Step 1: The key feature described is dividing the population into meaningful, homogeneous subgroups before sampling.
Step 2: These subgroups are called strata (for example by age, sex, or income). When the population is split into strata and a random sample is drawn from within each stratum, the method is stratified random sampling.
Step 3: Compare the alternatives: simple random sampling draws directly from the whole population without subdivision; cluster sampling divides the population into clusters and then selects whole clusters (not units from every cluster); systematic sampling picks every k-th unit from a list.
Step 4: Because units are drawn randomly from each subgroup (stratum), this is stratified sampling. The correct answer is option 2 (Stratified sampling).