Spatial reasoning correlates with academic success in mathematics and science, is highly malleable, and can be learned. Yet, spatial reasoning is often underused, underdeveloped, and ignored in current grade-school curriculum and teaching. This study explores changes in elementary students' spatial reasoning skills after participation in either a short-term or a long-term robotics intervention. The robotics intervention and the spatial elements within the intervention tasks are described. Measures of spatial reasoning elements were administered before and after each intervention. Two different groups of students ages 9-10 years were tested: a short-term group (N = 11) and a long-term group (N = 48). Statistical analysis revealed significant improvements to several different elements of spatial reasoning in both groups. Findings suggest that programming robots in either the short- or long-term intervention was associated with improvements in spatial reasoning.