Definite Algorithm Based on Causal Effect in SEM for the Analysis of Earthquake-Stricken Masses´ Mental Health

In recent years, SEM is widely used in many fields. This paper builds a mental health evaluation model for earthquake-stricken masses through the principle of structural equation model. Based on our MCLS solution for measurement equation by adding a constraint with modular length to the latent variable, we change the modular lengths of latent variables so that the path coefficients between any latent variables and its manifest variables satisfy prescription condition, at the same time, consider making the sum of error square least, we obtain a definite algorithm for SEM. With our model, the relationship of structure variables in index system can be analyzed, as well as the relationship between the construction and observation variables. Thus, we can make good analysis of the earthquake-stricken massespsila mental health, so related experts can take effective interventions.