Scaffolding for complex problem solving in science: What works better, general strategies or context-based ones?

Schoolchildren and teacher in science classMany middle school students don’t know where to begin when trying to solve problems that are open-ended and vague with multiple solutions or no solutions at all. Their meta-cognitive, self-regulation and reflection skills are often too limited for complex, ill-structured problems, says a recent study in Education Technology Research and Development.

These types of problems are of increasing importance in education as ill-structured problems are more often encountered in real life than are well-structured problems, says the study, which looked at how best to help middle school science students solve such problems.

Specifically, the study looked at what kind of scaffolding (e.g. prompts) is more helpful to students. Should teachers provide general scaffolding that gives students metacognition strategies or should teachers provide domain-specific (e.g. science-related) scaffolding that reflects the content the students are working in to solve the problem? Should scaffolding be continuous or faded (gradually removed)?

The study of 415 6th-grade students in 3 middle schools in the southwestern U.S. found that each type of scaffolding played its role in the more complex process of solving ill-structured problems. In general, domain-specific seemed to facilitate learning of scientific content, however, the researchers report.

The researchers built their study on a 4-step model for solving ill-structured problems developed in previous research (Ge and Land 2003):

1) Problem representation – The problem solver describes his or her interpretation and understanding of the problem including the constraints.

2) Develop solutions – Conflicting assumptions, evidence, and opinions lead to multiple solutions. The problem-solver s must select from among the various possible solutions to find the most suitable one.

3) Develop justifications – The problem-solver needs to develop justifications for the selected solutions by explaining why it will work as well as its possible difficulties and how these difficulties may be resolved.

4) Monitor and evaluate – The problem-solver must constantly monitor and evaluate the chosen solution to apply appropriate strategies and stay on track.

The researchers found that students who received domain-specific prompts were more effective in problem representation and those who received general prompts were more effective at evaluating solutions, discussing drawbacks and providing alternate solutions. Students performed equally well at the other two steps in the process—developing solutions and developing justifications—whether they received domain-specific or general scaffolding.

“Intelligently devised combinations of these two types of scaffolds can be used to improve students’ performance in justification and scientific explanation,” the researchers write.

Continuous scaffolding was more effective than faded support with students who were provided domain-specific support, the researchers report. The use of continuous or faded scaffolding seemed to make no difference in performance with students provided with general scaffolding. The results suggest that students really do learn to apply general strategies in other contexts, the researchers write.

Engaging students with complex problems

The 6th-graders in the study worked with a problem-based learning hypermedia program called Alien Rescue which is designed to engage middle school students in solving complex, ill structured problems. The primary objectives are for students to learn about the solar system and the tools and procedures that scientists use to study it. Students work at an “international space station” to find new homes for species of aliens whose planetary system has been destroyed. The students make these placements based on characteristics of the planets and of the displaced aliens.

One of the two researchers in this study was the lead instructional designer of Alien Rescue. The study was conducted during daily 45-minute science classes for a total of 13 class periods. Students took a multiple choice pretest 10 days before the study and were tested after the study to measure their learning of scientific content.

Students were assigned to one of 4 experimental conditions (general-continuous support, domain-specific continuous support, domain-specific faded support and domain-specific continuous support); a control group was not used in this study. Students recommended new environments and alternate environments for 4 alien species. Each student worked on the software and recommendation worksheet s individually. The recommendation forms were graded by 2 raters, including the primary researcher.

Examples of general prompts were:

To solve this problem, I need to find________________

Evidence to support my solution is________________

Examples of domain-specific prompts were:

Needs of the Akona (one of the alien species) are_______________

Characteristics of the selected world are____________

“The literature has already established that both domain-general and domain-specific scaffolds can be effective, but has not yet provided a solid base of data that can be used to determine the conditions under which they are effective,” the study says.

“Scaffolding middle school students’ content knowledge and ill-structured problem solving in a problem-based hypermedia learning environment,” by Saniye Tugba Bulu and Susan Pederson, Education Technology Research and Development, 2010, Volume 58, pps. 507-529.

Leave a Reply

  • (will not be published)