A long-term study of educational robotics and achievement in math and science

In recent years, educational robotics has become a popular tool in STEM programs, such as afterschool clubs, and summer camps, as well as classrooms. However, the research on the benefits of robotics have shown mixed results. In addition, many of the studies lack strong controls and focus on short-term effects, while the programs they investigate have few contact hours and do not have a consistent curriculum. This situation indicates that more research is needed. This work focuses on a public high school in Texas with a year-long robotics class. The first part examines a set of students who enrolled in the robotics class in the 9th or 10th grade, and a comparison group of students who did not enroll in robotics. The robotics and comparison groups were matched on 8th grade standardized math test scores, and demographic factors. Using multiple linear regression and logistic regression, I found that robotics enrollment was not a significant predictor for 11th grade math standardized test scores, or high school enrollment in Physics 1, Physics 2, or Calculus classes. The second part examines a series of video recordings of student teams in the robotics class working on a capstone project. Using grounded theory, I coded and analyzed recordings of two of the teams, focusing on the math and science discussions between the students and the contexts in which the math and science occur. Three themes emerged from the data. First, students use math/science more frequently to identify and fix problems than in their initial design. Second, students use math/science at a conceptual level and do not perform math calculations. Last, students have a “good enough” attitude and do not prioritize precision. These results may help explain the lack of effect robotics have on math test scores.