Agriculture Reference
In-Depth Information
in the social science researches, there should be
provision for cross-checking the responses in the
methods of collection of information.
4.1
Characteristics of a Good
Research Design
A good research design is one that helps in
reaching the objective of a research program
in the best possible way.
4.1.3 Validity
Objectivity
,
reliability
,
validity
are the main charac-
teristic features of a good research design.
, and
generalization
While using any measuring instruments, a
researcher should be sure that the instrument
selected for use is a valid one. An originality
test constructed for measuring originality should
measure none other than originality. In literature,
there are a number of procedures for establishing
the validity of a test.
4.1.1 Objectivity
By objectivity, we mean observations free from
bias, from the observer's point of view. When we
say “blood is red,” it is an objective statement,
but when we say “blood is the most useful ele-
ment in the human body,” then the statement may
not be a purely objective one, because of the fact
that the human body cannot survive only with
blood; for an effective functioning of the human
body, other useful components are needed along
with blood, meaning blood can be a useful com-
ponent, but most probably one cannot assign the
word “most” before useful. In many research
programs, various types of instruments and
measures are being used. Good research designs
use only those instruments and measures which
are precise in nature. Generally it is felt that
objectivity can be achieved easily, but in actual
situations, it is very difficult to attain the objec-
tivity from a host of competitive available
sources.
4.1.4 Generalization
Generalization of the outcome of a research pro-
gram is one of the important points to be noted
during the formulation of the research design. In
most of the research programs, though the
researcher starts with a particular problem for a
particular situation, at the end of the day, the
researcher would have the intent to explore the
possibility of using such outcomes in other rele-
vant fields and would precisely want to general-
ize the research outcome. That is, how best the
information or solution generated by the present
research work would be applied to a large group,
from which the samples have been drawn. This is
possible only when effective steps are taken dur-
ing the formulation of the hypothesis, data col-
lection, and of course analysis and interpretation
of information. For example, if a sample size is
large (
4.1.2 Reliability
30 generally), then one can use the infor-
mation applying the central limit theorem or
normal approximation. But if a sample size is
low, then specific probability distributions are
required to be ascertained and followed; as such
generalization of the result becomes restricted.
>
Reliability refers to consistency and authenticity
in responses. Consistency means the respondent
should not provide a different answer every time
the same question is asked by the investigators in
different forms for a particular problem. An
instrument is consistent if its response remains
same under repeated observations, essentially
under the same condition. Authentic information
is obtained from a source which has the authority
and credential to report about the problem. To
verify the reliability of the responses, particularly
4.2
Types of Research Design
Research designs vary depending upon the type
of research. By and large, research designs
are framed in three types: (a) exploratory type
