ATTITUDES TOWARD BULLYING ROLES
The attitudes regarding roles in bullying were assessed with a measure
devised for the present study. Participants read the following lines: “In this
table there are several behaviors that can be observed in young people. Say
what you would think if someone behaved like that with other colleagues.”
The same behaviors listed for the normative beliefs were presented (see
Table 2) and again participants evaluated them on a 5-point Likert-type
scale: 1 = very bad, 2 = bad, 3 = more or less, 4 = well, and 5 = terrific,
which is in agreement with the definition of attitude as an evaluative tendency
to impute some degree of goodness or badness to an entity (Eagly &
Chaiken, 1993). The internal consistencies for the roles of bully (a = .90)
and defender (a = .67) were moderate to strong. Given the shortness of the
latter scales, the observed alphas can be seen as sufficiently reliable. Principal
components analysis of all the items with varimax rotation revealed the
presence of two components with eigenvalues greater than 1, explaining
35.19% and 17.63% of the variance (52.82% in total). The factor loadings
presented a simple two-factor structure, with the items related to the bully
role loading on the first component and the defender role items loading on
the second. All factor loadings are above .43. For all measures, scores were
computed by averaging across items, with higher scores indicating stronger
endorsement of the construct. Gender was depicted by a dummy variable
with 0 = male and 1 = female.