Chapter 5: Selecting Research Participants

The chapter establishes essential terminology by distinguishing between the target population—the entire group researchers wish to understand—and the accessible population—the subset realistically available for study. A representative sample accurately reflects the characteristics of the broader population, while a biased sample systematically excludes or overrepresents certain groups, threatening the validity of conclusions. The chapter then contrasts two broad sampling frameworks. Probability sampling methods assign each population member a known, calculable likelihood of selection; these include simple random sampling where selection occurs through unbiased processes, systematic sampling using fixed intervals from a list, stratified random sampling that divides the population into meaningful subgroups before random selection, and cluster sampling that randomly selects naturally occurring groups. Nonprobability sampling methods lack known selection probabilities and include convenience sampling, which relies on easily accessible participants, and quota sampling, which attempts to match population proportions through deliberate selection. The chapter discusses how each method influences sampling error—the difference between sample and population statistics—and selection bias, the systematic misrepresentation arising from the sampling process itself. Practical comparisons reveal that while probability methods generally provide stronger grounds for generalization, nonprobability approaches remain common in applied settings where resources or access constraints exist. The chapter concludes by presenting strategies to minimize bias when ideal sampling conditions are unavailable, helping researchers make informed decisions about participant selection that balance methodological rigor with practical feasibility.