Purposive Sampling

A non-probability sampling strategy in which participants are selected because they meet criteria specifically relevant to the research question — proficiency band, learner profile, classroom role, exam-history, or a defined level of expertise. The selection logic is theoretical rather than statistical, common in qualitative and mixed-methods research where rich, focused data on particular cases matter more than generalisation to a population.

Variants

Patton's typology, taken up in applied-linguistics methods texts, distinguishes several purposive strategies:

• Maximum variation: selecting participants spanning the full range of a target characteristic (beginner through advanced, multiple L1s) to surface common patterns despite diversity.
• Typical case: selecting participants who represent the average member of a population, useful for portraying the unremarkable middle.
• Extreme or deviant case: selecting outliers (very high achievers, very low achievers, fossilised learners) to study what makes the unusual unusual.
• Critical case: selecting cases where, if the phenomenon does not occur, it is unlikely to occur anywhere else.
• Snowball / chain: recruiting through participant referrals, useful for hard-to-reach populations.
• Theoretical sampling: iterative selection driven by emerging analytic categories, central to grounded theory.

Use in Applied Linguistics

Purposive sampling is the default in qualitative SLA — case studies of individual learners, classroom ethnographies, teacher-cognition research — and in mixed-methods studies where a quantitative phase identifies cases for deeper qualitative follow-up. Dörnyei (2007) treats purposive selection as the qualitative analogue of stratified random sampling: both ensure coverage of theoretically important variation, but purposive does so through researcher judgement rather than probability.

Reporting

A purposive sample requires explicit selection criteria, the rationale linking those criteria to the research question, and full descriptive information for each case. Without that documentation, transferability — the qualitative counterpart to external validity — cannot be assessed by readers.

References

• Dörnyei, Z. (2007). Research Methods in Applied Linguistics: Quantitative, Qualitative, and Mixed Methodologies. Oxford: Oxford University Press.
• Patton, M. Q. (2015). Qualitative Research and Evaluation Methods (4th ed.). Thousand Oaks, CA: Sage.
• Mackey, A., & Gass, S. M. (2016). Second Language Research: Methodology and Design (2nd ed.). New York: Routledge.