Criteria For Selecting A Sampling Procedure: Basically, two costs are involved in a sampling analysis, which govern the selection of a sampling procedure.
1) The cost of data collection.
2) The cost of drawing incorrect inference from the selected data.
There are two causes of incorrect inferences, namely systematic bias and sampling error.
Systematic bias arises out of errors in the sampling procedure. They cannot be reduced or eliminated by increasing the sample size. Utmost, the causes of these errors can be identified and corrected.
Generally, a systematic bias arises out of one or more of the following factors:
a. Inappropriate sampling frame
b. Defective measuring device
d. Indeterminacy principle
e. Natural bias in the reporting of data
Sampling error refers to the random variations in the sample estimates around the true population parameters.
Because they occur randomly and likely to be equally in either direction, they are of compensatory type, the expected value of which errors tend to be equal to zero.
Sampling error tends to decrease with the increase in the size of the sample. It also becomes smaller in magnitude when the population is homogenous. Sampling error can be computed for a given sample size and design.
The measurement of sampling error is known as ‘precision of the sampling plan’.
When the sample size is increased, the precision can be improved. However, increasing the sample size has its own limitations.
The large sized sample not only increases the cost of data collection, but also increases the systematic bias. Thus, an effective way of increasing the precision is generally to choose a better sampling design, which has smaller sampling error for a given sample size at a specified cost. In practice, however, researchers generally prefer a less precise design owing to the ease in adopting the same, in addition to the fact that systematic bias can be controlled better way in such designs.