Sampling Methods and Bias: Learn It 2

Systematic Sampling

In systematic sampling, every individual in the population is given a number, and individuals/entities are chosen at regular intervals, with a random starting point (usually among the first several).

The following figure illustrates a systematic sample where every 4th individual is selected, starting at the 3rd individual (starting point selected at random). The individuals selected for the sample are highlighted.

Appropriate alternative text can be found in the description above.
Figure 1. A systematic sample selects individuals at regular intervals, starting from a randomly chosen point.

Stratified Sampling

In stratified sampling, a population is divided into two or more groups (these groups are collectively called strata, referring to one of several groupings is known as a stratum) according to some criterion (i.e., geographic location, grade level, age group, income group, etc.), and a sample is selected from each stratum using simple random sampling or systematic sampling.

In the illustration below, the population is divided into two groups (blue and green), and 4 samples were selected from each group.

The population is split evenly in two groups of 18, shown in blue and green. In the first group (blue), individuals 2, 3, 9, and 14 are selected. In the second group (green), individuals 2, 5, 6, and 13 are selected.
Figure 2. A stratified sample is created by dividing the population into groups and then selecting individuals from each group.

Cluster Sampling

Similar to stratified sampling, for cluster sampling, a population is divided into two or more groups (called clusters) according to some criterion. Clusters are then randomly selected, and every individual in the cluster(s) is included in the sample.

In the illustration below, the population is divided into four groups (green, purple, blue, and gray), and an entire group (blue) was chosen as the sample.

The population is divided into four groups (green, purple, blue, and white) and one entire group is selected for the sample.
Figure 3. A cluster sample selects an entire group from a divided population.

Convenience Sampling

A convenience sampling is a sample of individuals who are most accessible to the researcher. A convenience sample is usually not random or representative of the population. This is an example of a biased sampling method because it tends to produce samples that are not representative of the population. For example, you might take a sample of your friends because it is easy to collect information about them.