Survey evidence is often introduced in court cases involving trademark violation and
employment discrimination. there has been controversy, however, about whether
nonprobability samples are acceptable as evidence in litigation. jacoby and handlin
(1991) selected 26 from a list of 1285 scholarly journals in the social and behavioral
sciences. they examined all articles published during 1988 for the selected journals,
and recorded (1) the number of articles in the journal that described empirical research
from a survey (they excluded articles in which the authors analyzed survey data which
had been collected by someone else) and (2) the total number of articles for each
journal which used probability sampling, nonprobability sampling, or for which the
sampling method could not be determined. the data are in file journal. dat.
a explain why this is a cluster sample.
b estimate the proportion of articles in the 1285 journals that use nonprobability
sampling, and give the standard error of your estimate.
c the authors conclude that, because "an overwhelming proportion of . .
recognized scholarly and practitioner experts rely on non-probability sampling
designs," courts "should have no problem admitting otherwise well-conducted
non-probability surveys and according them due weight" (p. 175). comment on
this statement.