StATS: What is a convenience sample?

A convenience sample is a sample where the patients are selected, in part or in whole, at the convenience of the researcher. The researcher makes no attempt, or only a limited attempt, to insure that this sample is an accurate representation of some larger group or population. The classic example of a convenience sample is standing at a shopping mall and selecting shoppers as they walk by to fill out a survey.

In contrast, a random sample is one where the researcher insures (usually through the use of random numbers applied to a list of the entire population) that each member of that population has an equal probability of being selected. Random samples are an important foundation of Statistics. Almost all of the mathematical theory upon which Statistics are based rely on assumptions which are consistent with a random sample. This theory is inconsistent with data collected from a convenience sample.

In general, the Statistics community frowns on convenience samples. You will often have great difficulty in generalizing the results of a convenience sample to any population that has practical relevance.

Still, convenience samples can provide you with useful information, especially in a pilot study. To interpret the findings from a convenience sample properly, you have to characterize (usually in a qualitative sense) how your sample would differ from an ideal sample that was randomly selected. In particular, pay attention to who might be left out of your convenience sample or who might be underrepresented in your sample.

Also note whether the people who were left out might behave differently than the people in your convenience sample. An interview on the street corner, for example, would exclude non-ambulatory patients. If your outcome measures are not strongly related to this factor, you might be okay. An assessment of eye color is probably safe in this setting. But a street corner interview would be a disaster if you were measuring something like the degree of disability.

You also have to be careful to qualify your findings appropriately. Findings from a convenience sample would be considered less definitive and would usually require replication in a more controlled setting. You can also qualify your results by extrapolating them only to a much more targeted and narrowly defined population.

Further reading

1. The Research Sample, Part I: Sampling. Lunsford TK, Lunsford BR (1995) Journal of Prosthetics and Orthotics, Volume 7, Number 3, pp. 105-112. http://www.oandp.org/jpo/73/73105.htm
2. Nonprobability Sampling. Trochim WMK. Accessed on 2004-01-26. trochim.human.cornell.edu/kb/sampnon.htm

This work is licensed under a Creative Commons Attribution 3.0 United States License. It was written by Steve Simon on 2002-10-15, edited by Steve Simon, and was last modified on 2010-04-01. This page needs minor revisions. Category: Definitions, Category: Research designs.