Random selection from the larger population is the gold standard for creating representative samples. It gives each member an equal chance of being chosen, minimizing selection bias. Volunteer samples are biased, coercion is unethical, and mail-order lists are not representative.