This study was designed to explore the relations between different metrics of performance on the number line task and performance on mathematics achievement tasks. Schoolchildren were given these tasks in the spring of second grade and the fall of third grade. Their performance on the number line task was assessed using the percent absolute error, linearity, and slope of each childs results. These values were compared to the childrens performance on mathematics exercises requiring mental arithmetic and word-problem solving. In most cases, models including percent absolute error showed the strongest correlations and the greatest likelihood of being the correct model.