哪些研究问题需要用结构公式模型回答

结构公式模型是根据因果关系假设，同时求解由多元变量构成的一组线性方程的方法。它由反映观测变量与潜在变量之间关系的测量模型和反映潜在变量之间关系的结构模型两部分组成，其最典型特征是要求研究者在理论指导下对变量间关系作出明确假设。结构公式模型是目前惟一能够对所有变量关系同时进行完整分析的统计技术，它的产生与应用，反映了社会科学研究的复杂化和多元化。通过实例介绍了将研究问题与结构公式模型有机结合的方法：

What Kinds of Research Questions Need to Be Answered by Structural Equation Modeling

Structural Equation Modeling (SEM) is a statistical method that simultaneously find solutions for a set of linear equations consisted of multiple variables according to causal relationship hypotheses. SEM includes two parts, i.e. a measurement model that reflects the relationship between observable variables and latent variables and a structural model that reflects the relationship between latent variables. Its most typical characteristic is to require researcher propose before data collection a clear hypothesis about the relationship among variables according to a theory. Until now SEM is the only statistical method allowing simultaneous and integrated analysis of all relationships among variables. Its birth and use are the indication of more complicity emerged in research questions. The method of dynamically integrating research questions and SEM was introduced with empirical research examples.