EDUCATIONAL RESEARCH
SAMPLING DESIGN
DEFINITION:-
According
to W.G.Cochran, “In every branch of science lack the resources to study more
than a fragment of phenomena that might advance our knowledge”. For study any problem it is difficult to
study the whole population or universe.
Study the whole population or universe.
Studing is entire universe is not viable in many ways.
SAMPLE :-
It is a small portion of a population selected
for observation and analysis. By
observing the characteristics of the sample, one make certain inferences about
the characteristics of the population from which it is drawn.
SAMPLING:-
It is the process of selecting a sample
from the population.
SAMPLING ERROR:-
It is the differences between the value of
parameter and that of the corresponding statistic.
STANDARD ERROR:-
The standard error of any statistics is the
standard deviation of sampling distributions.
TYPES OF SAMPLE DESIGN
Sampling Strategies |
Probability Based Non – Probability
Based
 |
 |
||
Judgement Incidental Quota
Snow ball sampling Simple Stratified Systematic Double Cluster Multiple
Random Random Random Sampling
Proportionate Disproportionate
CHARACTERISTICS OF PROBABILITY SAMPLING:-
v
It refers
from the sample as well as the population.
v
It may be
representative of the population.
v
Inferential
parametric statistics are used.
v
There is
a risk for drawing conclusions.
v
Its
represtatireness refers to characteristics.
SIMPLE RANDOM:-
Researcher
assigns each member of the sampling frame a number, then selects sample units
by a random method.
ADVANDAGES:-
v
Only
minimal advance knowledge of population needed.
v
Easy to
analyse data and computer error.
DISADVANTAGES:-
v
It cannot
ensure the representativeness of a sample.
v
It does
not use the knowledge about the population.
SYSTEMATIC SAMPLING:-
It is an improvement over the simple random
sampling. It requires complete
information about all the individuals of the population in a systematic way.
ADVANTAGES:-
v
It is a
simple method of selecting a sample.
v
It
reduces the field cost.
v
Inferential
statistics may be used.
DISADVANTAGES:-
v
It can
not ensure representativeness.
v
There is
a risk in drawing conclusions from the observations of the sample.
v
Knowledge
of population is essential.
STRATIFIED SAMPLING:-
It
is an improvement over systematic method.
The researcher divides his population in strata on the basis of some
characteristics from smaller homogenous groups, (Strata). The researcher draws at random a
predetermined number of units.
ADVANTAGED:-
v
It is a
good representative of the population.
v
It is an
improvement over the earlier.
v
It is an
objective method of sampling.
DISADVANTAGES:-
v
Only the
criterion can be used for stratification.
v
It is
costly and time consuming.
v
There is
a risk in generalization.
DOUBLE SAMPLING:-
It is a new
application of the alone samplings. It
is most frequently used for establishing the reliability of the sample.
ADVANTAGES:-
v
It
reduces the error.
v
It
maintains the procedure of the finding to evaluate the reliability of the
sample.
DISADVANTAGES:-
v
It can
not be used for a large sample.
v
It is
time consuming and costly.
v
It requires
more competition.
MULTI – STAGE SAMPLING:-
It
is more comprechensive and representative of the population.
ADVANTAGES:-
v
It is a
good representative of the population.
v
It is an
improvement over the earlier methods.
v
It is an
objevtive procedure of sampling.
DISADVANTAGES:-
v
It is
difficult and complex.
v
It
involves errors while considering the
primary and secondary stages.
v
It is
subjective.
CLUSTER SAMPLING:-
It is the sample
units contain groups of elements (Clusters) instead of individual members.
ADVANTAGES:-
v
It is the
good representative of the population.
v
It is an
easy method.
v
It is an
economical method.
DISADVANTAGES:-
v
It is not
free from error.
v
It is not
comprehensive.
NON-PROBABILITY SAMPLING:-
CHARACTER’S:-
v
There is
no idea of population.
v
It has
free distribution.
v
There is
no risk for drawing conclusions.
JUDGEMENT SAMPLING:-
It
involves the selection of a group from the population on the basis of available
information, it should be representative of the total population.
ADVANTAGES:-
v
In it
knowledge of the investigator can be best used.
v
It is
economival.
DISADVANTAGES:-
v
It is
objective.
v
It is not
free from error.
v
It
includes uncontrolled variation.
INCIDENTAL ASSIGNMENT:-
Samples
are taken because they are most frequently available.
ADVANTAGES:-
v
It is an
easy method of sampling.
v
It is
frequently used in behavioural sciences.
v
It is an
economical method.
DISADVANTAGES:-
v
It is not
a representative of the population.
v
It is not
free from error.
v
Parametric
Slatisties cannot be used.
QUOTA SAMPLING:-
It combines both judgement sampling
and probability sampling . The
proportion of population folling int’s each category is decided on the basis of
judgement.
ADVANTAGES:-
v
It is an
improvement over the judgement sampling.
v
It is and
easy sampling technique.
v
It is
most frequently used in social surreys.
DISADVANTAGES:-
v
It is not
a representative sample.
v
It is not
free from error.
SNOWBALL SAMPLING:-
Initial
respondents are selected by probability samples.
ADVANTAGES:-
v
Useful in
locating members of rate populations.
DISADVANTAGES:-
v
High bias
because sample units not independent, projecting data respondents beyond
inappropriate.
REFERENCES:-
v
Research
in education.
Sotishivendra. Chandra.
v
Research
methods in education.
Dr.Radha Mohan.
0 comments:
Post a Comment