Sample

Definition and Examples

A sample is a subset of a population. The population can be a set of people, animals, or things. For example, the population might be made up of all children who, on a given day, are enrolled in a particular elementary school; the dogs, during a given month, that are treated at a particular veterinary hospital; or the boats, on a given evening, that are docked at a particular marina. With these entities as the possible populations, a sample would be a subset of school children, a subset of dogs, or a subset of boats, respectively. The notion of subset implies that a sample must be smaller than the population to which it is connected. If n represents the size of the sample and N represents the size of the population, then it is mandatory that n < N. If the sample and the population are identical in quantity, then only the population exists.

Reasons Why Researchers Use Samples

Most researchers use samples in their studies because they do not have the resources—time, money, or access—to measure or observe all members of the population of interest. If a researcher wants to describe the eating behavior of the dogs observed at a particular veterinary hospital, the researcher’s available time and money would likely prohibit the in-home observation of every dog. However, observing a subset of those dogs might be feasible given the study’s budget and duration. Here, as in many other situations, the researcher chooses “what’s possible” (even though it is not best) instead of “what’s best” (because it is not possible).

In certain studies, researchers use samples because they have no choice. In laboratory and field experiments, and in clinical trials, a group of people, animals, or things is given some form of treatment. Posttreatment measurements are used as a basis for comparing the treated group against a control group, a placebo group, or a different treatment group. Or the treatment group’s posttreatment scores are sometimes compared against the same group’s pretreatment scores. In such investigations, the treatment group typically is considered to be a sample of a larger group that, in the future, might also receive the treatment. For obvious reasons, it is not possible to observe or measure the entire population.

Typical Use of Sample Data

After the members of a sample are identified and measured, researchers normally summarize the sample data and then use what they know about the sample to make an educated guess (i.e., a statistical inference) about the population. For example, if the population of interest is defined as the freshmen who enter a particular university in the fall of a given year, a sample of those freshmen might be interviewed to find out what they worry about. If these interviews reveal that 60% of the students in the sample say they worry about “doing well academically,” this information concerning the sample—referred to as the statistic—would be used to make a guess about the percentage of students in the freshman class who worry about grades—the parameter.

...