The Partial Least Squares (PLS) method is a machine learning technique developed based on the Ordinary Least Squares (OLS) approach, capable of handling nonlinear relationships between independent and dependent variables as well as complex data structures. Due to its features, PLS is often used in the literature as an alternative to methods such as OLS, Principal Component Analysis, and Multicollinearity techniques. The advantages of PLS arise from its semi-parametric nature, which exempts it from certain assumptions such as normality, and its ability to mitigate issues like multicollinearity within the model. Although initially applied in the fields of chemistry and biology, PLS has gained popularity in social sciences—including economics and psychology—due to its ease of use and flexible model structure. Recently, the method’s algorithm has been increasingly employed in Structural Equation Modeling (SEM) because of its ability to examine relationships between observed and latent variables. Known as PLS-SEM, the method offers significant advantages, including applicability in small sample sizes, robustness when prior knowledge is limited, no requirement for normality assumptions, and facilitation of estimation in complex relationships. In addition to these methodological advantages, user-friendly software such as SmartPLS, WarpPLS, and ADANCO has contributed to the widespread adoption of PLS-SEM in social science research. This chapter first presents the key characteristics and mathematical foundation of PLS. Next, the estimation steps of PLS in Structural Equation Modeling are explained. Finally, practical applications using example datasets are demonstrated to illustrate the use of PLS-SEM in applied research.