Understanding of sampling methods

Simple Random Sampling is the most basic form of sampling. Every individual or item from the population has an equal chance of being selected.
This method is analogous to a lottery. For fair results, it needs the availability of a complete list of the population.
Whilst it is straightforward, simple random sampling is susceptible to sampling error because the collected sample may not perfectly represent the population.

Systematic Sampling involves selecting items at regular intervals from a list or from sequential files.
Key to this method is that the starting point must be randomly chosen within the first interval to ensure randomness.
Systematic sampling can introduce bias if there is a pattern in the list that matches the interval.

Stratified Sampling involves dividing the population into ‘strata’, or subsets, based on shared characteristics.
A random sample is then selected from each stratum. Strata should be mutually exclusive and collectively exhaustive.
Stratified sampling provides a more accurate reflection of the population structure, but the stratification process can be complex and time-consuming.

Cluster Sampling is similar to stratified sampling but in this case, the population is divided into clusters, usually geographically defined.
With cluster sampling, all individuals within chosen clusters are included in the sample. This can save time and money but may lead to higher variability between clusters.
Unlike stratified sampling, clusters are often naturally occurring (like neighbourhoods or schools).

Quota Sampling is a non-probability sampling method in which the researcher selects people according to some fixed quota.
This method ensures that the sample represents specific characteristics of the population, which are proportionally represented.
The difference from stratified sampling is that quota sampling allows the researcher to choose the subjects, introducing a potential selection bias to the sampling.

Convenience Sampling is another non-probability sampling method where subjects are chosen because of their convenient accessibility and proximity to the researcher.
This process is cheap, quick and easy but the results can be unreliable as they may not accurately reflect the overall population.

Lastly, one must understand the difference between sampling errors and non-sampling errors. Sampling errors stem from the fact that not every individual from the population is included in the sample, and thus, the sample may not perfectly represent the population.
Non-sampling errors, however, are errors that arise due to human errors, such as incorrect data entry, misunderstanding of survey questions, or non-responses. Note that these errors can occur even if the whole population is being studied. This is why sample integrity should always be maintained, and non-sampling errors should be minimised as far as possible.