SCALE® and ESAP® Cronbach’s alpha Reliability Statistics
|SCALE® (N = 98)||ESAP® (N = 98)||ESAP® (2004, N = 1,389)|
|α||No. Items||α||No. Items||α|
Overall, the SCALE® provided strong evidence of internal reliability (Hammett, in press). The SCALE’s composite measure of Interpersonal skills (combining Assertion, Comfort, & Empathy) was high (α > .80), as were the composite measures of Career and Life Skills (combining Drive Strength, Decision Making, Time Management, Influence, & Commitment Ethic), and Personal Wellness (combining Stress Management & Physical Wellness). The Problematic Indicator composite (combining Aggression, Deference, & Change Orientation) was also high. Finally, the overall reliability when combining all items less those making up the Problematic Indicators was very high (α = .94). Overall based on these reliability statistics, it can be said that the SCALE® assessment reliably measures the global and composite skills constructs purportedly assessed by the instrument.
Reliability statistics for the ESAP® are also provided in Table 2. Including them was necessary because it was the instrument used for the SCALE® validity comparisons (Hammett, in press). Its baseline statistics for reliability may also be informative, therefore, as a comparison to SCALE®. Overall, the ESAP® yielded very high internal estimates of reliability with all composite scales exceeding α = .92. Finally, combining the ten skills resulted in a total instrument skills reliability of α = .96. One likely reason for the slightly higher composite scale alphas for the ESAP® compared to SCALE® is the increased number of items in each scale. Consider, for example, the 25 items that assess Stress Management for ESAP® compared to only 7 items for the same skill for SCALE®. As explained by Kaplan and Saccuzzo (2018),
According to the domain sampling model, each item in a test is an independent sample of the trait or ability being measured. The larger the sample [of items representing the domain], the more likely that the test will represent the true characteristic. In the domain sampling model, the reliability of a test increases as the number of items increases. (p. 124).