Cluster sampling is highly efficient when the population is geographically dispersed. By selecting clusters, researchers can reduce travel costs and logistical burdens associated with collecting data from widely separated units, even if the units within a cluster are heterogeneous. Note: The provided answer key suggests 'Far a part', which is a common justification for cluster sampling efficiency.

When a population is homogeneous, meaning the units are very similar to one another regarding the variables of interest, simple random sampling is highly effective. Since any individual unit is representative of the whole, a random selection process provides an unbiased and efficient way to gather data without the need for complex stratification or clustering.

Systematic sampling involves selecting a random starting point from the first k elements of a population and then selecting every kth element thereafter. Because the sampling interval is fixed, the selection of the first unit (the random start) dictates the entire sequence of subsequent selections. Therefore, the initial choice determines the composition of the entire sample set.

Random numbers are frequently called random digits, especially when referring to the individual components of a random number table. These digits are generated such that each digit (0-9) appears with equal frequency and independence, which is essential for unbiased sampling and simulation.

Probability sampling ensures that every member of the population has a chance of being selected. The four standard methods generally taught are simple random sampling, stratified random sampling, systematic sampling, and cluster sampling. These methods are widely used to ensure that the sample is representative of the population from which it is drawn.

Both simple random sampling and systematic sampling are types of probability sampling. In probability sampling, every member of the population has a known, non-zero chance of being selected. This allows researchers to make statistical inferences about the population based on the sample data, which is a key requirement for rigorous scientific research.

In probability sampling, every individual or unit within the target population has a non-zero, known probability of being selected for the sample. This feature is fundamental to the design, as it allows researchers to use statistical theory to make valid inferences about the population and estimate the margin of error.

In stratified sampling, the population is divided into subgroups called strata. A critical requirement is that these strata must be mutually exclusive, meaning they are non-overlapping. This ensures that every member of the population belongs to exactly one stratum, which allows for precise estimation and reduces sampling error compared to simple random sampling.

Stratified sampling is ideal for heterogeneous populations because it involves dividing the population into distinct, homogeneous subgroups or 'strata'. By sampling from each stratum, the researcher ensures that all diverse segments of the population are adequately represented in the final sample, which significantly improves the precision of the estimates compared to simple random sampling.

Sampling with replacement means that after a unit is selected and observed, it is returned to the population before the next selection. Consequently, the same unit has a non-zero probability of being selected again in subsequent draws, allowing for multiple selections of the same unit.