Analysis and Results
An analytical plan was devised in order to summarize the procedural steps in interpreting the results. This plan outlined the steps in two phases (1) the reliability and validity of the measures and (2) the statistical analysis of the data.
A reliability analysis of the Mach IV, NPI and SAS instruments was conducted in order to assess the internal consistency of the items. For the Mach IV and NPI, internal consistency was determined by the coefficient alpha. These are scales (as opposed to indices) because they are intended to measure a single underlying construct (Machiavellianism and narcissism) and one would expect a strong inter-item correlation, if the instruments measured the phenomena they were constructed to measure (DeVellis, 1991). The reliability of the SAS was assessed by arranging the items, based on level of progressive violence, into a Guttman model, for which a coefficient of reproducibility was calculated.
The reliability analyses were then used to make decisions about the internal consistency of the instruments, and to determine whether there were items that for whatever reason might be best excluded to improve the instrument's homogeneity. These reasons might include obsolete wording or references, ambiguous syntax (to include sexist