Chapter 29: Recruitment and sampling

What is sampling?

Sampling is defined as selecting a suitable group of people (population) for a study.¹ Inviting people to the study who have the information to address the research question is a key consideration in sampling. Sampling is guided by the research question and will also influence data collection.

In qualitative research, different forms of non-random sampling, also known as non-probability sampling (not everyone has the chance of being selected), are utilised.¹ Using non-random sampling means the likelihood of a potential research participant being selected is not known in advance. This form is limited in generalisability; however, it aligns with qualitative research principles of sampling for meaning rather than frequency. Qualitative researchers tend to say that qualitative research is not generalisable, but is representative.¹ While qualitative studies often include non-random sampling, simple random sampling can be conducted when it is important to select a random set of participants from a large population, in which everyone has the same chance of being selected. This could be done by randomly selecting names from a telephone list or voter registration roll.

There are many ways to select a sample (sampling techniques). Among these are:

Snowball sampling, whereby study participants recruit or refer people they know to the study. This method is commonly used when potential participants can be hard to find through other means but potential participants are likely known to each other. For example, drug users or patients with rare diseases are likely to know others like themselves. This method may pose challenges for privacy because people may not want to share their contacts.²

Convenience sampling, in which study participants are those most available to participate in the study. Participants may be those who are easily accessible to the researchers – such as a practitioner who is a member of a professional organisation and uses that organisation to recruit participants or patients at a hospital where the researcher works. This method can introduce bias because participants are drawn from within the researcher’s own networks or spheres of influence.²

Purposive sampling, also known as purposeful sampling or selective sampling, involves the selection of participants on the basis of their ability to provide in-depth and detailed information about the phenomenon under investigation. For example, a study on the experience of working in a public hospital as a frontline emergency nurse during the COVID-19 pandemic requires participants to be nurses, working in an emergency department, having worked during the pandemic and at a public hospital. A general practitioner, for example, could not provide in-depth information on the phenomenon being investigated. This method may present challenges in locating potential participants because it can be difficult to find participants who are able to provide in-depth information about the phenomenon being studied.²

Quota sampling is sometimes referred to as purposive sampling with more structure. Categories that are important to the study and for which there is likely to be some variation are identified and then subgroups are identified on the basis of each category. The researcher decides how many people to include from each subgroup and collects data from that number of participants in each subgroup. This method requires the investigator to have prior information about the sample. For example, in a study researching students and their experience of attending university, many subgroups need to be considered; for example, those living on or off campus, the course of study, faculty or discipline, age, gender identity, ethnicity and more. This method can present a challenge to fill quotas for each category identified.²

Snowball, convenience purposive and quota sampling are the most commonly used techniques for sampling in qualitative research. Other, less commonly used techniques include stratified sampling, theoretical sampling, extreme case sampling, typical case sampling, systemic sampling and intensity sampling. The technique used will depend on the research aim and questions.

Sample size in qualitative research

There are no clear guidelines for sample sizing in qualitative research. While researchers often propose a sample size, in general it is not decided on prior to data collection, but rather when data saturation occurs. Data saturation is a controversial concept because it is usually considered the point at which no new data is identified in interviews or focus groups. Some qualitative researchers, such as Braun and Clarke⁴ hold that data saturation can never be fully achieved because each participant will have something new to add to the data. For other researchers, data saturation is an acceptable concept, and is often given as 8–17 participants.^1,3 While 15 might be a proposed sample size for a study involving interviews, when the researcher has completed 10 interviews they may feel they have reached saturation as far as new themes or ideas are concerned. It is important to remember that sample size is not used to generalise and validate findings⁵, but rather to ensure in-depth understandings of the phenomenon under investigation.

Criteria for sampling

Another consideration in sampling is determining the inclusion and exclusion criteria for the study. This is a standard practice in qualitative research and is used to define who will and will not be able to participate in the study.⁶ For example, inclusion criteria might include gender identity, age and health diagnosis. People who do not meet the inclusion criteria would not be eligible to participate in the study. Exclusion criteria are more than simply the opposite of inclusion criteria; they cover specific factors such as not being able to provide informed consent, using a specific type of medication, having a certain diagnosis and more.⁶ Inclusion and exclusion criteria are broad, and should not be so narrow that they limit the characteristics of participants who can be recruited to the study.

Well-developed inclusion and exclusion criteria and well-considered sampling methods will assist researchers with the next step, recruitment to the study.

What is recruitment?

Broadly speaking, recruitment to a research study involves presenting potential participants with detailed information about the research to help them decide on whether to participate in the study. The information presented during recruitment contributes to the consent process. Researchers who initiate participant recruitment should have a basic knowledge about the research and be part of the study team.

Before any research is to commence, the study must have ethical approval from a human research ethics committee. For academic researchers, this is the university’s human research ethics committee. For researchers outside the academic setting, it is the organisation that has been identified as the review agency. For example, research conducted within a hospital setting is reviewed by the hospital’s research ethics committee or panel. Research ethics committees review applications against the benchmarks set out in the National Statement on Ethical Conduct in Human Research.

As part of their data collection activities, researchers need to consider how they will invite research participants (recruitment) and the process of consent (see Chapter 30).

Recruitment methods need to take account of whether participant contact information is available and if there is permission to use it for the purpose of the research. Having email, phone or in-person (e.g. at a clinic) information will enable the researchers to correspond directly with potential participants, to invite them to participate in the research. Most researchers do not have the contact information of potential participants or are not able to contact them for the purpose of the research. In this instance, more passive methods of recruitment are needed and need to take into account settings frequented by potential participants, in-person or online. Passive methods of recruitment include advertising on social media, posting flyers on clinic noticeboards to advertise the research, asking clinics, sporting clubs, social clubs, schools, professional organisations, patient groups and other agencies to distrubute printed newsletters or to send emails on behalf of the researcher. The author has been involved in many studies in which contact details were not available. In a hospital study examining staff perceptions of how well a program was implemented, the hospital sent out invitations to potential participants (hospital staff) on behalf of the researcher. Staff who wished to participate were able to contact the investigators independently. In other instances where no contact details were available, the author advertised for participants with the assistance of peak bodies and community networks.

Recruiting participants through social media is increasingly common, but researchers need to be mindful about privacy and public availability of information. For example, potential participants may believe their comments in response to a recruitment advertisement on a social media platform are private, when in fact the information they share is available or visible to all users on that platform, or to the public.

Problems that may be encountered during recruitment include (but are not limited to) participants who are not fluent in English, participants who are hard to find and participants who do not trust research. Researchers need to consider these challenges in their recruitment activities and adjust them as necessary. Adjustments may include providing documents translated into the preferred language(s) of potential participants, recruiting research staff who speak the language, being mindful of the gender identity of research staff (e.g. women participants from some cultural backgrounds may prefer to deal with women researchers) and ensuring the research purpose is clearly communicated.

Incentives for research participants

Sometimes recruitment can be enhanced by providing an incentive for participants. This approach must be approved by the human research ethics committee before being offered to participants. Examples of reasonable incentives include providing reimbursement for parking at a hospital, offering a gift/shopping card or a coffee voucher for a nearby café, in recognition of time spent participating in the study. Incentives should not be excessive and therefore potentially coercive. The National Health and Medical Research Council (NHMRC) provides guidance on incentives in research.

Examples of sampling and recruitment methods are presented in Table 29.1. Notice how often multiple recruitment methods are used.

Table 29.1: Examples of sampling and recruitment

Strengths and challenges

Each sampling and recruitment method has strengths and challenges. The one chosen depends on the study’s research question(s) and aim(s). Choosing the appropriate methods will bring rigour to the research, while choosing inappropriate methods will reduce rigour and affect the research results. Consider a study in which women’s experience of episiotomy is being sought.⁷ It may be possible to recruit many women based on how relatively easy the birth was, or only a few women based on how willing the women are to talk about the use of forceps during delivery. Or consider a descriptive qualitative study in which up to 40 participants are recruited to provide slightly more generalisable results about an experience with a health service.⁸ Each sampling and recruitment method is valid, depending on what is being researched.

Summary

Sampling refers to the selection of a suitable group of people from a broader population, to participate in a study. Selecting the people suitable for the research study is important because that will affect the study’s findings. There are many ways to sample, and these depend on the research being undertaken as well as the availability of participants. Recruitment refers to providing potential participants with information about the research and gaining their agreement to participate. There are many recruitment methods, and the one(s) chosen depend on the research being undertaken as well as participant agreement to become involved. Issues of privacy, confidentiality and consent need to be fully considered when sampling and recruiting participants to a research study.

References

1. Lopez V, Whitehead D. Sampling data and data collection in qualitative research. In Schneider Z, Whitehead D, LoBiondo-Wood G, Haber J. Nursing & Midwifery Research: Methods and Appraisal for Evidence-Based Practice (4th ed). Elsevier Mosby; 2013:123-140.
2. DeCarlo M, Cummings C, Agnelli K. Graduate Research Methods in Social Work: A Project-based Approach; 2020.  Accessed September 26, 2023. https://viva.pressbooks.pub/mswresearch/
3. Hennink M & Kaiser B. Sample sizes for saturation in qualitative research: a systematic review of empirical tests. Soc Sci Med. 2022;292:114523. doi.org/10.1016/j.socscimed.2021.114523
4. Braun V & Clarke V. To saturate or not to saturate? Questioning data saturation as a useful concept for thematic analysis and sample-size rationales. Qual Res Sport, Exerc Health. 2019;13(2):1-16. doi:10.1080/2159676X.2019.1704846
5. Vasileiou K, Barnett J, Thorpe S et al Characterising and justifying sample size sufficiency in interview-based studies: systematic analysis of qualitative health research over a 15-year period. BMC Med Res Methodol 2018;18:148. doi:10.1186/s12874-018-0594-7
6. ICORD, Vancouver Coastal Health Research Institute. Inclusion & exclusion criteria explained. Accessed May 28^th 2023. https://icord.org/research/studycriteria/
7. He S, Jiang H, Qian X et al Women’s experience of episiotomy: a qualitative study from China BMJ Open 2020;10:e033354. doi:10.1136/bmjopen-2019-033354
8. Williams M, Jordan A, Scott J et al Service users’ experiences of contacting NHS patient medicines helpline services: a qualitative study. BMJ Open 2020;10:e036326. doi:10.1136/bmjopen-2019-036326
9. Gabriel P, Kaczorowski J, Berry N. Recruitment of refugees for health research: a qualitative study to add refugees’ perspectives. Int J Environ Res Public Health. 2017;14:125. Doi:10.3390/ijerph14020125
10. Jola C, Sundström M, McLeod J. Benefits of dance for Parkinson’s: the music, the moves, and the company. PLoS ONE. 2022;17(11): e0265921. doi:10.1371/journal.pone.02659214
11. Gauche C, de Beer L & Lizelle Brink L. Managing employee well-being: a qualitative study exploring job and personal resources of at-risk employees. SA Journal of Human Resource Management/SA Tydskrif vir Menslikehulpbronbestuur. 2017;15:a957. doi:10.4102/sajhrm. v15i0.957
12. Hoernke K, Djellouli N, Andrews L et al Frontline healthcare workers’ experiences with personal protective equipment during the COVID-19 pandemic in the UK: a rapid qualitative appraisal. BMJ Open. 2021;11:e046199. doi: 10.1136/bmjopen-2020-046199