Systematic Sampling Explained: What Is Systematic Sampling?

Systematic sampling, or systematic random sampling, is a probability sampling method that picks respondents from a larger population by choosing them at regular intervals. This is the method of sampling used when a researcher picks every nth person in a group to be part of a study.

For an example of systematic sampling, imagine that a group of 300 people walk into a room. They are what's known as a sampling frame—the total population from which respondents will be selected. As they enter, they pick slips of paper out of a bowl, and each slip of paper has a number on it, from one to 300, inclusive. A statistician could announce that everyone whose number ends with a nine has been chosen for a study. This will produce a sample size of thirty people because there are thirty numbers between one and 300 that end in a nine. In addition to getting the desired sample size, the statistician should also get a representative sample of the people in the room with an equal probability of having certain traits because everyone drew a random number.

The systematic sampling procedure features the following core characteristics.

1. 1. More manageable form of simple random sampling: When a target population size is too unwieldy for a simple random sample, systematic sampling lets you create a manageable data set while still selecting members of the population at random. If properly conducted, a systematic random sample will produce the same results as a more time-consuming random sample of a larger population.
2. 2. A form of probability sampling: Even though systematic sampling may pick respondents at fixed intervals, it still is a type of probability sampling that features random selection. This makes it different from convenience sampling or stratified sampling techniques, which feature a non-random selection of participants.
3. 3. A random starting point: Systematic sampling, where every nth person is chosen for a study, only works if the entire population is assembled in random order. If they start out in subgroups—like the first twenty people in the room are all doctors and the next twenty people are all lawyers—you will not get random results when you select people in fixed intervals. You also won't get random results if people are already arranged in a particular order, such as from oldest to youngest. You must begin with true randomness in order to get representative sampling units.

Systematic sampling is used when two factors converge.

1. 1. Researchers or statisticians need a random sample from a target population. Random sampling (also called probability sampling) is the gold standard for most types of research. (One exception is stratified sampling, which is a form of non-probability sampling that addresses imbalance in a sampling frame.) As such, many studies will require randomly selected people from a larger population.
2. 2. The size of the population is overwhelming. This type of sampling shows its value when statisticians need to pull a representative sample out of an otherwise overly large group. By creating a consistent sampling interval (in other words, studying every nth person in a group), the statisticians can get a manageable sample size that should still be representative of the entire population.

The advantages of systematic sampling are clear. It is an organized variation on simple random sampling with a low risk for data manipulation. Most people, from professional statisticians to laypeople, can understand its methodology. When done properly, it presents random results that should be representative of the target population at large.

The key disadvantage to systematic sampling is that it only produces high-quality data when researchers truly understand the size of a population and are able to select people from that population at random. For instance, if a California-based researcher wanted to create a sample population from people who moved to California from a different state, they would need to know exactly how many Californians did in fact start in a different state. They would then need a reliable method for reaching a truly random sample of those transplants. If they couldn't identify the total population size and reach a random sampling of that population, their survey could devolve into convenience sampling, where respondents are selected based on ease, not randomness. Such sampling is not considered as qualitatively valuable as true systematic random sampling.

Get the MasterClass Annual Membership for exclusive access to video lessons taught by science luminaries, including Terence Tao, Bill Nye, Neil deGrasse Tyson, Chris Hadfield, Jane Goodall, and more.