific sampling design, and repeated sampling
(usually 1000 repetitions) under this design, we can compare the resulting
population estimates with the known true population values for each run of
the simulation.
The resulting distribution of these estimates is called the sampling
distribution, and the average squared deviation from the underlying
population value is the Mean Squared Error (MSE) or when taking its square
root, the Root MSE (RMSE). To facilitate the comparison, we use the relative
version expressed in percentage deviation.
Empirical sampling simulations can be considered as the “[…] ultimate tool
for investigators who want to know if one sampling strategy will work better
than another for their population.” (Thompson, 2013). However, this requires
the underlying simulation population to replicate as realistically as possible
the target population.
The target variables chosen were collected during the last census. We have
chosen variables of sufficient quality as well of different types (i.e. continuous
vs. ratio) and with different proportionality to the MOS.
With the simulation set up in this way, we conducted the following
experiments and compared the resulting estimates with each other:
i. Sampling from a conventional sampling frame stratified, by the available
census variables. This is the baseline scenario, and the commonly applied
approach for this type of survey. As mentioned at the outset, this approach

237 | I S I W S C 2 0 1 9