A study was conducted to demonstrate the most commonly used predictive quantitative structure-property relationship (QSPR) modeling methods and their applications to materials design. QSPR methods were based on the hypothesis that changes in molecular structure were reflected in changes in observed macroscopic properties of materials. QSPR modeling was a supervised learning method that extracted the complex relationships between the microscopic structure and properties of materials and their macroscopic properties. The key requirement for QSPR modeling was a reliable data set of molecules or materials whose microscopic structures and properties were well-defined along with their measured macroscopic properties of interest. The reliability of the experimental property chosen to be modeled was important, as it was one of the factors that determined the stability and predictivity of models.