The purpose of this study was to explore middle grade mathematics students' uses of scaffolding and its effectiveness in helping students solve non-routine problems. Students were given two different types of scaffolds to support their learning of sixth grade geometry concepts. First, students solved a math task by using a four square graphic organizer that included the identification of critical components, strategy selection, computation and analyses of answers. The second type of scaffolding occurred with different grouping formats, alternating work led by the teacher, working alone. Measures included extended response daily tasks, teacher-created unit test, state standardized assessments, and surveys gauging student's satisfaction with types of scaffolding. Classroom use of a variety of scaffolds led to an increase in the number of correct responses and more detailed explanations. No significant differences occurred on the teacher-made tests, yet significant increases were found on students' state standardized tests. Students indicated that scaffolding by groups was helpful in initiating solution pathways.