There are distinct advantages and disadvantages of using systematic sampling as a statistical sampling method when conducting research of a survey population.
Systematic Sampling: An Overview
Systematic sampling is simpler and more straightforward than random sampling. It can also be more conducive to covering a wide study area. On the other hand, systematic sampling introduces certain arbitrary parameters in the data. This can cause over- or under-representation of particular patterns.
Systematic sampling is popular with researchers because of its simplicity. Researchers generally assume the results are representative of most normal populations, unless a random characteristic disproportionately exists with every "nth" data sample (which is unlikely).
To begin, a researcher selects a starting integer on which to base the system. This number needs to be smaller than the population as a whole (e.g., they don't pick every 500th yard to sample for a 100-yard football field). After a number has been selected, the researcher picks the interval, or spaces between samples in the population.
- Because of its simplicity, systematic sampling is popular with researchers.
- Other advantages of this methodology include eliminating the phenomenon of clustered selection and a low probability of contaminating data.
- Disadvantages include over- or under-representation of particular patterns and a greater risk of data manipulation.
Systematic Sampling Example
In a systematic sample, chosen data is evenly distributed. For example, in a population of 10,000 people, a statistician might select every 100th person for sampling. The sampling intervals can also be systematic, such as choosing one new sample every 12 hours.
Advantages of Systematic Sampling
The pros of systematic sampling include:
Easy to Execute and Understand
Systematic samples are relatively easy to construct, execute, compare, and understand. This is particularly important for studies or surveys that operate with tight budget constraints.
Control and Sense of Process
A systematic method also provides researchers and statisticians with a degree of control and sense of process. This might be particularly beneficial for studies with strict parameters or a narrowly formed hypothesis, assuming the sampling is reasonably constructed to fit certain parameters.
Clustered Selection Eliminated
Clustered selection, a phenomenon in which randomly chosen samples are uncommonly close together in a population, is eliminated in systematic sampling. Random samples can only deal with this by increasing the number of samples or running more than one survey. These can be expensive alternatives.
Low Risk Factor
Perhaps the greatest strength of a systematic approach is its low risk factor. The primary potential disadvantages of the system carry a distinctly low probability of contaminating the data.
Disadvantages of Systematic Sampling
There are also drawbacks to this research method:
Assumes Size of Population Can Be Determined
The systematic method assumes the size of the population is available or can be reasonably approximated. For instance, suppose researchers want to study the size of rats in a given area. If they don't have any idea how many rats there are, they cannot systematically select a starting point or interval size.
Need for Natural Degree of Randomness
A population needs to exhibit a natural degree of randomness along the chosen metric. If the population has a type of standardized pattern, the risk of accidentally choosing very common cases is more apparent.
For a simple hypothetical situation, consider a list of favorite dog breeds where (intentionally or by accident) every evenly numbered dog on the list was small and every odd dog was large. If the systematic sampler began with the fourth dog and chose an interval of six, the survey would skip the large dogs.
Greater Risk of Data Manipulation
There is a greater risk of data manipulation with systematic sampling because researchers might be able to construct their systems to increase the likelihood of achieving a targeted outcome rather than letting the random data produce a representative answer. Any resulting statistics could not be trusted.