Increasingly, computer-based learning systems are used by educators to facilitate learning. Evaluations of several math learning systems show that they result in significant student learning improvements. Feedback provision is one of the key features in math learning systems that contribute to its success. We have recently been uncovering feedback design patterns as part of a larger pattern language for math problems and learning support in online learning systems. In this paper, we present three feedback design patterns developed from the application of the data-driven design pattern methodology on a large educational dataset collected from actual student data in a math online learning system. These design patterns can help teachers, learning designers, and other stakeholders construct effective feedback for interactive learning activities that facilitate student learning.