ABSTRACT
The subjects to be observed may be identified through subject sampling, subject assignment, or single-subject selection. Subject sampling is most appropriate for an ex post facto study. With ex post facto studies, the central concern is how well the subjects selected represent a larger population. In an experimental group design, subjects generally are divided into experimental and control groups. With a single-subject study, the questions of subject representativeness and equivalence of experimental and control group assignments are irrelevant because only one subject is involved. Subject sampling can be categorized into probability sampling and nonprobability sampling. There are three common probability sampling alternatives to simple random sampling such as stratified sampling, cluster sampling, and systematic sampling. The more commonly used nonprobability sampling methods include accidental sampling, purposive sampling, and quota sampling. The relationship among sample size, population heterogeneity, and sampling error just described exists only when the sample approximates a simple random sample.
