CHAPTER TWO: LITERATURE REVIEW
The present study investigates the effects of learning task and learning sequence on programming learning based on the aspects of problem-solving, and adopts the strategy of task-oriented learning and individualized learning. Thus, this chapter will explore problem-solving, task-oriented learning, and individualized learning respectively.
Problem-Solving
The Intention of Problem-Solving
Because people encounter many difficulties everyday, problem-solving is regarded as the most important cognitive ability which everyone should possess to live in the world (Jonassen, 2000). For this reason, problem-solving ability is taken seriously in school in recent years, and becomes a quite major element of instructional objectives (Ministry of Education, R.O.C., 2004).
In the process of solving problems, learners integrate their prior knowledge and experience to the problem space, and elaborate their existing knowledge to solve problems that then results in the formation of new knowledge (Dahlgren & Öberg, 2001; Harland 2003; Jonassen, 2000; Morrison, 2004; Tennyson & Breuer, 2002). It shows that learners could administer the demands, progress, and process of learning in
problem-solving by themselves to constitute self-paced and self-directed learning.
Actually, a problem is no other than a task. Chi and Glaser (1985, p.229) described that a problem is a “situation in which you are trying to reach some goal, and must find a means for getting there”. Some other researches (Bransford & Stein, 1993;
Hayes, 1989; Jonassen, 2000) then proposed that a problem is typically defined as a gap or barrier between a goal state and one’s current state. The same, especially in the area of education, a task is “a piece of work or an activity, usually with a specified objective, undertaken as part of an educational course, at work, or used to elicit data for research” (Crookes, 1986). In other words, accomplishing a task could be looked upon as solving a problem, so this study would implement task-oriented learning in the aspect of problem-solving, and the intention of it will be specified in the following section which is named “task-oriented learning”.
The Process of Problem-Solving
The goal of problem-solving is to “find a way to overcome many difficulties and obstacles, then to attain an aim that was not immediately attainable” (Heh, 1999, p.197). As regards the details inside the process of problem-solving, there are many statements from the different researchers. For example, Dewey (1910) articulated the scientific aspect for problem-solving, and proposed the following assertions. First,
problem solvers should discover the problem and define it. Second, suggest possible solutions for the problem and identify alternative to select the best one. Finally, problem solvers implement and test the solution for revising it to the most suitable solution. Nevertheless, Wallas (1926) analyzed the progress of problem-solving through the creative view, and he described four stages for solving problems: (1) Preparation – it concerns the conscious accumulation of knowledge and draws upon influences from previous experiences. (2) Incubation – the conscious thought pertaining to the problem is rested and leaves to the unconscious mind. (3) Illumination – it occurs when one experiences a sudden flash of insight or sudden inspirations. (4) Verification – it tests for accuracy.
However, Norris and Jackson (1992) indicated that problem solving only comprises two major processes: understanding and solving. Understanding is the process that helps you ascertain and comprehend the nature of the gap, and solving is to execute the procedures that bridge the gap. Bransford and Stein (1993) then proposed a well-known IDEAL problem solver – Identify problems and opportunities, Define goals, Explore all possible strategies, Anticipate outcomes and act, Look back and turn – to help solve problems more effectively and efficiently. Carder, Willingham, and Bibb (2001) and Ronteltap and Eurelings (2002) suggested more detailed steps to solve a problem, that learners analyze the case, take the initiative to define their
learning needs, formulate the learning issues or goals, identify needed resources for learning, select and implement learning strategies, and evaluate the results.
Through overview of above propositions, problem-solving methods include at least four steps (Chang, 2002; Deek, Turoff, & McHugh, 1999; Huitt, 1992; Polya, 1945): (1) understanding and defining the problem (Input Phase), (2) planning solutions and then collecting necessary information (Processing Phase), (3) designing and implementing the solution (Output Phase), and (4) verifying and presenting the results (Review Phase). Huitt’s (1992) list is just a mirror of the information processing model (Figure 2-1) of problem-solving.
Figure 2-1. Information processing model (cited from Schunk, 1996)
Input Sensory
Register
Working Memory
Long-Term Memory
Response Mechanisms Control (Executive) Processes
active
In the “Input Phase”, problem solvers have to find out and comprehend the problem, meaning to form a mental representation. This behavior requires translating
known information into a representation in working memory. Then, information in working memory activates related knowledge in long-term memory, and solvers could select a suitable problem-solving strategy to plan possible solutions. During solving problems, solvers would often revise their initial representation and elaborate it into new knowledge when problem-solving does not succeed. Finally, they carry out the solution and evaluate the goal progress so-called “Review Phase”. Therefore, this is the evidence that problem-solving is a vital ability for knowledge acquirement.
Programming Instruction and Problem-Solving Skills
According to the curriculum standards of senior high school education in Taiwan, the objectives of programming instruction include (1) guiding learners to learn the conception and the theorem of computer science, (2) fostering learners the skills of applying computer to solve problems, and (3) laying the foundation for learners to learn computer science further (Ministry of Education, R.O.C., 1995). Linn and Dalbey (1985) stated that an ideal chain of cognitive accomplishments has three main links: (1) single language features, (2) design skill (includes templates and procedural skills), and (3) general problem-solving skills. In other words, when programming, learners will decompose the problem into component parts, and combine language features and use procedural skill of planning, testing, and reformulating to form a
program. Fay and Mayer (1988) again concluded that “programming can serve as a vehicle for learning general thinking skills” (p. 73). Thus, students are expected to transfer the very similar concepts such as loop structure to a new programming language environment, even to a new problem-solving domain in the real-world through programming instruction (Palumbo & Reed, 1991). Through overview of above propositions, the main instructional objective and the main function of programming courses consist in how to use programming to solve a problem, that is to say, problem solving transfer (Mayer, 1992).
Singh and Zwirner (1996) have proposed that program creation involves four stages: (1) understanding of a problem, (2) method establishment, (3) transformation of method into code, and (4) evaluation. Singh and Zwirner (1996) also argued that debugging comprises another four stages, including (1) realization that a bug exists, (2) diagnosis of the bug, (3) repair of the bug, and (4) evaluation. Through thinking and ruminating over both these stage processes, it can be seen that the flow paths of programming and debugging are consistent with the flow path of problem-solving. In other words, if a student learned how to systematically design programs, he or she also learned the skills of logical analyzing, critical thinking, concise writing, problem solving, and fact checking, all of which are generally useful (Felleisen, Findler, Flatt,
& Krishnamurthi, 2004).
Furthermore, Linn and Dalbey (1985) concluded that problem-solving skills were useful for learning new programming languages. For syntactic knowledge, semantic knowledge, schematic knowledge, and strategic knowledge, Mayer (1992) stated that an expert programmer would have stronger knowledge of programming language, more integrated conceptual models of the interior of computer system, and use high-level planning strategies for program design. Robins, Rountree, and Rountree (2003) again suggested that expert programmers could agilely use general problem-solving strategies to efficiently decompose programs. However, novices usually are limited to surface knowledge, fail to apply relevant knowledge, lack mental models of programming, approach programming “line by line” rather that at the level of meaningful structure, and spend very little time planning, so novices are often in a maze when programming (Winslow, 1996). Due to the importance of problem-solving, and the close relevance between programming and problem-solving skills, this study was designed to conduct students learning programming by the task-oriented learning activity from the aspect of problem-solving.
Task-Oriented Learning
Task-Oriented Learning
Task-oriented learning or task-based learning originally is an approach usually used in language learning. On the whole, there are six main types of tasks, including listing, ordering and sorting, comparing, sharing personal experiences, creative tasks, and problem-solving (Willis, 1996). For the task of problem-solving type, students are presented with open-ended, challenging, complex problems or scenarios that may occur in real world, but the most important is that they do not have all resources they need to develop the solutions (Enger, Brenenson, Katy, MacMillan, Meisart, Meserve,
& Vella, 2002; Morrison, 2004; Pederson & Liu, 2002; Wood, 2003), so the learners have to perform analysis, inquiry, and search to problems voluntarily.
Problem-solving tasks make demands upon people’s mental representation of the problem and perform operations to reduce the disparity between the initial state and the goal state. The approach of task-oriented learning is to break down the barriers between separate areas of tasks or problems, and bridge the gaps between discrete elements, by linking them through a well-designed matrix of interdependent and interrelated subtasks (Race, 2000). ChanLin and Chan (2004) stated that this kind of task-oriented learning approach organizes the task around a series cases profiling dilemmas of practice, which learners read, diagnose, discuss and explore strategies for
solving problems.
Thus, students are able to learn how to identify their learning needs, to identify the knowledge through these activities, and to identify how to best acquire the relevant knowledge (Dahlgren & Öberg, 2001). It can be seen that the activities of task accomplishment or problem-solving can stimulate learning from student-center standpoint, and then try to cause learners’ motivation in self-directed and collaborative learning. Task-oriented learning is to help students effectively acquire knowledge and apply it to real life, enhance problem-solving skills and metacognitive skills, promote the ability to work cooperatively in groups and communicate effectively, as well as encourage the development of their critical thinking and lifelong learning skills in a professional area. In sum, the task-oriented learning can increase the relevance between the skills, knowledge, and competences (Race, 2000).
Instructional Strategy of Task-Oriented Learning
Of course, if instructors hope that the students will perform fruitful learning outcomes, then it must need a well-designed problem case or scenario. Dolmans, Snellen-Balendong, Wolfhagen, and van der Vleuten (1997) proposed some principles for effective problem design and suggested that problems should connect to the students’ knowledge and experiences. The problems should be complex but not
overloaded or too structured. Furthermore, the problems should stimulate self-learning by encouraging students to generate learning issues and conduct literature searches, enhance student’s interest in the subject matter and match one or more of the learning objectives by sustaining discussion about possible solutions and facilitating them to explore alternatives. Sometimes, several cues or concepts can also be provided in the problems to encourage integration of knowledge.
But why do experts emphasize that the tasks or the problems should not be too structured in the task-oriented or problem-based learning situation? However, there are three main types of problem: well-structured, moderately-structured and ill-structured (Table 2-1), and each kind of problem calls on different skills. In fact, the few problems that students do encounter are normally well-structured problems, while problems that are encountered routinely in everyday life are exactly “ill-structured” or
“ill-defined” (Jonassen, 2000).
Table 2-1
The Definition and the Characteristics of Well-structured, Moderately-structured and Ill-structured Problems (revised from Foshay & Kirkley, 2003)
Type of problem Definition Characteristics
Well-structured Problems that always use the same step-by-step solution.
y Solutions strategy is usually predictable.
y Problem-solving only require a limited number of rules and principles.
y Convergent (one correct solution).
y All initial information is usually part of the problem statement.
(These problems are usually found at the end of textbook chapters and on examinations)
Moderately-structured Problem that require varying strategies and adaptations to fit particular contexts.
y Often more than one acceptable solution strategy.
y Convergent (one correct solution).
y Needed information often must be gathered.
Ill-structured Problems with vague and unclear goals.
y Solution is not predictable or convergent.
y Multiple aspects, goals, and solutions.
y There may not be a fully satisfactory solution at all.
y Needed information often must be gathered.
(These problems are often encountered in professional practice.)
Ill-structured problems appear ill-structured because they: (1) have vaguely defined or unclear goals and unstated constraints, (2) have multiple solutions, solution paths, or no solutions at all, (3) have multiple criteria for evaluating solutions, and (4) require learners to express personal opinions or beliefs about the problem (Jonassen, 1997; Jonassen & Kwan, 2001). Due to lack of clear identification of the problem, lack of procedures for identifying solutions, and lack of criteria for evaluating solutions, learners have to begin the process of identifying what they already know, what they need to find out, and generating a plan for solving the ill-structured problem (Carder et al., 2001; Lohman & Finkelstein, 2002). Therefore, their cognitive skills, metacognitive skills, and argumentation skills could be enhanced from the inside. This
is the reason why ill-structured tasks have more instructional functions than well-structured in the task-oriented learning strategy.
Enger, et al. (2002) found that well-designed student problems should be provided with the following features:
(1) Engage students’ interests and motivate them to probe for deeper understanding of a topic, and be complex enough so cooperation is necessary to solve the problem.
(2) Relate to real-world situations that are not limited to one correct answer and connect to previously learned knowledge.
(3) Require students to make decisions based on facts, requiring them to define what assumptions are needed, what information is relevant, and what steps are needed to solve the problem. (p.356)
Wood (2003) then provided another set of principles to create effective task-based learning scenarios:
(1) The scenario should be consistent with the faculty learning objectives.
(2) Problems should be appropriate to the level of the students’ understanding.
(3) Scenarios should have sufficient intrinsic interest for the students or relevance to future practice.
(4) Basic science curriculum should be presented in the context of clinical scenario to
encourage integration of knowledge.
(5) Scenarios should contain cues to stimulate discussion and promote participation in seeking information from various resources.
(6) The problem should be sufficiently open, so that discussion is not curtailed too early in the process.
As for how the task-based learning strategy lead learners to integrate knowledge through ill-structured problem-solving, Lohman et al. (2002) referred to four events, including (1) discussion of the facts – what is known about the problem, (2) information gaps – what information is needed but not know, (3) hypotheses – a list of possible causes or explanations of the problem, and (4) learning issues – areas where learners lack knowledge. Wood (2003) again stated a curriculum module (Figure 2-2) which included mixed teaching methods (including problem-based learning, PBL). As shown in Figure2-2, the staff of PBL could be a tutor or as facilitators for small group learning. Besides, students need to have sufficient time to achieve the learning outcomes in knowledge, skills, and attitudes.
Figure 2-2. Designing and implementing a curriculum module using PBL (cited from Wood, 2003)
Define learning outcomes for module
How will individual outcomes be achieved?
Metacognitive skills
Lectures PBL Communication
skills
Practicals
How many PBL sessions in the module?
Write learning objectives for each PBL
Write PBL scenarios
Write tutor notes
Pilot with staff
Implement module
Evaluate
Refine scenarios and tutor notes
Pilot with group of students
Design timetable for module;
write module handbook for students
The present study designed a programming task which conforms to the demands for the context of task-oriented learning strategy. This task was to ask students to accomplish a computer game in Visual Basic programming language. In the process of the activity, instructional system would guide them to decompose the task into subtasks
and give them some cues and resources as feedback. The learners were expected to pick up the skills of integrating knowledge effectively.
Individualized Learning
Individualized Learning
Individualized instruction was rising and flourishing in the 1960s. Heathers (1977) stated that “individualized instruction consists of any steps take in planning and conducting programs of studies and lessons that suit them to the individual student’s learning needs, learning readiness, and learner characteristics or ‘learning style’”
(p.342). Teachers have to vary or modify materials and procedures of the instructional setting which will accommodate a diversity of students (Anderson, 1979; Wang &
Lindvall, 1984). Generally, individualized instruction or personalized instruction is used interchangeably with adaptive instruction (Park & Lee, 2003; Wang & Lindvall, 1984). Como and Snow (1986) claimed that instructional approaches and techniques which are geared to meet the needs of the individual student are called adaptive instruction. On the contrary, individualized learning means allowing the learner to take the instructional setting in accord with his/her learning pace, sequence and content for achieving mastery learning (Anderson, 1979). For describing more precisely, the goal of individualized learning or adaptive learning is delivering the right content to the right person at the proper time in the most appropriate way (National Association of State Boards of Education Study Group [NASBE], 2001); that is, time, place, path and pace are all important. Hence, individualized learning and individualized instruction
are obviously two opposite sides of an organic whole.
Fletcher (1992) has defined individualization of pace, sequence, and content respectively. First, individualized pace controls the rate at individual learner’s progress through instructional content. It could be implemented by allowing learners to proceed as rapidly as they can or as slowly as they wish. There is also another way to carry out individualized pace by presenting different numbers of items to different learners. For example, learners would pass to the next lesson if they really achieve the criteria of mastery learning; otherwise the learner has to take the supplementary instruction.
Second, individualized sequence indicates that individual learners determine the sequence of topics for themselves, such as theory-based model (theory → examples
→ practice) or example-based model (examples → theory → practice), and so forth
(Magoulas, Papanikolaou, & Grigoriadou; 2003). That is, the components of the course material should be sequenced according to the characteristics and needs of the particular learner until the topic is mastered, and the topic selection process is then repeated until all nodes have been mastered (Figure 2-3) (Govindasamy, 2002; Shute &
Towle, 2003). Third, individualization of content is usually implemented in a diagnostic-prescriptive method. Based on the results of the assessment, content is diagnostically adjusted to the learners by skipping the component entirely, taking supplementary material before working on it, or viewing the component from one or
more perspectives.
Figure 2-3. Individualized learning ensure students achieving the criteria of mastery through the gateway of assessment (revised from Govindasamy, 2002)
Take the whole of current unit
Take part of
current unit Skip current unit
Next unit
Summative Assessment Prerequisite Test
Formative Assessment
Take more basic unit of prerequisite knowledge otherwise
pass pass
Formative Assessment
finish
Based on the above beliefs, individualized learning will give students more options, provide sustained feedback to them, and inspect the learner’s progress frequently. The ultimate objectives of individualized learning attempt to strengthen an individual’s skills to meet the demands of available educational opportunities and adapt effectively to a variety of situations in the complex world (Glaser, 1977;
Anderson, 1979).
Individualized Learning and Mastery Learning
For adapting instructional process to the student, a number of individualized systems were developed accommodate different student abilities. The following list present six modes of individualization according to the priority in learning content, learning sequence, learning rate, and so forth (Anderson, 1979; Fletcher, 1992; Glaser, 1977; Heather, 1977).
(1) Different students can work on different learning tasks toward different goals.
(2) Different students can use different learning materials or equipment in working toward the same goals, e.g. Program for Learning According to Needs (PLAN);
(3) Different students can study a given learning task in different types of individual or group settings, e.g. Individually Guided Education (IGE).
(4) Different students can work on a given learning task with use of different methods of teaching or learning, e.g. Learning for Mastery (LFM).
(5) Different students can be assigned to different teachers to produce effective student/teacher match-ups.
(6) Different students can be allowed different amounts of time as needed to complete a learning task, such as the systems of Individually Prescribed Instruction (IPI), Adaptive Learning Environment Model (ALEM), IGE, and PLAN.
In order to give consideration to both sides of learning method and learning rate
(the 4th and the 6th mode of individualization), and emphasize whether the student’s learning achieves mastery, the present study created the flexible learning environment which ensured students achieving the criteria of mastery through learning in their own paces. However, what is mastery learning? Bloom (1974) argued that students can master a great deal of what they are taught in school if the instruction is approached systematically, if students are helped when and where they have learning difficulties, if they are given sufficient time to achieve master, and if there is some clear criteria of what constitutes mastery. The goal of learning for mastery tends to establish at least the two lowest levels of Bloom’s taxonomy, that is to say, knowledge and comprehension (Anderson, 1979), while Guskey (1985) stated that mastery standard is applicable between 80% and 90% correct in most instructional situations except the objectives are so important and so critical for learning that demands 100% mastery.
From a psychological point of view, for the student, mastery learning gives more learners the sense of pride and well-being. When successful in learning, the student would feel more confident in future learning activities (Charles, 1976; Guskey, 1985).
For the teacher, mastery-referenced instruction reduces the burden of remedial instruction and offers heightened satisfaction resulting from success with students who previously had been failing (Heathers, 1977). Thus, the use of mastery learning is able to provide more powerful and positive influence on learning. Mastery learning process
could not only pinpoint individual learning problems but also help students overcome their specific difficulties.
Implementing Individualized Learning in CAL
Glaser (1977) has argued that students’ learning preferences and teachers’
instructional strategies are actually variable, so that the development and implementation of individualized learning program are complex and difficult in the existing environment. Thus, how to make the instructional setting suit to the learner is the key problem. Nevertheless, the education medium so-called computer-assisted learning (CAL) or computer-assisted instruction (CAI) was provided with the unique strength that the pace of the presentation could be generally controlled by the student.
The computer presents parts of the content for an appropriate amount of time and then pauses. After the student presses the key, the remainder would be shown continuously.
Moreover, individualization may have a pretest and several formative assessments. The CAL system is able to diagnose the results of tests and ascertain the student’s status automatically. Mcmanus (2000) has claimed that technology-based learning systems could offer educators the ability to individualize instruction for learners consistently and automatically. Park and Lee (2003) also suggested that computer-assisted instructional systems allow learners making their own path in
learning to comply with the needs of planning and implementing of adaptive instructional systems more easily. In fact, the reason of these functions was that the one-to-one nature of CAI makes it possible to monitor student understanding constantly and to respond based on the needs of each individual student efficiently (Ross, 1984).
As regards implementing an individualized learning system, Wu and Lin (1977) stated that the instructional design of individualized learning usually consists of six important elements: (1) deciding the instructional objectives, (2) selecting and editing the content of course materials, (3) setting an individualized learning environment and learning resources, (4) rendering individualized learning, (5) carrying out the assessment, and (6) revising the defect of initial instructional design. Heathers (1977) has brought up the typical steps of designing individualized instruction model, which are revealed in the following.
(1) Determine what learning task the student needs to accomplish next in the curriculum.
(2) Assess the extent to which the student already has mastered and place him/her at appropriate points.
(3) Diagnose the student in terms of learning readiness and style to match him/her with appropriate instructional modes or procedures, and determine how best he/she can
work on the task.
(4) Provide the student with help as needed in performing the learning task.
(5) Assess the student’s performance of the lesson to determine whether mastery has been reached. If the student does not achieve the standard of mastery, providing supplementary instruction in order to correct his/her errors and misunderstandings.
Anderson (1979) then described three characteristics of good adaptive curricula.
First, the cognitive goals beyond recall of knowledge focusing on the development of higher information processes should be included. Second, the curricula must help students to learn and develop their self-confidence when they have difficulties. Third, adaptive instructional curricula should present explicit goals and explicit performance standards. With explicit goals, students can know what they learn; with explicit performance standards, students can learn independently. Gates, Lawhead, and Wilkins (1998) also suggested that an adaptive hypermedia system should (1) integrate information from heterogeneous sources into a unified system, (2) provide a filtering mechanism so that users interact with a view which matches their needs, (3) deliver the information through the computer-based interface, and (4) support the automatic creation to help with ongoing maintenance of the application.
According to these important instructional design components, instructors can understand the meaning of individualized learning certainly and can establish an
individualized computer-assisted instructional system in a well-regulated way.
Individualized CAI systems not only suit the learner and enhance the learning performance (Shute, Lajoie, & Gluck, 2000), but also allow students to adapt effectively to a variety of situations (Anderson, 1979). Thus, an adaptive hypermedia instructional system should give a instructional presentation adapted to the learner’s knowledge of the subject matter and a suggested set of the most relevant links for the student to pursue, rather than gives all students the same information and the same links (Park & Lee, 2003).
Summary
Since the flow paths of programming and debugging are consistent with the flow path of problem-solving, a student would learn the skills of programming when he or she grasps the useful skills of critical thinking and problem solving (Felleisen et al., 2004). Furthermore, the novice programmers just lack the problem solving strategies and detailed mental models of program design. In contrast to experts, novices do not have stronger knowledge of programming language and do not know how to use problem-solving strategies agilely (Mayer, 1992; Robins et al., 2003).
In order to make students understand the purposes of programming consist in problem-solving rather than just coding, the programming instruction of this study was integrated with the problem-solving strategy by means of task-oriented learning for students to be familiar with the steps and skills of problem-solving. The experimental instruction also provided conceptual models to dynamically visualize the flowchart of programs for enhancing the students’ program comprehension performance.
Besides, because individual differences also affect program comprehension strategies and performance (Ko & Uttl, 2003), this study hoped to permit every student acquiring adaptive, efficient, and mastery learning by the self-paced individualized learning sequence.
