Skip to main content

Sampling Methods — Previous Year Questions

📊 Topic Overview​

MetricValue
Papers Appeared12 of 28 (42.9%)
Priority🔴 HIGH
Section DistributionSec A: 6

🎯 Must-Know Concepts​

  1. Definition of sampling, population, and sampling frame
  2. Probability sampling: simple random, systematic, stratified, cluster, multistage
  3. Non-probability sampling: convenience, purposive, snowball, quota
  4. Advantages and limitations of each type
  5. Factors affecting sample size

Key Question: Types of Sampling Methods (Probability and Non-Probability)​

Answer (Section A — ~450 words):​

Sampling is the process of selecting a subset of individuals (sample) from a larger population to make inferences about the population. Because studying entire populations is usually impractical, samples are used as representatives. The quality of a sample — how well it represents the population — determines the validity of generalisations.

Key Concepts: Population: The complete set of individuals, events, or objects sharing defined characteristics. Sampling frame: A list of all accessible members of the population from which a sample is drawn. Sample: The subset selected for study. Representativeness: The degree to which a sample mirrors the population on relevant characteristics.

Probability Sampling Methods: In probability sampling, every member of the population has a known, non-zero probability of selection. This allows statistical inference and error estimation.

1. Simple Random Sampling (SRS): Every member has an equal probability of selection. Achieved via random number tables or computer-generated randomisation. Requires a complete sampling frame. Unbiased but may miss rare subgroups.

2. Systematic Sampling: Select every k-th element from the sampling frame after a random start. Efficient; approximates SRS if the list has no periodic pattern. Risk: if the list has a cyclical pattern (e.g., alternating high/low scorers), systematic sampling is biased.

3. Stratified Random Sampling: Divide the population into mutually exclusive strata (e.g., gender, age groups) and randomly sample from each stratum. Ensures representation of all strata. Proportional (strata sampled in proportion to their size) or disproportional (rare strata oversampled). More precise than SRS for heterogeneous populations.

4. Cluster Sampling: Divide the population into naturally occurring clusters (e.g., schools, cities); randomly select clusters; study all or sample members within selected clusters. Economical for geographically dispersed populations. Less precise than SRS (cluster members tend to be similar to each other).

5. Multistage Sampling: Combine several sampling techniques in stages (e.g., select districts, then schools, then students). Used in large-scale national surveys.

Non-Probability Sampling Methods: Every member does not have a known probability of selection. Cannot calculate sampling error; generalisability is limited but feasibility is higher.

1. Convenience Sampling: Select whoever is readily available. Inexpensive and quick but highly prone to bias. 2. Purposive (Judgement) Sampling: Deliberately select cases that meet specific criteria relevant to the research question. Used in qualitative research (theoretical sampling). 3. Snowball Sampling: Initial participants recruit further participants from their networks. Used for hidden or hard-to-reach populations (drug users, undocumented migrants). 4. Quota Sampling: Set quotas for subgroups (similar to stratified) but fill them non-randomly. Useful when a sampling frame is unavailable.

🧠 Memory Aids​

Probability (SSSCMu): Simple random, Systematic, Stratified, Cluster, Multistage

Non-Probability (CPSQ): Convenience, Purposive, Snowball, Quota

📚 Related Content​

✅ Self-Assessment​

  1. What is the difference between stratified and cluster sampling?
  2. When is snowball sampling most appropriate?
  3. Why is systematic sampling potentially biased?
  4. What is the advantage of stratified over simple random sampling?
  5. What is a sampling frame and why is it important?

💡 Exam Tips​

  • Section A: Definition → Key concepts → Probability methods (5, each defined and evaluated) → Non-probability methods (4) → Factors in sample size.
  • Distinguish stratified (divide then randomly sample within strata) from cluster (randomly sample clusters, then study within) — commonly confused.
  • Non-probability methods are increasingly asked in Section B — prepare snowball and purposive in particular.