Abstarct: In some studies, there are variables that can not be directly measured or observed, often referred to as latent variables. Structural equation modeling provides a direct method for modeling latent variables, which combines structural model and ‎measurment‎ model to formulate structural equations. When there is a strong correlation between predictor variables, it is said multicollinearity exists. ‎In such cases, the least squares estimator is not practical and often shrinkage estimators are used instead. One of such, is the ridge estimator.‎ Here we examined the ridge method for solving structural equation modeling in ordinal and continuous data. In this regard, a constant value is added to the diagonal element of the correlation matrix. Then we minimized the objective function in the structural equation model on the basis of this change in order to estimate the model coefficients. We evaluated this method by a real example.‎ Finally, we also gave a systematic modeling of the structural equations in the form of an example and compared the ridge method with some other alternatives.