Random Sampling

A sampling procedure in which every member of a defined target population has a known, non-zero probability of selection. Random sampling is the foundation of inferential statistics: standard errors, confidence intervals, and p-values are all derived under the assumption that the sample was drawn from the population by a probability mechanism.

Variants

Simple random sampling gives every population member an equal probability of selection; in practice it requires a complete sampling frame (a numbered list of every unit) and is implemented through random-number generation. Systematic random sampling draws every k-th unit from an ordered frame after a random start, equivalent to simple random when the ordering is unrelated to the outcome. Cluster random sampling selects intact groups (schools, classes, centres) at random and then includes all members of selected clusters; it is administratively cheaper but reduces effective sample size because units within a cluster are correlated. Stratified random sampling divides the population first, then samples randomly within each stratum.

Distinction from Random Assignment

Random sampling concerns how participants are recruited into the study; random assignment concerns how recruited participants are allocated to conditions. The two are independent. A study can have random assignment without random sampling — the typical quasi-experiment in classroom SLA — supporting causal claims about treatment effects but not population-level generalisations. Random sampling without random assignment supports descriptive population estimates but not causal inference.

Status in SLA Research

Genuine random sampling from a defined L2 learner population is rare in applied linguistics. Mackey and Gass (2016) and Dörnyei (2007) both note that the discipline relies overwhelmingly on convenience samples, with random procedures applied at the assignment-to-condition stage rather than the recruitment stage. The implication is that inferential statistics in SLA studies should be read as conditional on the participants studied, with Replication and meta-analysis carrying the load of population-level inference.

References

• Dörnyei, Z. (2007). Research Methods in Applied Linguistics: Quantitative, Qualitative, and Mixed Methodologies. Oxford: Oxford University Press.
• Mackey, A., & Gass, S. M. (2016). Second Language Research: Methodology and Design (2nd ed.). New York: Routledge.
• Cohen, L., Manion, L., & Morrison, K. (2018). Research Methods in Education (8th ed.). London: Routledge.