03 · What You Need to Know
How to Determine Whether Your Required Sample Is Recruitable
Start with the sample your research question actually requires
Recruitment feasibility cannot be judged in isolation from the research question and study design. Before asking where you will find participants, determine who actually qualifies and approximately how many participants the study needs.
For quantitative studies, the required sample may be informed by the planned analysis, expected effect, desired precision, statistical power, design, and anticipated missing data or attrition. Qualitative studies usually involve a different logic. The number and composition of participants should reflect the research question, methodological approach, sampling strategy, and the depth and diversity of information required.
The important point is that recruitment planning begins with a defensible sampling requirement rather than an arbitrary target such as "I think I can get 100 respondents."
Once that requirement is reasonably clear, you can ask whether the population and recruitment environment can realistically supply it.
The target population is not the recruitment pool
Suppose there are 8,000 teachers in the geographical area relevant to your research. It may be tempting to conclude that recruiting 300 participants should be easy. But perhaps only 1,200 teach the grade level you need, 500 use the technology being studied, 220 work in institutions you can approach, and only some of those teachers will receive the invitation and choose to participate.
The headline population figure of 8,000 therefore says surprisingly little about recruitment feasibility.
You need to distinguish between several progressively smaller groups.
Target population The broader population to which your research question refers.
Potentially eligible population People who appear to satisfy the study's inclusion and exclusion criteria.
Accessible pool Eligible or potentially eligible people you can legitimately identify and approach through your available sites, organizations, records, communities, platforms, or other recruitment channels.
Approached participants People who actually receive or encounter the recruitment invitation.
Enrolled participants Eligible people who provide the required consent and enter the study.
Retained participants Those who remain sufficiently engaged to provide the data required for the intended analysis.
The distinctions matter because losses can occur at every stage. Recruitment feasibility therefore depends less on how many people theoretically belong to your population than on how much of that population is genuinely accessible to your study.
Eligibility criteria can shrink the pool quickly
Every inclusion criterion filters the population. Sometimes those restrictions are essential because the research question genuinely concerns a specific group. Problems arise when researchers combine several restrictive criteria without checking how many people satisfy all of them simultaneously.
Imagine that you want to study teachers who teach mathematics, work at private secondary schools, have at least five years of teaching experience, regularly use generative AI, and have implemented it in classroom assessment. There may be thousands of teachers in the region, but the intersection of those characteristics could be much smaller.
Before finalizing the design, estimate the number of people who satisfy the complete eligibility criteria rather than estimating each characteristic separately.
If the eligible population becomes extremely small, the recruitment problem deserves separate consideration because rare or hard-to-reach populations may require different sampling, recruitment, site, and design decisions.
Accessibility depends on a route to the participants
Knowing who the participants are is different from having a legitimate mechanism for contacting them.
A university may have 20,000 students, but that does not mean a researcher can obtain their email addresses. A hospital may treat hundreds of eligible patients, but privacy requirements may prevent researchers from directly identifying or approaching them. Employees may be accessible only through organizational permission. Members of a professional association may be reachable only if the association agrees to distribute the invitation.
Ask specifically how potential participants will encounter your study. Will a cooperating institution distribute the invitation? Will authorized staff identify potentially eligible individuals? Will you recruit through clinics, classes, community organizations, professional networks, advertisements, online platforms, registries, or another ethically and methodologically appropriate channel?
If your answer is simply "I will find them somehow," recruitment has not yet been demonstrated to be feasible.
Gatekeepers can determine whether recruitment happens at all
Many recruitment plans depend on someone else's cooperation. Schools, hospitals, companies, government agencies, community organizations, professional associations, platform administrators, and dataset custodians may control access to potential participants.
A supervisor's belief that an organization will probably cooperate is not the same as organizational permission. Neither is an informal conversation with one employee necessarily sufficient when authorization must come from another office.
If recruitment depends critically on one or two external organizations, consider whether you should contact potential research sites before finalizing the study. What can appropriately be confirmed at an early stage will vary according to the setting and applicable ethics or institutional procedures, but a critical dependency should at least be identified rather than ignored.
Being eligible does not mean someone will participate
Even when you can identify and contact enough eligible people, only a proportion may agree to participate. There is no universal recruitment rate that you can safely assume across studies.
Participation can be influenced by the population, research topic, recruitment approach, study burden, perceived relevance, trust, incentives, scheduling requirements, privacy concerns, length of participation, and the relationship between researchers and the community. The National Institute of Mental Health accordingly recommends that investigators consider recruitment and retention strategies in relation to the particular participants, institutions, and communities involved rather than relying on a generic recruitment formula.
Historical information from comparable studies or from the proposed recruitment site can therefore be valuable. If a research unit knows that previous studies involving a similar population approached 500 eligible people to enroll approximately 100, that evidence is more useful for planning than assuming that most eligible individuals will agree.
Work backward from the number of completed participants you need
A useful way to think about recruitment is to work backward from the number of participants whose data you ultimately need.
If your analysis requires 200 usable cases, recruiting exactly 200 people may leave no room for ineligibility discovered during screening, nonresponse after initial contact, withdrawal, incomplete questionnaires, missed follow-ups, or unusable data.
You can therefore construct a recruitment funnel using plausible assumptions for your particular setting. These are planning estimates, not universal rates.
This calculation does not prove that 189 participants will be obtained. The percentages are uncertain and should be informed by the best evidence available from your setting, comparable studies, preliminary enquiries, or pilot work. Its value is diagnostic: it makes your assumptions visible.
If you require 300 completed participants but your most plausible recruitment pathway yields approximately 189, you have identified a feasibility problem before spending months discovering it empirically.
Recruitment capacity is also a rate, not just a total
Suppose a clinic can provide access to enough eligible participants eventually, but historically only ten suitable participants become available each month. If you need 200 and have four months for recruitment, the total population may be adequate while the recruitment rate is not.
This is why recruitment should be connected to time. Estimate how many eligible people are likely to become available during the actual recruitment window, how quickly they can be screened and enrolled, and whether recruitment activities can operate continuously during that period.
Recruitment also has start-up costs in time. Site approval, ethics review, staff training, preparation of recruitment materials, scheduling, and coordination may occur before the first participant is enrolled. The NIMH specifically recommends incorporating preparatory requirements and recruitment and retention considerations into study timelines rather than treating enrollment as an isolated activity.
Once the likely rate is understood, you can more realistically determine how much time recruitment and data collection may require.
Recruitment burden matters
A technically reachable population may still be difficult to recruit if participation demands too much from potential participants.
Compare a five-minute anonymous questionnaire with a study requiring four laboratory visits, repeated blood collection, weekly diaries, or twelve months of follow-up. The eligible population could be identical, yet willingness to participate and remain involved may differ considerably.
Examine what participation looks like from the participant's perspective. Consider travel, scheduling, physical or psychological burden, privacy concerns, repeated contact, technological requirements, compensation where appropriate, and the clarity of what participants are being asked to do.
Reducing unnecessary burden can improve feasibility, but convenience should not override scientific or ethical requirements. If the measurements necessary to answer the question are inherently demanding, that burden belongs in the recruitment assessment.
Recruitment and retention are connected
Recruiting the required number is not always enough. In longitudinal, intervention, repeated-measures, or multi-stage studies, participants may need to remain involved for weeks, months, or years.
The relevant feasibility question may therefore be not "Can I recruit 150 people?" but "Can I recruit and retain enough people to obtain the endpoint data my analysis requires?"
Retention planning may involve clear communication about study expectations, appropriate scheduling, reliable follow-up procedures, and other strategies suited to the study population. NIMH guidance emphasizes planning for both recruitment and retention because enrollment alone does not guarantee that the intended data will ultimately be available.
Recruitment feasibility can sometimes be tested empirically
When major uncertainty remains, preliminary or feasibility work may help. This does not mean conducting a casual miniature version of the study and assuming that everything will scale perfectly.
In formal pilot and feasibility research, recruitment itself can be an explicit feasibility objective. The CONSORT extension for randomized pilot and feasibility trials, for example, identifies eligibility, recruitment, refusal, and follow-up rates as quantities that may be estimated to inform a future definitive trial. It also emphasizes reporting how participants were identified and approached.
Such work can help answer practical questions: How many potentially eligible participants appear each month? What proportion passes screening? How many consent? Which recruitment channels work? Where does the process fail?
Watch Out
A large target population does not guarantee an adequate sample. If you cannot identify, legitimately approach, enroll, and where necessary retain enough eligible participants within your study period, the population is not effectively recruitable for that particular design.
