Teaching is more effective when students are taught through research-based teaching approaches that have been found to be successful in producing the outcomes they advocate. However, use of an effective instructional strategy does not ensure learning, because the success of a sound approach is determined by how well it is implemented. The purpose of this study was to evaluate the effect of the fidelity of implementation of the Real Engagement in Active Problem Solving (REAPS) teaching model on changes in creative problem solving in mathematics. REAPS is a teaching model that was initially designed to meet the needs of exceptionally gifted and talented learners. However, teachers and researchers have found it to be an effective method for teaching all students. Using a repeated measures t-test, we found that creative problem solving of students in a school in which all teachers implemented the REAPS model improved significantly in total scores as well as accuracy and concept subscores. In addition, we found that the level of fidelity of implementation of the model explained 14.8% of the variance in total creative problem solving in mathematics, 20.1% of the variance in accuracy subscore, and finally 4.5% of the variance in concept subscore. We found evidence that teachers can make a significant impact on students' creative problem solving in mathematics through a high level of fidelity of implementation of the REAPS model.