Following field testing, three quota samples
of 90 were used for each of the three roles.
In the Wilson & Cairns (in press) study a quota sample
of residents from Enniskillen (who responded anonymously) were compared, some 8 months after the bomb-blast, to residents of a nearby less-violent town (Lowtown) in percepts of violence, coping behaviour and levels of psychological disorder.
This presumably influenced the choice of a research design consisting of a non-random quota sample
and employing semi structured interviews with 168 men and women who had quit high school in Edmonton during 1984-85.
The sampling error for a quota sample
is assumed to be about twice as large as that of a probability sample.