On the Use of Confirmatory Measurement Models in the Analysis of Multiple-Informant Reports
研究了验证性测量模型如何分解多信息源数据的方差，并指出不同因子模型设定会影响特质效度和测量特异性的含义，从而影响模型评估。
In applied research it is important to understand the implications of the factor analytic model used to represent the covariance structure underlying a set of observed measures. Here the focus is on the use of confirmatory measurement models in the analysis of multiple-informant reports. By effecting a variance decomposition that partitions the variation in measurements into constituent components, the authors investigate the implications of first-order and second-order confirmatory measurement models as they apply to key informant data. Among other things, the authors demonstrate that depending on the particular factor analytic specification used, trait validity and measure specificity take on different meanings and consequently affect the evaluation of the model being considered.