Systematic sampling is a form of random sampling in which units are selected in an ordered fashion according to a periodic interval, i.e. having the 3rd or 5th person on every list sampled. This type of sampling method follows a particular pattern which is chosen and carried out throughout the whole process. For example, say there are 1000 individuals in a sample and each individual is numbered off, then a researcher could chose to sample every 50th person in that population and would have to be consistent with this pattern.
In comparison to other sampling methods, one disadvantage of systematic sampling is that the problem of bias can arise. For example, say a population of 1000 individuals is made up of 500 females and 500 males. If each individual was numbered off and every 50th person was sampled, it is possible that the sample could end up being almost entirely comprised of only females or males. In this case, the collected data would be biased towards one gender and not necessarily representative of the whole population.
The interval chosen in systematic sampling remains constant throughout the entire process. Additionally, the point of origin can be randomly chosen to be anywhere in the sample. For instance, using the previous example of 1000 individuals and an interval of 50, the starting point could be individual number 10. Then every 50th person after this starting individual would be sampled. Generally, a random starting point is implemented to eliminate the possibility of bias.
One technique which can be employed to minimize the chance of bias arising is to have more than one randomly selected starting point. For instance, maybe 4 starting points could be randomly chosen and then every 200th person from that starting point, in a total population of 1000 individuals, could be sampled.
Although there are limitations to systematic sampling, it is still useful to practice. It can also be considered advantageous in the sense that using a consistent interval ensures that the population will be evenly sampled. Furthermore, if the population you have is randomly distributed and there is no distinct pattern present, then the use of systematic sampling will not result in biased observations.
© BrainMass Inc. brainmass.com April 17, 2024, 12:38 pm ad1c9bdddf