What is sampling in the context of social research? Discuss different forms of sampling with their relative advantages and disadvantages. (Answer in 350 words)

- What is sampling in the context of social research? Discuss different forms of sampling with their relative advantages and disadvantages. (Answer in 350 words)
Introduction
Sampling is the claim that a part, properly drawn, can speak for the whole.

Body

1. What sampling is and why research uses it: A sampling frame lists the units of a population, from which a subset is drawn; the inference back is valid in proportion to how the drawing was done. Research uses it because complete enumeration is slow and costly — the Census repeats only once a decade — and because a well-drawn sample of a few thousand can describe millions, often more accurately, since fewer cases can be better supervised.

2. The different forms: Probability sampling gives every unit a known, non-zero chance of selection and so allows sampling error to be calculated. Simple random sampling draws by lot; systematic sampling takes every kth unit from a list; stratified sampling divides the population into strata — rural and urban, caste, income — and samples within each; cluster and multi-stage sampling select villages or urban blocks first and households within them, which is how the National Sample Survey is built. Non-probability sampling abandons that calculus: quota sampling fills preset category totals at the interviewer's discretion; purposive sampling picks informative cases; convenience sampling takes whoever is available; snowball sampling grows from one respondent's contacts, the standard route into hidden populations; and theoretical sampling, in Glaser and Strauss, chooses the next case to develop an emerging category.

3. Their advantages and disadvantages, set against each other: Simple random sampling is unbiased but needs a complete frame; stratified sampling avoids that and guarantees small groups appear, and is more precise for the same sample size when strata are internally homogeneous, but demands prior knowledge of the stratifying variable. Cluster sampling is the cheapest to field and the least precise, since neighbours resemble one another. Systematic sampling is simple and fails if the list has a periodicity. Quota sampling is fast and carries hidden interviewer bias; the collapse of the Literary Digest poll in 1936 showed what a skewed frame and non-response do even to an enormous sample. Snowball sampling reaches the unreachable at the cost of over-representing the well connected.

4. What decides which form suits a study: The question. Generalisable estimates need probability sampling and a usable frame; depth, rare populations and emerging categories need purposive or theoretical selection. Budget, frame quality and the heterogeneity of the population then settle the design.

Conclusion
The size of a sample matters far less than the procedure by which it was drawn.