Individuals with better spatial thinking have increased interest and greater achievement in science, technology, engineering, and mathematics (STEM) disciplines (Wai, Lubinski, & Benbow, 2009). This relationship means that STEM education may benefit from leveraging spatial thinking, but measures of spatial thinking as they relate to specific STEM disciplines are needed. The current work presents an assessment of spatial and mathematical reasoning, called Make-A-Dice. In Make-A-Dice, individuals are presented with a cube net (i.e., flattened cube) with numbers on two sides. Their goal is to "make a dice" by filling in the blank sides using two rules: opposite sides add to seven and use the numbers one through six once each. Make-A-Dice, math, spatial, and other measures were given to adults (Study 1) and elementary students (Studies 2 and 3), both across two sessions. Make-A-Dice had both internal and test-retest reliability, with items ordered by difficulty. Further, performance was related to spatial and mathematical reasoning. In Study 1, adults reported a range of strategies used to complete Make-A-Dice and one strategy predicted performance. Studies 2 and 3 showed that Make-A-Dice is age-appropriate for elementary students. Make-A-Dice shows promise as an individual difference measure linking spatial and mathematical thinking and has the potential of identifying elementary-aged children who may benefit from spatial training. [This is the in press version of an article published in "Behavioral Research Methods."]