Answer to Question 1
In both cluster sampling and stratified sampling, the population is first divided into subgroups. However, the similarity is rather illusory. In stratified random sampling, a sample is taken from every stratum of the population in an attempt to ensure that important segments of the population are given due weight. By contrast, in cluster sampling a random sample of clusters is taken, so that some clusters will have no members in the sample. Since, within clusters, population members will probably be fairly homogeneous, the danger is that important subgroups of the population may be either not represented at all or grossly underrepresented in the final sample. In consequence, while the great advantage of cluster sampling lies in its convenience, this convenience may well be at the cost of additional imprecision in the sample estimates. A further distinction between cluster sampling and stratified sampling is that in the former a complete census of chosen cluster members is taken, while in the latter a random sample of stratum members is drawn. This difference, however, is not essential. Indeed, on occasions an investigator may draw a random sample of chosen cluster members rather than take a full census.
Answer to Question 2
A
Kente prestige cloth (detail), Ghana, Ewe peoples.
The Annunciation (The Mérode Altarpiece) (detail).
Imperial Procession (detail), south frieze, Ara Pacis Augustae, Rome.
A sampling of the modified and unusual bases found in tRNAs