2006 Volume 3 Pages 1-10
The purpose of this study was to examine the factorial validity and cross validity of an expanded soccer attacking skill scale (Expanded SASS) measured by the location of players in soccer games using multiple-group analysis by structural equation modeling. The samples were 388 attacking performances in the final of the FIFA World Cup Korea/Japan 2002TM, the final of the 27th all Japan University Prime Minister’s Cup, and the final of the 18th Japan Club Youth Football Championship which were measured by five-point interval scales by distance, and number of players. From the result of confirmatory factor analysis (CFA) that was constructed using one of the samples that were randomly split into two groups, the same model as was used in a previous study, except for an addition of correlated uniqueness, was accepted. The result of cross-validation by the other sample accepted the model with equality constraints of all parameters across samples. These results confirmed that the Expanded SASS was highly cross-validated. Nevertheless, there were two items that were interpreted as low factor loading; therefore, we need to add evidence of validity according to the intended use of the Expanded SASS.
