Besides the factor structure of the

Besides the factor structure of the SWLS, its factorial invariance needs to be addressed. Factorial invariance is an important property for a measurement that was developed to comparethe mean level of a certain construct or trait among different groups, since interpretation ofthe mean differences may be problematic unless the underlying constructs are the same acrossgroups. Therefore, when a measurement is used to compare mean differences among differentgroups, the measurement should have the same meaning across groups. Moreover, factorialinvariance is an important property for case selection. Millsap and Kwok (2004) indicated thatpartial factorial invariance (when some, but not all, parameters are invariant) would debase theaccuracy of selection when the composite score was used as the criterion for case selection.Thus, if a measurement is used to select cases with certain characteristics, the property of factorial invariance should be tenable if different groups are combined within a large sample.