Data Collection Methods
Ever wondered how researchers gather information from large populations? They use either a census (measuring every member of a population) or sampling techniques. While a census gives accurate results, it's expensive and sometimes impractical.
Sampling involves selecting sampling units (individuals) from a sampling frame (complete list). Random sampling methods give everyone an equal chance of selection. Simple random sampling uses random number generators, while systematic sampling selects every kth unit after a random starting point. Both are bias-free but require a sampling frame.
Stratified sampling ensures your sample represents different groups (strata) within the population. This reflects population diversity but requires clear classification of strata. Non-random methods include quota sampling (similar to stratified but filled by researchers) and opportunity sampling (using whoever's available), which are easier but potentially biased.
Real-world application: Weather data sets often contain special notations like 'tr' (trace, calculated as 0.025mm), 'na' (unavailable data), cloud cover percentages (0-8), and wind speeds in knots 1km=1.15mph. Understanding these conventions is essential when analysing meteorological information!