Objective: It has been argued that ADHD characteristics children have difficulties in selecting the best strategy when they accomplish cognitive tasks. The detrimental influence of these poor strategy skills may be crucial for several aspects of academic achievement such as mathematical learning. Method: Fourth- and fifth-grade children with ADHD symptoms and matched controls were asked to select the better of two rounding strategies in a computational estimation task (i.e., finding the best estimate of two-digit addition problems). Results: (a) Both control and ADHD children correctly executed a selected strategy, (b) ADHD children selected the best strategy less often than controls, (c) ADHD took more time to estimate sums of two-digit addition problems and provided poorer estimates, and (d) different factors predicted best strategy selections in each group. Conclusion: These findings have important implications for further understanding the sources of differences in cognitive performance between ADHD and control children.