An overview of sampling methods
In the study of a particular phenomena present in the population it may not be possible to examine each member of the population. This is because populations under research are often too large. In this context, researchers use samples to draw conclusions on phenomena under study for large populations. Sampling is thus a critical research method. There are probability-sampling methods such as random sampling and stratified sampling (Linda 2008).
The other probability-sampling method is systematic sampling. The other category of sampling methods is non-probability sampling. Non-probability sampling includes convenience sampling and quota sampling. Other methods of non-probability sampling include snowball sampling and judgement sampling. The basic difference between probability and non-probability sampling methods is that in the former the sample set is constructed in a random manner.
Different sampling methods have their own advantages and disadvantages. The basic characteristic of the sample is that it ought to be representative of the population from which it is drawn. This means that the sample ought to have a distribution of traits as those of the population. This is critical in ensuring that the sample results can be generalized to the wider population. This offers validity and reliability to the results of the research.
Reliability of results
Probability sampling methods implies that the sample is drawn from the population in a random manner (Linda 2008). This means that each member of the population has a certain chance of being picked into the sample set unlike in the non-probability sampling.
This is one of the chief advantages of the probability sampling as the research is generalizable to the wider population. This is because the sample is representative of the general population traits. In this context, the research results achieved from this type of sampling can be generalized to the whole population.
While probability sampling offers some degree of objectivity in the construction of sample set, certain contexts necessitate non-probability sampling methods. Such contexts include instances where the researcher has limited time and financial resources to conduct probability sampling.
In instances where non-probability sampling method is used there is a limitation to the generalization of the results to the wider population (Black 1999).This is because the sample may be picked from a certain section of the population accessible to the researcher. In this context, the sample may not be truly representative of the whole population and as such, the results of the population may not be generalized to the whole population (Jacobs 2011).
Accuracy of the research findings
The question of accuracy of research findings may be closely related to that of the reliability of the results. However, there might be a slight difference depending on how a researcher looks at the two concepts. Reliability of the results means that the results generally do give the trend or the traits of the studied phenomena. On the other hand, accuracy of the research findings do try to give a measure or degree of how the findings apply to the studied phenomena.
In the context of accuracy of the research findings, the probability research findings do give a quantitative measure on the accuracy of the findings. This is achieved through the calculation of the sampling error. Sampling error is the degree of variation of the sample traits from those of the population. Thus in reporting the research findings, the results are given with a qualitative margin of error. Such qualitative margin of error may be useful to some researches depending on the phenomena under study.
On the other hand, the non-probability research methods are more subjective in nature and lacks the randomness associated with probability sampling (Black 1999).In non-probability, sampling the sample error is not calculated. In this sense, it is difficult to establish the degree to which the sample traits differ from those of the population. This may not necessarily mean that the accuracy of the research finding is in doubt.
In some instances, the researcher may not be interested in putting a quantitative measure on the accuracy of the findings of the research. On the contrast, he may be interested in understanding the nature of the phenomena under study. This is especially so in research on social aspects.
A researcher’s dilemma; which sampling method to use?
Several aspects influence a researcher’s choice of sampling method. Availability of resources such as time and finances is one aspect that greatly influences the choice of sampling methods. Probability sampling methods are relatively expensive in obtaining an optimal sample size from the population (Jacobs 2011).
This gives the researcher to use non-probability sampling techniques in which he may draw sample members that are convenient and easily accessible to him. The nature of the phenomena under study also influences the choice of the sampling methods. Scientific studies are generally quantitative in nature and thus may require probability-sampling methods. This is in contrast to the social studies.
References
Black T 1999, Sampling techniques: Advantages and disadvantages. Web.
Jacobs R 2011, Educational Research: Sampling a population. Web.
Linda W 2008, Sampling methods. Web.