An investigation into the measurement invariance of the performance index

[摘要] The leadership-for-performance framework designed by Spangenberg and Theron (2004) aspiresto explicate the structural relationships existing between leader competency potential, leadershipcompetencies, leadership outcomes and the dimensions of organizational unit performance. ThePerformance Index (PI) and Leadership Behaviour Inventory (LBI) comprise the leadership-forperformancerange of measures. The PI was developed as a comprehensive criterion measure ofunit performance for which the unit leader could be held responsible. The basic PI structuralmodel has been developed to explain the manner in which the various latent leadershipdimensions measured by the LBI affect the eight unit performance latent variables that areassessed by the PI. Although preliminary research suggests the basic PI structural model could berefined, continued research in this regard can only be justified if the basic PI measurement modelis shown to be measurement invariant across independent samples from the target population. Aspart of ongoing research of the leadership-for-performance range of measures, this crossvalidationstudy investigated the extent to which the PI measurement model may be consideredmeasurement invariant across two independent samples from the same population. Two sampleswere collected through non-probability sampling procedures and included 277 and 375 completecases after imputation by matching. Item analysis and dimensionality analysis were performed oneach of the PI sub-scales prior to the formation of item parcels. No items were excluded basedon item- and dimensionality analysis results. Two composite indicator variables (item parcels)were created from the items of each sub-scale and were treated as continuous variables in thesubsequent statistical analyses. Structural equation modelling, using robust maximum likelihoodestimation, was used to perform a confirmatory first-order factor analysis on the item parcels foreach sample. The measurement model was fitted to both samples independently and close fit foreach sample was established. The measurement model was cross-validated using a progressiveseries of measurement invariance tests. Results indicated the PI measurement model did notdisplay full measurement invariance across the two samples although it did cross-validatesuccessfully under the configural invariance condition. Statistically significant non-equivalencewas found to exist in both the measurement error variances and the factor covariances (p<0,05),although the p<0,05 critical value was only narrowly surpassed in both cases. The measurementmodel did, however, display metric invariance across the samples as no significant differenceswere found between the factor loadings, suggesting the content of each item is perceived andinterpreted in a similar manner across samples from the target population. When considered incombination, these results may be viewed as quite satisfactory as they indicate that themeasurement model does not appear to vary greatly when fitted to data from the two samples. Asthis study has established at least metric invariance of the PI, it therefore provides some basis ofconfidence for proceeding with subsequent research aimed at establishing the structuralinvariance of the basic PI structural model and eventually research that links the leadershipbehaviour to work unit performance as measured by the PI. Limitations of this study arediscussed.

[发布机构] Stellenbosch University