Page 75 - SREENARAYANAGURU OPEN UNIVERSITY

Discussion

2.2.1 Sample Size

There are many myths about the sample size determination like
if we take larger samples, we may get accurate results but the fact
is larger sample size would help to increase the precision of results
not exactly accuracy. There is no fixed sample size but we have ◆ Number of items to
references, formulae and concepts to determine the sample size. be selected
The number of items to be selected from the universe is known as
the sample size. An optimum sample size would be representative
of the population. Large samples are considered as 30 or more and
less than 30 are known as small sizes. The statistic obtained from
a large sample can effectively estimate population parameters.

2.2.2 Factors Influencing the Sample Size

The following are the factors which influence sample size.
a. Confidence level: It indicates the probability, with which the
estimation of a statistical parameter (e.g., an arithmetic mean) in ◆ Degree of
a sample survey is also true for the population. It is the degree of confidence
confidence with which we can estimate a population parameter
with a sample statistic. It is usually taken as 90 percent or 95
percent.

b. Margin of error: A margin of error shows the percentage
that the results will differ from the real population value. The
margin of error represents the maximum amount of error you can
tolerate in your results. It is usually expressed as a percentage ◆ Precision
and determines the precision of your estimates. A smaller margin
of error requires a larger sample size. For example, a 95 percent
confidence interval with a 3 percent margin of error means that
your statistic will be within (+/- 3) percentage points of the real
population value 95 percent of the time.

c. Population variability: The variability or dispersion within ◆ Dispersion
the population under study influences the sample size. Higher
variability requires a larger sample size to accurately represent the
population.
d. Population size: The population may be known and finite
or unknown and infinite. Different approaches are adopted in
each case. Larger population size requires a larger sample size to ◆ Finite and infinite
ensure representativeness. The reason is that when the population population
is large, there is a higher chance of variation and diversity within
it. Therefore, a larger sample is needed to capture that variability

SGOU - SLM - MCom Research Methodology 69