The Limited Usefulness of Convenience Sampling

Convenience sampling is a type of sampling by which the researcher selects a study environment and/or study participants primarily based on ease of access, availability, and/or familiarity. Convenience Convenience sampling radiussampling is not uncommon in qualitative research when researchers may need to complete their research in a short time frame and at a relatively low cost. For example, an ethnographer who wants to study how people behave in a confined space might design her research to observe people on her daily commute on the local subway. Or a graduate student might select clergy within a narrow radius of his university to conduct in-depth interviews to understand the roles clergy play in the lives of their congregations. Or focus group discussions might be conducted at a geriatric facility where the researcher visits her parents in order to learn about skilled nursing care.

In each case, the researcher may come away with insightful information about people in confined spaces on that particular subway car on a particular day, or clergy roles among the particular clergy drawn from religious groups within the neighborhood, or skilled nursing care at that particular geriatric facility. However, the important limitation of these studies lies in the fact that the subway car, the religious groups, and the geriatric facility were not selected because they were somehow representative of confined spaces, religious organizations, or senior medical care facilities, but rather because these locations and participants were in easy access and familiar to the researchers. As a result, and without other research to help triangulate the data, the researcher (and users of the research) have no way of knowing how (or if) the particular subway car on the particular day and time of day, or the clergy in the neighborhood, or the geriatric facility where the researcher’s parents live relate to (i.e., is the same or different than) the broader context of confined spaces, religious organizations, or geriatric facilities.

This raises an important limitation to convenience sampling. From a quality standpoint, convenience sampling limits the ultimate usefulness of a qualitative study because the data based on a convenience sample do not allow the researcher (and users of the research) to apply the findings to other contexts, i.e., convenience sampling limits the transferability of the research. Transferability is a vital aspect of the Total Quality Framework Transparency component and is fundamental to contributing something of value. And in the end, contributing something of value – that is, maximizing the usefulness of the research – is the researcher’s ultimate goal.



Image captured from:


  1. It’s absolutely true that convenience samples do not create data that can be generalized to broader populations. However, there is much good to a convenience sample. They are a great starting point to brainstorm ideas and develop hypotheses. From convenience come ideas that can then be tested with more representative samples in more represented locations.

    The problem with convenience samples is not that they are unrepresentative and ungeneralizable but rather that they are misused and abused. We need to use convenience samples properly, to save time and money, before embarking on larger scale, more formal testing.


    1. Thank you, Carol, for your comment. I think you and I agree. Problems arise when the use of convenience samples is not properly acknowledged and the limitations of the learning is under appreciated.


  2. Margaret, The topics you always present/discuss are very important, meaningful, and relevant. If people dedicated themselves to understanding and applying all the things you shine a light on, in using (the Total Quality Framework), the scientific and business approaches and what we could achieve, would become much better. Many in the research and business worlds do not think systematically and quite often use data (quantitative and qualitative) to achieve political ends and don’t spend a lot of time on the means (methodology, ethics, quality). So often, I have seen people break most/all of the Total Quality Framework rules and they do so to justify some political/personal ends. And it usually ends badly (although many times, those ‘chickens don’t come home to roost’ in the short run) – so, it becomes a game of not getting caught, continually lie and obfuscate, or when caught, deny everything/ blame others. There is a great resistence to our collective efforts to get people to think and act more honestly and systematically, with integrity and ethics – but this is a good fight that is worth the effort (IMHO). In a great many respects, we are analogously living in a modern ‘dark ages’ – and we are selling ‘enlightenment’ via the application of systematic and rigorous science. We are hoping to tip the population scales with people that are well meaning, but simply uninformed. Our efforts will be a bit more problematic with those that are not well meaning (lets call them the Machivellians), who don’t really care for the greater good or doing things with a high degree of quality or doing things ethically/honestly (following a Total Quality Framework) – because the world is a zero-sum game to them – and their only goal is to ‘win’ (what is good for them). I certainly have worked with several people like this (and they are poison). They actively fight against the implementation of any part of the Total Quality Framework. I also think you will not have to dwell too long on recognizing many of these actors in the world today.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.