Page 1 :
—tt—“—O, , 226 8, POPULATION, SAMPLE, AND SAMPLING, up. In order to be able to observe one of the members ag m, ~ 2 Our, , disease may have a support group, initial subject then leads to more subjects for the study., , li owball Sampling, LINEAR SNOWE AMPLING, In this, each selected sample is as, a linear chain is created by the comp, , FIGURE 8.4 Linear snowball samplin, , ked to provide reference of only one similar subject; why,, letion of desired sample (Fig, 8.4). R, , , , EXPO! NONDISCRIMINATIVE SNO SAMPLING, , In this, each sample member is asked to provide reference of at least two similar subjects, , because of which the size of the sample grows exponentially and a large sample size can be, , achieved (Fig. 8.5)., (O) (O), (0), 6 (©), ©), , FIGURE 8.5 Exponential nondiscriminative snowball sampling, “, , , , , OWBA AMPLIN, , , , , .L_ DISCREMH BA, ily one sample is selected and asked for two references of similar subjects, out °!, r one subject must be active to provide further references and another could be nor, teferences. Similarly, each active reference subject is further asked for tw°, ‘subjects; out of them one should be active for further references (Fig. 8.6)