Scaffolding students' problem-solving processes on an ill-structured task using question prompts and peer interactions

Quick take
Metacognitive prompts help students solve problems, with or without student teams.
2001 Xun Ge
Polya proposed four problem-solving steps: understanding the problem, devising a plan, carrying out the plan, and looking back at work.
Technology-enhanced scaffolding can situate problem identification and engagement by providing vivid descriptions, visualizations, and related questions and resources to students’ experiences.
Such scaffolds can also assist students as they seek information.
KRM: hints - where to find clues.
Hint tag - button that will show content, but only after warning the student that s/he shouldn't rely on it.
Students have limited prereq knowledge
While students’ difficulties in problem solving are partly attributed to misconceptions or shallow conceptions of domain knowledge (P. J. Feltovich, Spiro, Coulson, and J.  Feltovich, 1996), they are, to a greater extent, due to alack of metacognitive knowledge (Brown, 1987).
supports should be provided to students during problem solving in cognition and metacognition through various scaffolding strategies, such as:
coaching through prompts
modeling through reciprocal teaching or peer-regulated learning
and guiding students to self-generate questions.
While this study confirmed the findings of previous research on the effectiveness of question prompts in facilitating students’ cognition and metacognition, it also showed the benefits of peer interactions, which were contingent upon group members’ active and productive engagement in peer interactions, that is, questioning, explaining, elaborating and providing feedback among peers.
in order for students to gain full benefits from peer interactions, the peer interaction process itself need to be scaffolded, especially when students were novice learners in problem solving; and question prompts, through expert modeling, may serve to facilitate this process.
questioning strategies helped students to focus attention on their learning process and to monitor their learning through elaboration on the question asked
question prompts designed to access prior knowledge or experience were more effective in enhancing comprehension.  
asking and answering metacognitive questions helped students to focus on the process of problem solving
KRM: iunstruction on group problem-solving in the course.
Through social interactions, more highly-skilled learners or peers can provide modeling of higher-level thinking
"reciprocal teaching" involving dialogue between the teacher and the student helped to foster comprehension and comprehension monitoring activities.
KRM: ask student to explain things to me. How does the program work? Why does it do that?
The benefits of dialogue in reciprocal teaching lie in the following: (a) extensive modeling of the comprehension-fostering and comprehension-monitoring activities by the teacher; and (b) task requirements for students to ask and respond to questions, which made their thinking explicit.
Scaffolds may be tools, such as cue cards, or techniques, such as teacher modeling, questioning, and explanation
Procedural prompting
questions can direct students' attention to important information, or alternative perspectives which students may have ignored
questions allow a learner to elaborate his or her thought process and develop one's understanding
"why" questions activated the students' prior knowledge related to the new concepts.
questioning strategies can help to activate one's schemata
The following are examples of generic question stems used in the studies by King: "how are ...and...alike?" "What is the main idea of ...?" "What are the strengths and weakness of...?" "How is this related to..." "What conclusion can you draw about?"
KRM: questions for programming
KRM: dude can ask questions, or be asked questions? Maybe give students questions stems, e.g., for debugging:
why does that variable have that value at that point?
How do you know that that code is being executed?
When is that code run? Conditions?
What does the value of that variable depend on?
Planning is an essential executive strategy and provides evidence of metacognition. It is necessary to make a problem-solving plan intentionally and to carry out that plan
KRM: my implicit plans for debugging?
KRM: program writing: ways of adding structure:
- what are the big pieces? Use experience to identify other pieces that I have made.
King (1991a) grouped metacognitive questions into three executive levels or categories—planning, monitoring and evaluating.
For example, to identify the problem (or redefine the problem, which is often necessary during problem solving process), questions such as, "What are we trying to do here?" can be asked which are expected to help students determine the nature of the problem more precisely.
Questions such as "What information is given to us?" would presumably help students to access prior knowledge
Based on information-processing theories of problem solving, two types of strategies are generally applied in problem representation: schema-driven strategies and search-based strategies (Gick, 1986).  
Search strategies may involve the comparison of problem states to the goal state and the use of information gathering strategies.
peer interaction during open-ended learning was effective when group members offered suggestions, negotiated ideas, and shared their experiences.
benefits of peer interactions are contingent upon a high level of peer interaction, that is, active engagement in asking questions and providing elaborations and feedback, which in turns is dependent upon group dynamics where group members are willing and see the need to engage in such a knowledge construction activity.
the following groups of question prompts:  
1)   Problem Representation Prompts: "How do I define the problem?"  (There are 3 questions and 4 sub-questions to Question #3 in this category.)  
2)   Solution Prompts: "How do I generate the solutions?"  (There are 3 questions in this category.)  
3)   Justification Prompts: "What are my reasons and argument for the proposed solutions?"  (There are two general questions in this category, with 1 sub-question to each question.)
4)   Monitoring and Evaluation Prompts: "Am I on the right track?"  (There are 2 questions and 3 sub-questions to Question #10.)  
KRM: students can make worksheets, with question prompts, for an exercise.
Could be a general document type. System creates a body field with some questions in it.
The findings of this study on the effects of question prompts have two implications for instructional design:  
First, question prompts are useful in supporting metacognitive knowledge in students' ill-structured problem-solving, such as reasoning, monitoring, reflection and evaluation, which makes learning activity "mindful" or "intentional"
Second, domain knowledge seems important for students to take full advantage of the strategy of question prompts, which are designed to facilitate both cognitive and metacognitive knowledge.
learning how to employ cognitive strategies may sometimes be dependent on having some relevant knowledge of the domain.
Instruction of domain knowledge could be interwoven into a complex, ill-structured learning environment to help students acquire domain knowledge, while scaffolding strategies, such as question prompts, could be embedded in the instruction to support ill-structured problem-solving processes.
The case study revealed that the individual students who received question prompts (e.g., Case 1, 2) were prompted by the questions to attend to the important aspects of the problem (e.g., factors and constraints) and were able to map those factors and constraints to their problem solution.  
students in the condition of peer interactions without receiving question prompts had significantly lower problem-solving performance than those in the condition of individuals who received question prompts.
KRM: questions can give some of the benefits of group work, without other people.
KRM: questions created by experts might be better than group of novices trying to get things done. THey don't know how.
The students working with peers but also receiving question prompts did significantly better in every problem-solving process than those who did not receive question prompts.  
KRM: questions scaffold group interaction, as well as individual thought.
KRM: students could use worksheets to record group interaction.
The results indicated that when peers with equal levels of competence work together, additional scaffolding to support peer interactions seemed critical.  The positive effects of reciprocal teaching among the peers as demonstrated by the previous studies (e.g., Palincsar & Brown, 1984; Brown & Palincsar, 1989) is perhaps due to the fact that question asking was modeled and dialog between the teacher and the student was provided before the students were asked to provide peer tutoring each other.  
when support is provided to students for group collaboration, the benefits of peer interactions can be maximized.  
A common pattern emerging from the qualitative data showed that the students frequently related the problem to their personal experience as a WalMart user
The WalMart situation helped the students to picture themselves as customers and allow them to form the representation of a given situation that was familiar to them.  Students' knowledge about a given situation helped them to activate appropriate schemata.
An interesting phenomenon found in this study was that the students tended to overrate their own problem-solving skills, which seemed to contrast with their actual competence of problem solving.
high self-efficacy resulted from inaccurate rating may also influence students' self-regulation process.  In this study, while the students' high self-efficacy about their problem-solving abilities may motivate them to set goals for problem-solving tasks, it may also limit their motivation to seek knowledge or strategies that may help them improve their problem-solving skills.