Jan 05, 2017, By: Maher M. El-Masri, RN, PhD
Sampling: The process of selecting research participants from a target population who have the characteristics of that population
In quantitative research, researchers are interested in drawing conclusions about a population that has common characteristics that interest them. This population is known as the target population. Given that most populations are too large to study in their entirety because of time, cost and logistical constraints, researchers often conduct research on samples that they assume to be representative of the target population. A representative sample is one that shares the characteristics of the population from which it was drawn. This is an important condition in research sampling because if the research sample is not truly representative of the target population (in other words, if it is biased), the study will probably generate non-generalizable results.
To obtain a representative sample, researchers conduct probability sampling or non-probability sampling (to be discussed in the next two columns). The vital distinction between these two types of sampling is that probability samples are randomly selected from the target population and non-probability samples are not. Because selection is random in probability sampling, everyone in the target population has an equal opportunity to be included in the study; any biases the investigators may have cannot influence the selection process. In non-probability sampling, participants are chosen on the basis of researchers’ assumptions about the characteristics of the target population, some of which may not be known to the researchers. Because non-probability sampling ignores these unknown characteristics, the generalizability of the research findings is limited.
Although probability sampling is the superior method, it is important to keep in mind that at times it may not be feasible, for example, in a clinical research study at an acute care setting in which the accessible population is not known beforehand (because patients enter into the population as they become ill).
Another important way to judge the appropriateness of the sample is by its size. Sample size should be determined on the basis of empirical calculation, which should be described in the paper. If the sample is too small, the study may lack the statistical power needed to detect a true effect, and the researchers may end up drawing incorrect conclusions about this effect (i.e., they may conclude that an effect doesn’t exist when it does).
Sampling procedures and sample size are too important to be ignored when one evaluates research reports. Readers need to be convinced that the study participants possess characteristics that reflect those of the target population. They also need to look for evidence that the sample size was not determined arbitrarily or on the basis of a hunch. Any evidence of sampling problems should raise serious concerns about the validity of the research findings.
Maher M. El-Masri, RN, PhD, is a full professor and research chair in the faculty of nursing, University of Windsor, Windsor, Ont.