Blog Viewer

Terminology 101: Non-probability sampling
May 07, 2017, By: Maher M. El-Masri, RN, PhD

Non-probability sampling: The process of selecting research participants non-randomly from a target population

Non-probability sampling procedures involve recruiting participants in a non-random fashion for a research study. Thus, individuals in the accessible population don’t have an equal opportunity to be selected. This creates the potential for selection bias, limiting the generalizability of the study findings. Selection bias occurs when the study sample is not truly representative of the population of interest: those who participate in the study share an attribute (or attributes) that may be systematically different from the attribute(s) of those who don’t participate. For example, in a survey of flu vaccination among nurses, those who support flu vaccination may be more likely to participate than those who are indifferent to, or have a negative view of, flu vaccination. In such a case, one can’t fairly claim that the findings will be generalizable to the entire population of nurses.

Like probability sampling, non-probability sampling is a three-step procedure. The first step is to identify the target population — nurses, in our example. The second step is to identify the sampling frame, which is the portion of the target population that is accessible to the researchers — for instance, the 2,500 nurses at Hospital X. The third step is to recruit the required sample of nurses from the sampling frame (e.g., 150).

The main non-probability sampling techniques used in quantitative nursing research are convenience sampling and quota sampling.

Convenience sampling involves recruiting study participants on the basis of their availability. It is by far the most commonly used sampling procedure by quantitative nursing researchers. How might the researcher in our flu vaccination survey example go about recruiting the required sample? One approach would be to visit various floors of the hospital during various shifts to recruit whoever is available, until the required sample is attained. Alternatively, the researcher could post a general invitation to participate on bulletin boards across the hospital. With these approaches, nurses would be recruited in a non-random fashion when they happened to encounter the study’s recruitment process.

Quota sampling is a type of convenience sampling in which researchers make sure that certain attributes of importance to their research are proportionately represented in the sample. Quota sampling therefore improves sample representation. In our flu vaccination survey example, the researcher may have reason to believe that male and female nurses are systematically different in their attitudes toward flu vaccination. It would therefore be important for male nurses to be proportionately represented in the sample. The researcher would do quota sampling to ensure that the percentages of men and women in the study reflect their percentages in the nursing population.

Selection bias can’t be fully ruled out in non-probability sampling, but it can be minimized by ensuring that the sample shares the attributes of the population. It is extremely important that researchers using non-probability sampling clearly describe their recruitment procedures and the characteristics of their study sample. Readers can use such information to assess for selection bias by comparing the characteristics of the study sample with those of the target population. resources on this topic


Maher M. El-Masri, RN, PhD, is a full professor and research chair in the faculty of nursing, University of Windsor, Windsor, Ont.