“…Newer, network-based sampling methods also allow the researcher to sample from a hidden population (Heckathorn, 1997, Klovdahl, 1985, Klovdahl, 1989, Klovdahl, et al, 1994, Spreen, 1992. Two designs, the random walk (Klovdahl, 1989, Klovdahl, 1990, Liebow, et al, 1995 and respondent-driven sampling (Broadhead, et al, 1995, Heckathorn, 1997, take advantage of social networks within a population and aim to avoid some biases of earlier sampling methods. Appropriate analysis may allow the researcher to minimize biases associated with a given design in order to improve the estimate of population parameters (Gile and Handcock, 2011, Heckathorn, 2007, Thompson, 2011.…”