结构方程模型.pptVIP

Duncan(1984)的一席话 Coupled with downright incompetence in statistics, paradoxically, we often find the syndrome that I have come to call statisticism: the notion that computing is synonymous with dong research, the na?ve faith that statistics is a complete or sufficient basis for scientific methodology, the superstition that statistical formulas exist for evaluating such things as the relative merits of different substantive theories or the ‘importance’ of the causes of a ‘dependent variable’; and the delusion that decomposing the covariations of some arbitrary and haphazardly assembled collection of variables can somehow justify not only a ‘causal model’ but also, praise the mark, a ‘measurement model.’ There should be no point in deploring such caricatures of the scientific enterprise if there were a clearly identifiable sector of social science research wherein such fallacies were clearly recognized and emphatically out of bounds. 因子 Factors are influences that are not directly measured but account for commonality among a set of measurements. X1 X2 X3 F X1 X2 X3 u1 u2 u3 F 探索性因子分析 将可测随机向量与潜在因子连接起来的线性模型 目的：寻找少数几个因子，以解释观察变量之间的相关性 没有如下先验知识 公因子数 因子载荷 因子间关系 假定误差项无关 提出假设： 公因子数 因子载荷 放松假定： Φ和Θδ为对称阵 验证性因子分析 EFA与CFA的关系 相同点：CFA类似于EFA的简单结构 不同点： 用途不同 EFA在于探索（归纳） CFA在于检验（演绎） 假定不同 因子相关性 测量误差相关性 估计方法不同 EFA常采用谱分解 CFA常采用MLE、GLS等 ξ1 X1 X2 δ1 X3 X4 δ2 δ3 δ4 φ12 ξ2 1 1 λ42 λ21 λ31 λ32 λ12 λ11 λ41 λ22 EFA CFA Joreskog’s major contribution arises from the fact that he appears to have realized that most researchers who use factor analysis cannot specify exactly all the loadings for the pattern matrix, not even the correlations among the factors when they generate hypotheses about the expected factors. However, frequently they can specify the zero elements of the pattern matrix, and sometimes, on the basis of past analyses, they can specify the exact values of a few of the nonzero loadings of the pattern matrix. Consequently, Joreskog sought to formulate the factor