Sample size
Sample size can be statistically calculated. However, the researcher can proceed without
calculating a scientifically precise sample size. The researcher needs to be guided by the
fact that a sample that is too small is not likely to be representative while one that is too
large is likely to increase costs and time demands without producing significant benefits.
•
Generally, a larger sample is advisable to ensure good respresentation.
• Sub group the sample into smaller
units for analysis
As variability increases, it means that people/aspects in the sample are more diverse. This
implies that more information is required to be representative of the sample.
It is very important that researchers should keep the following in mind when determining
the size of their samples:
•
In general, the larger the sample (within reason), the smaller the sampling error and
the higher the representativeness of the sample.
• If the researcher wishes to discuss the influence of variables such as gender, age,
and profession in the findings, then the initial selection of the sample must be large
enough to enable the eventual creation of statistically
sufficient numbers in each
category or sub-sample .
•
While a big sample is good, an appropriate-sized is better as it is more cost-effective
in terms of time and money.
The following image is a convenient summary of the various sampling methods.
Dostları ilə paylaş: