The purpose of this study was to investigate the effects of scaffolding type and prior knowledge during a hypermedia supported problem-based learning program. Each of twenty intact sections of a computer literacy course was randomly assigned to one of three treatment groups - no scaffolding, content scaffolding, and metacognitive scaffolding. Three hundred and twelve undergraduate students enrolled in one of the twenty sections were randomly assigned to project teams and worked on a group project to build a personal computer by using a hypermedia supported problem-based learning program. The study examined the effects of scaffolding type and prior knowledge on individual posttest achievement, group project performance, attitudes towards the instruction, cognitive load, navigation patterns, and time in the program.

Results for achievement indicated that students in the content scaffolding treatment performed significantly better on the individual posttest than those in the metacognitive treatment. However, project teams given content scaffolds performed significantly lower on the group project than those in the no scaffolding condition. In addition to the findings for the scaffolding type, students with high prior knowledge performed significantly better on the individual posttest than those with low prior knowledge.

The attitude data revealed that students in the metacognitive scaffolding treatment reported that they performed more group behaviors than those in the content scaffolding treatment. Furthermore, high prior knowledge students had a significantly more positive attitude toward the program than low prior knowledge students.

Findings for navigation patterns indicated that students given content scaffolds visited significantly more learning resource web pages than those given metacognitive scaffolds. However, students in the content scaffolding condition visited significantly fewer hardware warehouse web pages than those in the other two treatment conditions. In addition, results revealed that students in the no scaffolding treatment spent significantly less time on the group project than those in the content scaffolding treatment and the metacognitive scaffolding treatment.

The results of this study have implications for the design and delivery of hypermedia supported problem-based learning environment. The findings suggest that content scaffolds can direct student attention to important content and encourage deep understanding. However, considerations should be given to the difficulty level of the task, the time allotted to solve the problem, and other demands students face in a problem-based environment.