The survey used a six point Likert-type scale that allowed a respondent to chose
from strongly disagree (1), disagree (2), somewhat disagree (3), somewhat agree (4),
agree (5), strongly agree (6). This scale forced respondents to make a definitive choice
for each item rather than a more indecisive choice of either agreeing or disagreeing. To
prevent respondents from responding in patterns, patterned response bias, the authors
recoded several of the items in the survey by reversing the meaning of the response
(Grandzol & Gershon, 1998).