Participant Sampling in Dissertations: UK Guide (53 characters)

Written By

Edward Fletcher

✔️ 97% Satisfaction | ⏰ 97% On Time | ⚡ 8+ Hour Delivery

Sampling, deciding who participates in your research, is where many dissertations fall apart. Students either treat it as something simple (just ask whoever's available) or vastly overcomplicate it (agonising for weeks about the perfect sample size). It's neither. Good sampling is about being deliberate and then being honest about what you can and cannot claim.

Why Sampling Actually Matters

Your sample is the group of people or things you study. You cannot study everyone (that's called a census and is rarely possible in dissertations). So you study a sample and make claims based on that sample. The quality of your claims depends entirely on whether your sample is appropriate for your research question.

Get sampling right, and your findings are credible. Get it wrong, and no amount of brilliant analysis will rescue your dissertation. This matters more than most students realise.

The Two Main Sampling Approaches

Sampling divides into two broad approaches: random sampling and purposive sampling.

Random sampling means selecting participants entirely by chance. Every member of your population has an equal chance of being selected. Think of drawing names from a hat. Random sampling works when you want to represent a specific population fairly. If you want to claim something about "UK university students," random sampling helps you do that credibly.

Purposive sampling means selecting participants deliberately because they fit your research criteria. You choose people specifically relevant to your question. If you're researching how mature students experience university, you'd purposively select mature students. You wouldn't randomly sample across all students, most wouldn't be mature.

Here's the key distinction. Random sampling helps you represent a population. Purposive sampling helps you understand a specific phenomenon. Different research questions require different approaches.

Random Sampling Methods

Simple random sampling is the most straightforward. Every person in your population has an equal chance of selection. List all members (say, all students in your university), then randomly select. This is genuinely random, but it's tedious for large populations.

Systematic sampling is random sampling made practical. Instead of drawing names randomly, you select every nth person from a list. For example, every fifth person. It's not truly random (there's a pattern), but it's close and much more practical.

Stratified random sampling divides your population into groups (strata) and then randomly samples from each group. If you're researching UK university students, you might stratify by region (Southeast, Midlands, North, Scotland) and then randomly sample from each region. This ensures you get representation across important subgroups.

Cluster sampling divides your population into clusters (maybe different universities) and randomly selects clusters, then studies everyone or a random sample within those clusters. It's practical when your population is geographically spread.

The advantage of all random methods: they're defensible. You can claim fairly that your sample represents your population.

The disadvantage: they're time-consuming and require access to your full population. You need a list of everyone. That's rare in dissertation research.

Purposive Sampling Methods

Purposive sampling is far more common in student dissertations because it's practical and works well for qualitative research.

Criterion-based sampling means selecting people who meet specific criteria relevant to your research. If you're studying student dropout, you study students who've dropped out. You're not trying to represent all students. You're trying to understand the specific phenomenon.

Snowball sampling starts with one or two participants who then recommend others. You ask your first interviewee "do you know anyone else with this experience?" and they suggest someone. It's tremendously useful when you're researching hidden or vulnerable populations, perhaps students with eating disorders or undocumented immigrants. It builds trust because people recommend friends.

Maximum variation sampling means deliberately choosing people who differ from each other. If you're researching mature students' university experiences, you might include mature students who're successful and struggling, mature students of different ages, different subject disciplines, different entry routes. The variation helps you understand the full range of experience.

Typical case sampling means selecting examples that are representative of the phenomenon. If you're researching "typical" student experience, you select participants who represent typical students, not extreme cases.

Expert sampling means selecting people with specific expertise. If you're researching how teachers teach reading, you interview experienced reading specialists. You're not trying to represent all teachers. You're trying to access expertise.

Deciding Your Sample Size

Here's where students get confused. Qualitative and quantitative research expect different sample sizes.

For quantitative research, larger samples are generally better because they reduce random error. The specific size depends on your research design, but typically you're looking at minimum 30 or more for statistical analysis. Some designs need hundreds. Check with your supervisor.

For qualitative research, sample size is different. You're not trying to represent a population statistically. You're trying to understand experience deeply. Quality matters more than quantity. Most qualitative dissertations include 12 to 20 interviews, sometimes fewer. That's sufficient if those participants provide rich data and you've chosen them deliberately.

The key concept is saturation. You continue recruiting participants until you're not learning new information. When subsequent interviews are repeating what you've already learned, you've reached saturation and can stop. That might be 12 participants or 20. You don't know until you're there.

Common Sampling Mistakes

First mistake: convenience sampling, studying whoever's easiest to access. "I studied my classmates because I could easily ask them." That's weak unless your classmates actually fit your research criteria. Explain why they're appropriate, not just why they're convenient.

Second mistake: collecting more data than you need. You don't need 100 survey responses if 50 give you solid data. You don't need 30 interviews if 15 have reached saturation. More isn't always better. Appropriate is better.

Third mistake: not acknowledging sampling limitations. If you studied only students from one university, acknowledge that. "Findings reflect the specific context of [University name], and may not transfer directly to other institutions with different resources and cultures." That's honest and credible.

Fourth mistake: not being clear about who you included and excluded. "I recruited students" is vague. "I recruited first-year undergraduate students in psychology at [University], excluding students under 18 or those with diagnosed mental health conditions at study start" is clear. Readers need to know exactly who your sample is.

Fifth mistake: claiming your sample represents something it doesn't. If you interviewed 12 mature students, you cannot claim findings about "UK students" broadly. You can claim findings about "mature students' experiences" in that specific context. Be precise about what your sample represents.

How to Write About Your Sampling Method

Your methods chapter should explain: who you wanted to study (your population), how you selected them (your sampling method), how many you recruited (your sample size), and why this approach was appropriate for your research question.

Example: "This research employed purposive criterion-based sampling to recruit experienced secondary mathematics teachers. Criteria for inclusion were: a minimum of five years teaching experience, current employment in a UK secondary school, and willingness to discuss teaching practise in depth. Participants were recruited through three secondary schools in the [region], with participants ranging from 6 to 22 years experience. Twelve teachers participated, representing a range of school contexts (selective and thorough) and subject specialism (pure mathematics and applied mathematics). This sampling approach was appropriate because the research question required accessing experienced practitioners' reflections on their practice. A random sample of all teachers would have included many early-career teachers whose experiences differ substantially and whose inclusion would have diluted insights into experienced practice."

Notice that this explains: the sampling method (purposive criterion-based), the criteria (five years experience, employed, willing), how you recruited (through schools), how many (twelve), their characteristics (range of experience and context), and why this was appropriate. That's thorough.

Three FAQs

Q: If I'm doing qualitative research, how many participants do I actually need? There's no magic number. It depends on your question, your data richness, and when you reach saturation. Most qualitative dissertations work with 12 to 20 participants. But if your participants give rich, detailed data and you quickly reach saturation, fewer might suffice. If your question is complex and your data sparse, you might need more. Discuss sample size with your supervisor in relation to your specific question.

Q: Should I randomly sample or purposively sample? It depends on your research question. Quantitative research investigating whether something is prevalent in a population requires random or stratified random sampling. Qualitative research investigating how something is experienced requires purposive sampling. Mixed-methods studies might use random sampling for quantitative data and purposive for qualitative. Match your sampling to your research question.

Q: What if I can't access as many participants as I'd like? Be honest. "I initially aimed to recruit 20 participants but managed to recruit 12 due to recruitment challenges. This limitation means findings are based on a smaller sample than intended. Saturation was reached with twelve participants, suggesting the sample was adequate despite falling short of initial targets." Honesty about limitations is far stronger than pretending you got the sample you planned.

---

Related posts: How to Design Your Research, Conducting Interviews for Your Dissertation, Qualitative Research Methods

---

The bibliography at the end of your dissertation is more than a formal requirement; it is a reflection of the breadth and quality of your reading and an indication of your engagement with the scholarly literature in your field. A weak bibliography that includes only a small number of sources, or that relies heavily on textbooks and websites rather than peer-reviewed academic journals and primary research, will leave your marker with concerns about the depth of your research. As a general guideline, your bibliography should include a mix of foundational texts that have shaped thinking in your field and more recent publications that demonstrate your awareness of current developments and debates in the literature. Managing your references using a software tool such as Zotero, Mendeley, or EndNote will save you a great deal of time and reduce the risk of errors in your final reference list, allowing you to focus your energy on the quality of your writing.