The Cronbach's alpha coefficient was 0.67 for the complete
instrument, which should be acceptable for group level comparisons,
where a value of 0.7 often is considered as sufficient. When an
instrument is built up of items that are causal indicators, the demand
on high homogeneity is not so crucial, because items with different
substantial content contribute to the comprehensive phenomenon
that is intended to be measured. The demand is, thus, not as great as it
is for effect variables that have to reflect the underlying construct
(Streiner and Norman, 2003). The items of MDSP are partly such
causal indicators. But some of them may be considered as effect
indicators. However, moral development according to Kohlberg's
theory is a highly complex phenomenon that is influenced by several
causal factors, and it could be concluded that the demand on high
homogeneity for this instrument should not be very relevant (Terwee
et al., 2007).