Agriculture Reference
In-Depth Information
giving priority to the purpose of the study, and
the elements of the sample are selected in such
a way so as to fulfill the objective of the study.
This method is the simplest but may lead to
biased and inefficient sample depending upon
the efficiency of the investigator.
from the first 100 respondents whom the
researchers/interviewers have met and who are
willing to share information. Samples collected
in this process are subject to biasness, and the
only way to overcome the bias is to conduct
parallel studies.
3.
Purposive Sampling
: The selection of elements
to be included in the sample is entirely made on
the basis of the choice of the researcher.
Though easy to operate, it gives hardly a repre-
sentation of the parent population and as such
the results are biased and unsatisfactory.
6.3
Execution of the Sampling Plan
A good sampling plan is an “essential” condition
for the study to be undertaken, but it is
not sufficient to have a good sample leading to
efficient and accurate estimation of the population
parameter. It is necessary to have a “good” sam-
pling plan suitable for the study followed by the
“efficient execution” of the sampling plan. A
“good” sampling plan, if not executed properly,
may give “bad” (unreliable, inaccurate) results,
leading to wastage of entire time, energy, and
money used for the investigation. For the efficient
execution of the sampling plan, the investigator
responsible for the data collection must possess
the necessary qualifications. They must be prop-
erly trained before the data collection. They must
be taught how to handle the equipment and to
make correct observations and measurements and
note them down carefully. A proper supervision of
the fieldwork is a must, and scrutiny and editing of
the collected data is essential. Precautions must be
taken to identify the sample units, in specifying the
units(s) of measurements at various stages, and to
minimize the error in recording the data. Pilot
survey may be undertaken to select the suitable
sampling plan among the alternative plans. An
efficient execution of sampling plan cannot only
reduce both the sampling and non-sampling errors
but also helps in reducing the cost of the study.
4.
Snowball Sampling
: In researches, particularly
in the field of social studies, related to sensitive
issues like drug abuse and HIV victims in
which the respondents are very much hesitant
because of social taboos, reasons, or otherwise,
this type of sampling is very useful. In this type
of study, the researcher finds out initially one or
two or a few respondents for the study purpose
and then takes the help of these respondents to
find more and more respondents on specific
aspects. This is because of the fact that this
type of respondents may be hesitant to present
themselves when approached by a stranger
rather than a fellow friend. Thus, in this type
of studies, size of sample goes on increasing as
one gets more andmore number of respondents.
Accidental Sampling
5.
: This type of sampling is
mostly used in evaluative types of research
studies. Suppose one wants to know the extent
of adoption of a particular technology, say, birth
control program/hybrid rice technology in a
certain community. In this method of selecting
the respondents, the researchers/interviewers
take information from the respondents who hap-
pen to meet the researchers first, and in the
process, the researchers may take information
