the clia-model: a framework for designing powerful learning environments for thinking and problem...

A major challenge for education and educational research is tobuild on our present understanding of learning for designingenvironments for education that are conducive to fostering in studentsself-regulatory and cooperative learning skills, transferable knowledge,and a disposition toward competent thinking and problem solving.Taking into account inquiry-based knowledge on learning and recentinstructional research, this article presents the CLIA-model(Competence, Learning, Intervention, Assessment) as a framework forthe design of learning environments aimed to be powerful in eliciting instudents learning processes that facilitate the acquisition of productiveknowledge and competent learning and thinking skills. Next, twointervention studies are described that embody major components ofthis framework, one focussing on mathematical problem solving inprimary school, and a second one relating to self-regulatory skills inuniversity freshmen. Both studies were carried out in parallel with thedevelopment of the framework, and were instrumental in identifyingand specifying the different components of the model. They yielded bothpromising initial support for the model by showing that CLIA-basedlearning environments are indeed powerful in facilitating in studentsthe acquisition of high-literacy learning results, especially theacquisition and transfer of self-regulation skills for learning andproblem solving.

Introduction

For a large part of the past century the major mission of schools consisted in teachinglow-literacy skills, namely reading, writing, and calculating. Changes in society during the latepart of the 20th century have induced a growing need for the acquisition by all citizens of

European Journal of Psychology of Education2004, Vol. XIX, nº 4, 365-384© 2004, I.S.P.A.

The CLIA-model: A framework for designingpowerful learning environments for thinking andproblem solving

Erik De CorteLieven VerschaffelUniversity of Leuven, Belgium

Chris MasuiLimburg University Center, Belgium

aspects of high literacy, such as thinking (critically), solving complex problems, regulatingone’s own learning, and communication skills (National Research Council, 2000). However, ithas repeatedly been observed that education has not been able to keep up with thesedevelopments. For instance, in a report of the European Round Table of Industrialists (ERT)(1995) entitled Education for Europeans. Towards the learning society, a cry of alarm wasraised to alert society to the so-called educational gap, i.e., the fact that – due to its slowness inresponding to changes in society – there is “an ever-widening gap between the education thatpeople need for today’s complex world and the education they receive” (ERT, 1995, p. 6).This problem is even increasing because recently the pace of societal developments hasaccelerated dramatically due, among others, to the exponential knowledge explosion, toglobalization in many domains of society, and to the large-scale introduction of the newinformation technologies.

A major challenge for education and educational research is, then, to build on our betterunderstanding of productive learning for the design of novel environments that are conduciveto fostering in all students self-regulatory and collaborative learning skills, productive andtransferable knowledge, and a disposition toward competent thinking and problem solving.

Over the past decades researchers have taken up this challenge by designing andevaluating instructional environments that are aimed to be powerful in eliciting in studentslearning processes that facilitate the acquisition of productive knowledge and competentlearning and thinking skills. In this article a selection of recent perspectives relating to theissue of developing effective learning environments will be outlined briefly as background forthe presentation of a framework for the design of learning communities. This will be followedby a description of two intervention studies from our Center that are based on and embodymajor components of this framework, one focussing on learning and teaching mathematicalproblem solving in the upper primary school, and a second one relating to the acquisition ofself-regulatory skills in university freshmen.

Design of learning environments: Recent developments

Paralleling the shift in the goals of education from low-literacy to high-literacy,substantial advancement has been made in our understanding of cognition and learning (seee.g., Brown, 1994; Bruer, 1993; Mayer, 2001; National Research Council, 2000). Whereas“there is more to education than cognition” (Bruer, 1993, p. 2), and “learning theory does notprovide a simple recipe for designing effective learning environments” (National ResearchCouncil, 2000, p. 131), it is nevertheless the case that our current understanding of theprocesses of learning and development can guide the design of more powerful instructionalsettings to facilitate in students the acquisition of a disposition toward skilled learning,thinking, and problem solving. There is at present in the field of research on learning andinstruction a fairly broad consensus about the following definition of effective learning: it is aconstructive, cumulative, self-regulated, goal-directed, situated, collaborative, andindividually different process of meaning construction and knowledge building (De Corte,1996).

Taking into account the changes in the goals of education as well as the advances inresearch on learning, several related frameworks for the design of powerful learningenvironments have been proposed since the late 1980s. Based on an analysis of threesuccessful instructional intervention studies, namely Palincsar and Brown’s (1984) reciprocalteaching of reading comprehension, Scardamalia and Bereiter’s (see Scardamalia, Bereiter, &Steinbach, 1984) procedural facilitation of writing, and Schoenfeld’s (1985) heuristic teachingof mathematical problem solving, Collins, Brown, and Newman introduced in 1989 thecognitive apprenticeship model of learning and teaching. The model describes thecharacteristics of ideal learning environments in terms of four components: content, teachingmethods, sequence of learning tasks, and social context of learning strategies.

366 E. DE CORTE, L. VERSCHAFFEL, & C. MASUI

DESIGNING POWERFUL LEARNING ENVIRONMENTS 367

One of the most representative examples of a research-based attempt at changing theclassroom environment in line with the cognitive apprenticeship framework is the “FosteringCommunities of Learners” (FCL) project initiated and elaborated by Brown and Campione(1994, 1996, see also Brown, 1994). Basically the FCL system encompasses three keycomponents: students engage in groups actively in research on an aspect of an interestingtheme or a problem that in the end should be mastered by all pupils of the class. Therefore,when finishing their research they share their acquired knowledge with their fellow learners.To assess mastery all pupils of the class are then given a consequential task. The goal of theseresearch-share-perform cycles that are accompanied by continuous reflection by all pupils, isto bring them to understand deep disciplinary content. Key principles of learning that underlyFCL are: cycles of research-share-perform; a metacognitive, reflective environment; centralityof discourse and dialogue; focus on deep domain-specific knowledge; distributed expertise;instruction and assessment aligned; and community of practice, also beyond the classroomwall (Brown & Campione, 1996). Using so-called design experiments the Brown group hasalready reported quite substantial learning and transfer results of the FCL project (see e.g.,Brown & Campione, 1994; Campione, Shapiro, & Brown, 1995; for an extensive discussionof design experiments in educational research, see the theme issue of the EducationalResearch edited by Kelly, 2003).

A second example of a research-based program aiming at fundamentally changingtraditional classrooms is the so-called Jasper-project developed by the Cognition andTechnology Group at Vanderbilt (CTGV). A basic idea of the project is “AnchoredInstruction”, presented by the CTGV as one way of implementing cognitive apprenticeship:instruction is anchored in videodisc-based meaningful problem-solving environments (e.g.,planning a trip focusing on the concepts “distance/rate/time”) that provide a richer, moreauthentic, and more dynamic presentation of information than textual material (Cognition andTechnology Group at Vanderbilt, 1990). The stories on the videodiscs (such as TheAdventures of Jasper Woodbury) create an environment for active and cooperative learningand discussion in small groups as well as for individual and whole-class problem solving (forextensive reports on the project see Cognition and Technology Group at Vanderbilt, 1997,2000).

In a volume entitled How people learn: Brain, mind, experience, and school (NationalResearch Council, 2000) an extensive report is given of a two-year study during which aCommittee on Developments in the Science of Learning of the National Research Council(NRC) in the U.S.A. evaluated and synthesized a large amount of research on learning andteaching, complemented with the results of a second NRC Committee on Learning Researchand Educational Practice. Taking into account the available inquiry-based knowledge onlearning, and heavily drawing upon the ideas and findings of the two projects briefly discussedabove, this volume presents a framework for the design of learning environments involvingfour interconnected basic features: effective learning environments are learner-centered,knowledge-centered, assessment-centered, and community-centered; moreover, because of theinterconnectedness these four perspectives need to be aligned in ways that mutually supportone another. Learner-centered environments help students construct their knowledge andskills building on the understandings, beliefs, and cultural practices that they bring to theschool. Knowledge-centered environments help students to acquire well-organized bodies ofdomain knowledge that support future thinking and learning, and include also an emphasis onhelping students learn to monitor and regulate their own learning. Assessment-centeredenvironments provide opportunities for feedback that yield relevant information for improvingteachers’ instruction and students’ learning, and help students develop skills of self-assessment. Community-centered environments establish norms that support people’s abilitiesto learn from one another; these norms pertain not only to the class as a community, but alsoto the school as a whole as well as to links between the school and the broader community.

The preceding overview shows that since the late 1980s the concept of effective orpowerful learning environments has progressively been enlarged by stressing more explicitlythe important role of assessment and by adding the broader community dimension. However,

the backbone of the notion has remained, namely the focus on fostering and mediatingconstructive, self-regulated, and collaborative learning anchored in rich and meaningfulcontexts, and aiming at the acquisition of “deep” conceptual domain knowledge and higher-order thinking and learning skills.

CLIA: A framework for designing powerful learning environments

Our own work relating to the creation of instructional settings that facilitate in studentsthe acquisition of productive knowledge and learning and thinking skills has paralleled theresearch discussed in the previous section, and has also been influenced by the cognitiveapprenticeship model, the FCL and the Jasper-projects. This section presents the frameworkfor the design of learning communities that has resulted from our theoretical and empiricalstudies. Over the years this framework has also developed, esp. by highlighting moreexplicitly than in the projects reviewed above the importance of beliefs and affective aspectsfor learning, and also by stressing the need to modify the (traditional) classroom culture.

The framework for designing learning environments that are intended to be powerful, isstructured according to four interconnected components:

1. Competence: Components of competence in a domain.2. Learning: Characteristics of effective learning processes.3. Intervention: Principles and methods guiding the design of learning environments.4. Assessment: Forms of assessment for monitoring and improving learning and teaching.

These four components have been deliberately chosen building on the related views ofGlaser (1976), Resnick (1983), and Snow and Swanson (1992) concerning the core elementsof a theory of learning from instruction. As argued by Resnick (1983), such a theory must be“both descriptive, explaining why instruction works and why it does not, and prescriptive,what to do the next time for better results” (p. 6). In that perspective the theory must conformto several requirements. First, it must specify the objectives of instruction, thus thecompetence to be attained. Second, it should provide a theoretical account of the learningprocesses needed to acquire competence. Third, it should specify guiding principles forinstructional interventions to support those learning processes. In addition it is necessary toassess the outcomes of the interventions (Glaser, 1976; Snow & Swanson, 1992). As thesefour components are narrowly interconnected they need of course to be aligned.

Competence

Acquiring competence in a domain requires the acquisition of five categories ofcomponents: cognitive ones, on the one hand, and conative components involving motivationand volition, on the other hand. (see e.g., De Corte, Greer, & Verschaffel, 1996; Snow, Corno,& Jackson III, 1996; Corno, Cronbach, Kupermintz, Lohman, Mandinach, Porteus, et al., 2002).

1. A well-organized and flexibly accessible domain-specific knowledge base involvingthe facts, symbols, concepts, and rules that constitute the contents of a subject-matter field.

2. Heuristics methods, i.e., search strategies for problem analysis and transformation (e.g,decomposing a problem into subgoals, making a graphic representation of a problem)which do not guarantee, but significantly increase the probability of finding the correctsolution of a problem because they induce a systematic approach to the task.

3. Metaknowledge, involving knowledge about one’s cognitive functioning(metacognitive knowledge: e.g., believing that one’s cognitive potential can bedeveloped and improved through learning and effort), on the one hand, and



knowledge about one’s motivation and emotions that can be used to deliberatelyimprove volitional efficiency (e.g., becoming aware of one’s fear of failure inrelation to mathematics tasks and problems), on the other hand.

4. Self-regulatory skills, involving skills relating to regulating one’s cognitive processes/activities (metacognitive skills or cognitive self-regulation; e.g., planning andmonitoring one’s problem-solving processes), on the one hand, and skills forregulating one’s volitional processes/activities (metavolitional skills or volitionalself-regulation; e.g., keeping up one’s attention and motivation to solve a givenproblem), on the other hand.

5. Positive beliefs about the self in relation to learning and problem solving in adomain, about the social context in which learning activities take place, and aboutthe content domain and learning and problem solving in that domain.

However, research has shown that knowledge and skills that are available in students areoften accessible nor usable when necessary to solve a given problem (Cognition andTechnology Group at Vanderbilt, 1997). Acquiring a disposition to skilled learning andthinking should help to overcome this phenomenon of inert knowledge. Therefore, theintegrated mastery of the different components mentioned above should result in thedevelopment of a disposition toward skilled thinking and learning. According to Perkins(1995) such a disposition involves besides ability and motivation, two additional crucialaspects, namely sensitivity for situations in which it is relevant and appropriate to use acquiredknowledge and skills, and an inclination to do so. Perkins (1995) argues that this sensitivityfor situations and contexts, and the inclination to follow through are both fundamentallydetermined by the beliefs a person holds. For instance, a person’s beliefs about what counts asa mathematical context and what he or she finds interesting or important have a strong impacton the situations he or she is sensitive to and whether or not he or she engages in them.

Learning

Although important questions remain for continued inquiry (De Corte, 2004), the followingcharacteristics of productive learning are already well documented by a substantial amount ofresearch: it is an active/constructive, cumulative, self-regulated, goal-directed, situated,collaborative, and individually different process of meaning construction and knowledge building(De Corte, 1996; see also Mayer, 2001; National Research Council, 2000; Shuell, 1988).Therefore, these features of effective learning can and should guide educational practice.

1. Active/constructive: learning is an effortful and mindful process in which studentsactively construct their knowledge and skills through reorganization of their alreadyacquired mental structures in interaction with the environment.

2. Cumulative: this characteristic stresses the important impact of students’ priorformal as well as informal knowledge on subsequent learning.

3. Self-regulated: this feature refers to the metacognitive nature of productive learning;indeed, self-regulation of learning means that students manage and monitor theirown processes of knowledge building and skill acquisition. The more studentsbecome self-regulated, the more they assume control and agency over their ownlearning; consequently they become less dependent on external instructional supportfor performing those regulatory activities.

4. Goal-oriented: effective and meaningful learning is facilitated by an explicitawareness of, and orientation toward a goal. Because of its constructive and self-regulated nature, it is plausible that learning will be most productive when studentschoose and determine their own objectives. Therefore, it is desirable to stimulate andsupport goal-setting activities in students.

5. Situated and collaborative: learning is conceived as an interactive activity betweenthe individual and the physical, social and cultural context and artefacts, and

especially through participation in cultural activities and contexts. In other words,learning is mostly not a purely “solo” activity, but a distributed one: the learningeffort is distributed over the individual student, his partners in the learningenvironment, and the resources and (technological) tools that are available.

6. Individually different: the processes and outcomes of learning vary among studentsdue to individual differences in a diversity of aptitudes that affect learning, such asprior knowledge, conceptions of learning, learning styles and strategies, interest,motivation, self-efficacy beliefs, and emotions. To induce productive learning instudents instruction should take into account these differences in aptitudes.

Intervention

Taking into account the literature reviewed in the preceding section, the following majorguiding principles for the design of powerful learning environments can be derived from ourpresent conception of competence (first component of CLIA), on the one hand, and thecharacteristics of constructive learning (second component of CLIA), on the other. This showsat the same time the interrelatedness of the CLIA-components and the necessity to align them.

1. Learning environments should initiate and support active, constructive acquisitionprocesses in all students, thus also in the more passive learners. However, the viewof learning as an active process does not imply that students’ construction of theirknowledge cannot be guided and mediated through appropriate interventions such asmodeling, coaching, and scaffolding (Collins et al., 1989) by teachers, peers, andeducational media. Indeed, the claim that productive learning involves goodteaching still holds true. In other words, a powerful learning environment ischaracterized by a good balance between discovery and personal exploration, on theone hand, and systematic instruction and guidance, on the other, always taking intoaccount individual differences in abilities, needs, and motivation among learners.

2. Learning environments should foster the development of self-regulation strategies instudents. This implies that external regulation of knowledge and skill acquisitionthrough systematic intervention should be gradually removed so that students becomeagents of their own learning. In other words, the balance between external and internalregulation will vary during students’ learning history in the sense that progressivelythe share of self-regulation grows as explicit instructional support fades out.

3. Because of the importance of context and collaboration for effective learning,powerful learning environments should embed students’ constructive acquisitionactivities preferably in real-life situations that have personal meaning for the learners,that offer ample opportunities for distributed learning through social interaction, andthat are representative of the tasks and problems to which students will have to applytheir knowledge and skills in the future. But stressing the importance of socialinteraction for productive learning does not exclude the individual acquisition ofcomponents of competence (Salomon & Perkins, 1998).

4. Because domain-specific knowledge, heuristic methods, metaknowledge, self-regulatory skills and beliefs play a complementary role in competent learning,thinking, and problem solving, learning environments should create opportunities toacquire general learning and thinking skills embedded in the subject-matter fields.

5. Powerful learning environments should create a classroom climate and culture thatinduces in pupils explicitation of and reflection on their learning activities andproblem-solving strategies. For instance, Berry and Sahlberg (1996) argue that in orderto modify pupils’ ideas about learning in the direction of De Corte’s model describedabove, it is necessary to develop their conceptual metacognitive understanding aboutlearning through reflective practices and dialogues with peers in small groups.

6. Learning environments should allow for the flexible adaptation of the instructionalsupport, especially the balance between self-regulation and external regulation, in



order to take into account the individual differences in aptitudes among learners. Inaddition, the important impact of affective characteristics, especially emotions, onstudents’ learning activities and outcomes points to the necessity of alternatinginstructional interventions with emotional support, depending on whether theindividual student is in the learning or in the coping mode (Boekaerts, 1993).

Assessment

Forms and methods of assessment should be aligned with the preceding components ofthe CLIA framework, and integrated with instruction. This implies that classroom assessmentsshould satisfy the following conditions (see also Glaser & Silver, 1994; Shavelson & Baxter,1992; Shepard, 2001).

1. Assessment instruments should address and monitor students’ progress toward theacquisition of the full range of aspects of the competence component of the CLIA-model, i.e., the different kinds of knowledge, skills, and beliefs.

2. Assessment instrument should provide diagnostic feedback about students’ deepunderstanding of content and their mastery and productive use of learning andthinking skills, which is helpful for students and teachers in view of further learningand instruction. In that perspective assessment tools should not only address learningoutcomes but trace also students’ learning processes and strategies.

3. The conception of learning outlined above (the second component of the CLIA-model) also implies that alternative assessment forms should contain assignmentsthat are meaningful for the learners, and that offer opportunities for self-regulatedand collaborative – besides individual – approaches to tasks and problems.

4. Assessment practices should help students develop skills in individual and groupself-assessment.

In the remaining part of this article two intervention studies from our own Center will bebriefly reviewed in which major aspects of the CLIA framework were implemented: the firstinvestigation relates to mathematical problem solving in the upper primary school; the secondstudy pertains to fostering metaknowledge and self-regulatory skills in university freshmen inbusiness economics. Both investigations were not designed as formal tests of the CLIA model,but were carried out in parallel with the development of the framework, and were as suchinstrumental in identifying and specifying the different components of the model as describedabove. As such this approach is in line with the perspective of the Design-Based ResearchCollective (2003; see also Burkhardt & Schoenfeld, 2003) on the potential of interventionresearch, namely exploring possibilities for novel learning and teaching environments, anddeveloping contextualized theories of learning and teaching. In the following review we willfocus especially on the CLIA aspects involved in both studies.

A learning environment for mathematical problem solving in the upper primary school

In the Flemish part of Belgium new standards for primary education became operationalin the school year 1998-1999 (Ministerie van de Vlaamse Gemeenschap, 1997). With respectto mathematics these standards stress more than before the importance of mathematicalreasoning and problem-solving skills and their applicability to real-life situations, as well asthe development of positive attitudes and beliefs toward mathematics. As a contribution to theimplementation of those new standards we carried out a research project – commissioned bythe Department of Education of the Flemish government – aiming at the design and evaluationof a powerful learning environment, that elicits in upper primary school children theappropriate learning processes for acquiring the intended competence in mathematicalproblem solving as well as positive mathematics-related beliefs.

The learning environment in the classroom was fundamentally changed, and its design,implementation, and evaluation were done in close cooperation with the teachers of the fourparticipating experimental classrooms and their principals. The intervention consisted of aseries of 20 lessons taught by the regular classroom teachers (for a detailed report about thisstudy see Verschaffel, De Corte, Lasure, Van Vaerenbergh, Bogaerts, & Ratinckx, 1999). Interms of the CLIA- framework the learning environment can be described as follows.

Competence

The learning environment focused on the acquisition by the pupils of an overall cognitiveself-regulation strategy for solving mathematical problems consisting of five stages, andembedding a set of eight heuristic strategies which are especially functional in the first twostages of that strategy (see Table 1). Acquiring this strategy involves: (1) becoming aware ofthe different phases of a competent problem-solving process (awareness training); (2)becoming able to monitor and evaluate one’s actions during the phases of the solution process(self-regulation training); and (3) gaining mastery of the heuristic strategies (heuristic strategytraining). In addition fostering positive mathematics-related beliefs was aimed at.

Table 1The competent problem-solving model underlying the learning environmentStep 1: Build a mental representation of the problem

Draw a pictureMake a list, a scheme or a tableHeuristics: Distinguish relevant from irrelevant dataUse your real-world knowledge

Step 2: Decide how to solve the problem Make a flowchartGuess and checkHeuristics: Look for a patternSimplify the numbers

Step 3: Execute the necessary calculations Step 4: Interpret the outcome and formulate an answerStep 5: Evaluate the solution

Learning and intervention

To elicit and support constructive learning processes in all pupils, the six above-mentioned principles for the design of learning environments were applied in an integratedway. This led to an intervention that was characterized by the following three basic features.

First, a varied set of carefully designed meaningful, complex, and open problems wereused that differ substantially from the traditional textbook tasks. Moreover, these problems werepresented in different formats: a text, a newspaper article, a brochure, a comic strip, a table, or acombination of several of these formats. An example is described below. The self-regulationskills and the heuristic strategies were thus fostered embedded in the mathematical content.

School trip problem1

The teacher told the children about a plan for a school trip to visit the Efteling, a well-knownamusement parc in The Netherlands. But if that would turn out to be too expensive, one of the otheramusement parcs might be an alternative.Each group of four pupils received copies of folders with entrance prices for the different parcs.The lists mentioned distinct prices depending on the period of the year, the age of the visitors, andthe kind of party (individuals, families, groups).



In addition, each group received a copy of a fax from a local bus company addressed to theprincipal of the school. The fax gave information about the prices for buses of different sizes (witha driver) for a one-day trip to the Efteling.The first task of the groups was to check whether it was possible to make the school trip to theEfteling given that the maximum price per child was limited to 12.50 euro.After finding out that this was not possible, the groups received a second task: they had to find outwhich of the other parcs could be visited for the maximum amount of 12.50 euro per child.

Second, a learning community was created through the application of a varied set ofactivating and interactive instructional techniques. The basic instructional model for eachlesson period consisted of the following sequence of activities: (1) a short whole-classintroduction; (2) two group assignments solved in fixed heterogeneous groups of three to fourpupils, each of which was followed by a whole-class discussion; (3) an individual task also with asubsequent whole-class discussion. Throughout the lesson the teacher encouraged and scaffoldedpupils to engage in, and to reflect upon, the kinds of cognitive and metacognitive activitiesinvolved in the model of skilled problem solving. These instructional supports were graduallyfaded out as pupils became more competent in and aware of their problem-solving activity,and, thus, took more responsibility for their own learning and problem-solving processes.

Third, an innovative classroom culture was created through the establishment of new socialnorms about learning and teaching problem solving, and aiming at fostering positive mathematics-related attitudes and beliefs in children. Typical aspects of this classroom culture are: (1)stimulating pupils to articulate and reflect upon their solution strategies, (mis-)conceptions,beliefs, and feelings relating to mathematical problem solving; (2) discussing about what countsas a good problem, a good response, and a good solution procedure (e.g., “for some problems arough estimate is a better answer than an exact number”): (3) reconsidering the role of the teacherand the pupils in the mathematics classroom (e.g., “the class as a whole, under the guidance ofthe teacher, will decide which of the generated solutions is the optimal one after an evaluation ofthe pros and cons of the different alternatives”).

The learning environment was elaborated in partnership with the teachers of theparticipating experimental classes and their principals. The model of teacher developmentadopted emphasized the creation of a social context wherein teachers and researchers learn fromeach other through continuous discussion and reflection on the basic principles of the learningenvironment, the learning materials developed, and the teachers’ practices during the lessons.To enhance a reliable implementation of the learning environment, teachers received the supportmaterials, especially a general teaching guide containing an extensive description of the aims,content, and structure of the learning environment, and a list of ten guidelines comprisingactions that they should take before, during and after the individual or group assignments.

In the period of four months during which the intervention was implemented in theexperimental classes, pupils from seven control classes followed the regular maths curriculum.These classes were comparable with the experimental classes in terms of ability and SES, andduring the four month period they followed an equal amount of lessons in word problemsolving. Interviews with the teachers of these classes and analyses of the textbooks usedindicated that the teaching of word problems was representative of current instructionalpractice in Flemish elementary schools (De Corte & Verschaffel, 1989).

Assessment

Students’ progress toward the major goals of the learning environment was assessedsummatively using a variety of instruments. Three parallel versions of a written test consistingof ten difficult nonroutine problems were used as pretest, posttest and retention testrespectively. A questionnaire on beliefs and attitudes relating to the teaching and learning ofmathematical word problem solving, and a standardized mathematics achievement test thatcovers the entire mathematics curriculum were both applied as pretest and posttest. In addition,in each of the four experimental classes the solution processes of three pairs of children forfive nonroutine problems were videotaped and analyzed before and after the intervention.

Formative assessment was also substantially built into the learning environment resultinginto diagnostic feedback facilitating informed decision-making about further learning andinstruction. This was mainly obtained as a result of discussions about and reflection onarticulated problem approaches and solution strategies in small groups and in the whole class.

Results

Whereas no significant difference was found between the experimental and controlgroups on the word problem test during the pretest, the former group significantlyoutperformed the latter during the posttest, and this difference in favor of the experimentalgroup continued to exist on the retention test. The effect was .31 (Cohen, 1988).

The learning environment had also a significant, albeit small, positive impact on children’spleasure and persistence in solving mathematics problems, and on their mathematics-relatedbeliefs and attitudes as measured by a self-made Likert-type questionnaire (effect size .04).

The results on a standardized achievement test showed that the extra attention during themathematics lessons for cognitive and metacognitive strategies, beliefs, and attitudes in theexperimental classes did not have a negative influence on the learning outcomes for other,more traditional parts of the mathematics curriculum. To the contrary, there was even asignificant transfer effect; indeed, the experimental classes performed significantly better thanthe control classes on this test (effect size .38).

The analysis of pupils’ written notes on their response sheets of the word problem testshowed that the better results of the experimental children were paralleled by a verysubstantial increase in the spontaneous use of the heuristic strategies taught in the learningenvironment (effect size .76). This finding was confirmed by a qualitative analysis of thevideotapes of the problem-solving processes of three groups of two children from eachexperimental class before and after the intervention.

Furthermore, not only the high and medium ability pupils, but also those of low abilitybenefited significantly – albeit to a smaller degree – from the intervention in all aspects justmentioned. Indeed, there was no significant interaction between the variables experimentalgroup, time of test, and ability level.

As was derived from a number of videotaped lessons in each experimental class, all fourteachers implemented the intervention in an appropriate way. Taking this into account thepreceding results show that a CLIA-based intervention, combining a set of carefully designedword problems with highly activating and interactive teaching methods, and the introductionof new social classroom norms, can lead to the creation of a powerful learning environmentwhich significantly boost pupils’ cognitive and metacognitive competency in solvingmathematical word problems.

A learning environment for fostering metaknowledge and self-regulatory skills inuniversity freshmen

At the outset of the third millennium, higher education in Europe is facing several majorproblems. Firstly, universities have to adjust to larger and more heterogeneous populationsthan in the past. Secondly, certainly in Flemish tertiary education the output of studentscompleting a degree is largely insufficient. And, last but not least, there is an urgent need forgraduates who are prepared for lifelong learning. In response to these challenges we carriedout a research project aiming at the design, implementation, and evaluation of a powerfullearning environment for fostering learning competence in beginning university students (for adetailed report about this study see Masui, 2002; Masui & De Corte, 1999; Masui & De Corte,in press). Thereby we took into account previous research (Masui, Borremans, Van Damme, &Vandenberghe, 1986), as well as the growing knowledge base about self-regulated learning(see e.g., Schunk & Zimmerman, 1998; Vermunt, 1995).



Starting from a constructivist perspective on learning, several aspects of expertise,namely metacognitive, affective, and conative skills and related metaknowledge, wereintegrated into the real instructional context of an experimental group (E) of 47 first yearstudents in business economics. The intervention focused on the acquisition in students ofeight regulatory skills that were taught in a series of ten sessions of 90 minutes each,supplemented by numerous homework assignments aimed at practising and transferringknowledge and skills. As explained below, this learning environment likewise embodiesnumerous components of the CLIA framework. The study involved also two control groups of47 students each: in the first control group (C1) a treatment was applied that focused oncognitive activities such as ‘analyzing’ and ‘rehearsing’; the second control group (C2) was anon-treatment group. All students in the three groups were selected from the total group offreshmen (N=352) taking into account several entrance characteristics (prior academicknowledge, intelligence, cognitive study skills, attribution behavior, self-judgements aboutexecutive regulation activities, and gender). E and C1 were independent, but equivalentgroups in terms of average level of intelligence, and prior knowledge; E and C2 were matchedgroups.

Competence

The available literature shows convincingly that metacognitive knowledge and a largevariety of cognitive as well as volitional self-regulation skills have an effect on learningprocesses and outcomes (Masui, 2002). Because research also reveals intimate relationshipsbetween those skills, we opted for a multidimensional approach, i.e., a substantial number ofregulatory activities were addressed integratively in the learning environment. Taking theresearch findings as well as the context of the present study into account, we first selected fourcognitive self-regulation skills, namely ‘orienting’, ‘planning’, ‘self-checking’ and ‘reflecting’.They represent different aspects of metacognitive behavior, which are undoubtedly significantfor freshmen at the university.

Subsequently we chose four matching affective and conative skills. Since ‘orienting’implies evaluating one’s own weaknesses and strengths we firstly chose ‘self-judging’. Nextwe assumed that learning to ‘plan’ was a good occasion to learn making choices or to ‘value’.Thirdly, we included ‘coping with emotions’ as the affective counterpart of ‘self-checking’,and finally ‘reflecting’ seemed to provide sufficient opportunities for learning to ‘attribute’ ina constructive way. There is evidence regarding the effect of all these activities on studyresults in higher education, but a fully integrated approach using these types of skills is mostlylacking in previous training studies.

Learning and intervention

The characteristics of the learning component of the CLIA-framework mentioned abovewere taken as the starting point for developing a learning environment to elicit and stimulatethese learning qualities. Besides, learner-related parameters (esp. prior knowledge),instruction-related aspects (goals, domain content, support), and particular features of theresearch context (e.g., free entrance to the study program on the basis of a certificate ofsecondary education) were considered. Taking into account all these variables, the design ofthe experimental intervention was based on the following integrated set of seveninterconnected and partly overlapping instructional principles.

1. Embed the acquisition of knowledge and skills in the real study context, i.e., theselected activities have to be taught in the context in which students must applythem (situated learning). This principle was realized during the sessions incollaboration with the instruction team for macro-economics and managementaccounting.

2. Take into account the study orientation of the students and their need to experiencethe usefulness of the learning and study tasks (personal usefulness). In this respecteach part of the intervention was explicitly linked to students’ personal goals(especially being successful in their first year).

3. Sequence teaching methods and learning tasks and relate them to a time perspective(sequencing and time perspective). The intervention was spread over a period of sixmonths in which more and more disciplines and more and more complex tasksbecame involved.

4. Use a variety of forms of organisation and social interaction (variation inorganisation and social settings). More specifically a stimulating social environmentwas created by alternating modeling, individual assignments, working in pairs,small-group work, whole-class discussion, and homework.

5. Take into account prior knowledge and large differences between students(adjusting to prior knowledge and differentiating). The application of the third(sequencing) and fourth (variety of social settings) principles contributed to theimplementation of this differentiation.

6. Stimulate articulation of and reflection upon learning and thinking processes(verbalizing and reflecting). Verbalizing problem-solving processes was donemainly by thinking aloud, writing while thinking, and oral or written retrospection.

7. Create opportunities to practice and to transfer learned activities to new contentdomains (practice and transfer). Whereas the intervention focused on the coursesmacro-economics and management accounting, transfer exercises were assigned indifferent other disciplines, especially history and sociology.

The experimental sessions took place in groups of 15 students. A session started with anoverview of the goals to be attained, the activities that were planned, and the kind ofcontribution that was expected from the students. Next, the students made two or moreexercises in macro-economics or management accounting individually or in pairs. After eachassignment they were invited to draw some conclusions, both with regard to the specificcontent and with regard to the problem-solving process. At the end of the session studentsreceived all necessary information about the homework they had to make individually or incollaboration with a fellow student. All experimental sessions were audiotaped.

In the first control group (C1) the focus of the treatment was on cognitive activities. Thisimplied practicing ‘relating’, ‘analyzing’, ‘structuring’, ‘concretizing’, ‘applying’ and‘rehearsing’ during the intervention sessions (for macro-economics and managementaccounting), as well as in homework assignments. The second control group (C2) was onlyexposed to the usual instructional support and study guidance consisting of lectures, practicals,consulting hours, and individual feedback on assignments and examinations.

Assessment

A variety of summative assessment instruments were used spread over three posttestsessions to assess the effects of the intervention on self-regulation behavior. In the firstposttest session assignments for management accounting and multiple-choice questions formacro-economics were administered; besides solving the questions students were also askedto write while thinking, a variant of the thinking aloud technique. During the second posttestsession an attribution questionnaire was used, and metaknowledge of the regulatory skills onwhich the intervention focused was assessed with a direct knowledge test (e.g., “What do youhave to know at the start of a trimester in order to be able to organize and plan your study fora particular course? Also mention how you can obtain that information.”). In the last postteststudents had to fill in again questionnaires on self-efficacy, on self-regulation skills, and onattribution style that were already administered as pretests. At this stage transfer of regulationactivities to a course in statistics that was not involved in the intervention was also measured.



Therefore, a questionnaire containing eleven questions about study activities in the statisticscourse was administered (e.g., “How much time do you think you will have to invest in thetheoretical and practical parts of the statistics course, including the lessons?”). The overallexam result at the end of the academic year was used as indicator of academic performance.

As in the previous study multiple opportunities for formative assessment resulting indiagnostic feedback and coaching, were integrated in the learning environment. This wasrealized especially through discussion about and reflection on articulated problem approachesand verbalized difficulties experienced by the students, as well as through feedback on indi-vidual assignments.

Results

The results of the intervention were quite positive as is shown by the following majoroverall outcomes of the learning environment.

The experimental students demonstrated significantly more metaknowledge than thecontrol students about each regulatory skill included in the direct knowledge test. The effectsizes for the difference with C1 varied between .41 and .93, and with C2 between .26 and .56.This means, for instance, that the experimental students knew more about the impact ofpersonal characteristics (such as prior knowledge, being persistent, being careful andreflective) on studying and taking exams, and that they had a more extended knowledge ofcoping with negative emotions during learning.

Also a positive relationship was observed between metaknowledge of self-regulatoryactivities and academic performance. The entering characteristics of the students such as priorknowledge and intelligence, explained 43% of the variance in performance. When entering themetaknowledge variables in the regression equation the amount of criterion variance explainedincreased to 54%.

An important question was whether, as a result of the intervention, students had becomemore competent in learning, in the sense that they transferred the trained regulatory skills to acourse that was not involved in the intervention, more specifically statistics. Analysis ofstudents’ answers to the open-ended questionnaire with eleven questions, showed that theexperimental students were indeed more self-regulating for the statistics course than theirpeers in the control groups. For the difference with C1 the effect sizes varied between .27 and.69, and with C2 between .28 and .58. This means, for example, that the experimental studentsproved to be better informed about the statistics course, and, therefore, showed evidence ofmore orienting behavior; and, because they were able to formulate more studyrecommendations with regard to statistics, they were also more skilled in reflecting.Moreover, this transfer behavior explained a substantial part of the variance in the examscores for statistics: entering variables explained 41% of the criterion variance; this increasedto 67% when the transfer scores were included in the regression equation.

Finally, the students of the experimental group obtained better study results as measuredby exam scores, pass rates, and study careers. In the first year the experimental studentsoutperformed the control students as well in terms of the overall result (effect size .36 for thedifference with C1 and .38 for C2), as for the two major intervention courses: macro-economics (effect size .41 for C1, and .26 for C2), and management accounting (effect size.57 for C1 and .26 for C2).

From the 47 students in each research group significantly more experimental studentssucceeded in the first year, and obtained their master’s degree. In E, C1 and C2 respectively38, 28 and 34 students were successful in the first year, and respectively 37, 26 and 30 gottheir degree.

In sum, the implementation of this CLIA-based learning environment also resulted inpositive effects in the experimental group in comparison to the control groups. After theintervention students in the experimental group had more metaknowledge about regulationskills, they produced more self-regulation activities in the courses involved in the intervention,and were more in control of their academic performance. Moreover, they showed significant

transfer of the acquired self-regulation skills to a non-intervention course, and achieved betteracademic performance as measured by examination scores, pass rates, and study careers (for amore detailed discussion of these transfer results see De Corte, 2003).

Conclusions and discussion

In this article the CLIA-model was presented as a framework for designing learningenvironments that aim at pursuing in students high-literacy goals of education in response tonew societal demands. Although the two intervention studies summarized in the previoussections were not designed as formal test of the model, they yield nevertheless promisingsupport for it by showing that CLIA-based learning environments are indeed powerful inproducing in students high-literacy learning results, especially the acquisition and transfer ofself-regulation skills for learning and problem solving. As such the two studies prove to beuseful with respect to several promises of design-based research put forward by the Design-Based Research Collective (2003), namely the exploration of the potential of novel learningand teaching environments and the development of contextualized theories of learning andteaching, Major components of the CLIA-framework that were embodied in both learningenvironments are: metaknowledge and strategic aspects (heuristics and self-regulatory skills)of competence, but they were acquired embedded in the respective domain-specific contents;constructive, self-regulated, situated and collaborative learning representing essentialcharacteristics of effective learning processes; the six guiding principles for the design of theinstructional intervention; assessment tasks and instruments aligned and integrated withinstruction aimed at monitoring students’ progress toward competence in view of improvinglearning and teaching.

Admittedly, the observed effect sizes of the learning gains are mostly rather small. But, inthis respect one should take into account that several features of the interventions may havehad an oppressive impact on the learning outcomes. First of all, the scope and the duration ofthe both interventions were rather limited, and focused on only a restricted part of the students’curriculum; in the other components of the curriculum they were immersed in a moretraditional approach to teaching and learning. Moreover, even the fifth graders whoparticipated in the mathematics problem-solving environment, had already been taught forseveral years according to the more traditional approach. This approach has not only a lesserfocus – if any – on higher-order skills and interactive learning, but it may even result in habitsand beliefs of learning that are at right angles with the CLIA-framework, and have – so to say– to be “deconstructed” before the novel learning environment can be really productivelyimplemented. Besides, as argued by Gage (1996), the behavioral sciences – just as medicalscience – should take small effects seriously, especially when they are supported by relevanttheory and consistent with other research findings. And, our findings are fairly consistent withresults reported by others such as the FCL and the Jasper projects discussed briefly in the firstpart of this article.

Whereas instructional research of the kind represented by our two intervention studies,intends to contribute to the innovation of classroom practices (for a discussion see De Corte,2000), the primary goal should be to advance theory building about learning from instruction.An underlying idea is that an effective way at better understanding the processes of learning –and, thus, advancing theory – consists in designing innovative learning environments that arepowerful for eliciting in students the intended processes of knowledge and skill acquisition. But,although the contribution of intervention research to the elaboration of a theory of learning frominstruction is relevant and important (see e.g., Cobb, Confrey, diSessa, Lehrer, & Schauble,2003; Design-Based Research Collective, 2003), one should also be aware of its limitations. Inthis respect some methodological considerations are in order to conclude this article.

Due to the quasi-experimental design of our intervention studies, the complexity of thelearning environments, and the rather small experimental groups, it is impossible to establish



the relative importance of the different components of the interventions in producing learninggains. From an analytical perspective this is often considered as a methodological weakness ofteaching experiments. Moreover, intervention studies are criticized for their lack ofrandomization and appropriate control (Levin & O’Donnell, 1999). Notwithstanding thesecriticisms, we think that the more systemic approach of our studies, characterized by a highdegree of ecological validity, is perfectly suitable and defensible when the focus of interest isto evaluate the quality and the effectiveness of a multicomponential intervention asrepresented by our learning environments (see also Brown, Pressley, Van Meter, & Schuder,1996). In fact, it is plausible that it is precisely the combination of different aspects of thedesign, the content, and the implementation of the environments that is responsible for thelearning gains.

But, in view of ultimately building a highly credible theory of learning from instruction,based on convincing empirical evidence, we should, of course, also be concerned about theinternal validity of intervention studies. In this respect, it is useful to refer to the four stagemodel of educational intervention research proposed by Levin and O’Donnell (1999):

Stage 1 involves the development of preliminary ideas and hypotheses, as well ascarrying out observations and pilot work.

Stage 2 consists in conducting controlled laboratory experiments, on the one hand, andclassroom-based intervention and design experiments, on the other hand.

Stage 3 concerns the carrying out of so-called “randomized classroom trials studies”.Stage 4 refers to informed classroom practice.

According to Levin and O’Donnell stage 3 is crucial in view of upgrading the credibilityof educational intervention research, but has so far mostly been missing or neglected. Theyacknowledge that Stage 2 studies constitute a crucial step in the model, but they argue that inview of the intended credibility controlled laboratory experiments, and classroom-basedintervention and design experiments are only “preliminary”: the former lack a classroom-implementation component, while the latter do not provide scientifically credible evidencebecause of shortcomings in randomization and control. “The accumulation of classroom-basedscientifically credible evidence is precisely the function of the randomized classroom trialsstage. As in medical research, this consists of an examination of the proposed treatment orintervention under realistic, yet controlled, conditions” (Levin & O’Donnell, 1999, p. 204). Asimilar plea for randomized studies – in combination with matched experiments – was recentlymade by Slavin (2002).

One can hardly disagree in principle with this proposal, and there is no doubt that itsimplementation would contribute to improving the credibility of educational interventionresearch. For instance, one could conduct randomized experiments in which different versionsof complex learning environments are systematically contrasted and compared in view of theidentification of those components that contribute especially to their power and success. Inaddition, involving larger numbers of experimental classes in such investigations would allowto derive more reliable and generalizable conclusions about the effectiveness of the learningenvironments, but at the same time to study more systematically the relationship between theteachers’ implementation of those interventions, on the one hand, and their pupils’ learningoutcomes, on the other.

However, as alleged and shown by Slavin (2002), one should also be aware of “the factthat randomized experiments of interventions applying to entire classrooms can be extremelydifficult and expensive to do and are sometimes impossible” (p. 17).

Notes

1 The problem is not presented in its original format because it takes a lot of space. Moreover, translating it fromFlemish to English is somewhat cumbersome.

References

Berry, J., & Sahlberg, P. (1996). Investigating pupils’ ideas of learning. Learning and Instruction, 6, 19-36.

Boekaerts, M. (1993). Being concerned with well-being and with learning. Educational Psychologist, 28, 149-167.

Brown, A.L. (1994). The advancement of learning. Educational Researcher, 28(8), 4-12.

Brown, A.L., & Campione, J.C. (1994). Guided discovery in a community of learners. In K. McGilly (Ed.), Classroomlessons: Integrating cognitive theory and classroom practice (pp. 229-270). Cambridge, MA: The MIT Press.

Brown, A.L., & Campione, J.C. (1996). Psychological theory and the design of innovative learning environments: Onprocedures, principles, and systems. In L. Schauble & R. Glaser (Eds.), Innovations in learning: Newenvironments for education (pp. 289-325). Mahwah, NJ: Lawrence Erlbaum Associates.

Brown, R., Pressley, M., Van Meter, P., & Schuder, T. (1996). A quasi-experimental validation of transactionalstrategies instruction with low-achieving second-grade readers. Journal of Educational Psychology, 88, 18-37.

Bruer, J.T. (1993). Schools for thought: A science of learning in the classroom. Cambridge, MA: The MIT Press.

Burkhardt, H., & Schoenfeld, A. (2003). Improving educational research: Toward a more useful, more influential, andbetter-funded enterprise. Educational Researcher, 32(9), 3-14.

Campione, J.C., Shapiro, A.M., & Brown, A.L. (1995). Forms of transfer in a community of learners: Flexible learningand understanding. In A. McKeough, J. Lupart, & A. Marini (Eds.), Teaching for transfer: Fosteringgeneralization in learning (pp. 35-68). Mahwah, NJ: Lawrence Erlbaum Associates.

Cobb, P., Confrey, J., diSessa, A., Lehrer, R., & Schauble, L. (2003). Design experiments in educational research.Educational Researcher, 32(1), 9-13.

Cognition and Technology Group at Vanderbilt (1990). Anchored instruction and its relationship to situated cognition.Educational Researcher, 19(6), 2-10.

Cognition and Technology Group at Vanderbilt. (1997). The Jasper Project: Lessons in curriculum, instruction,assessment, and professional development. Mahwah, NJ: Lawrence Erlbaum Associates.

Cognition and Technology Group at Vanderbilt (2000). Adventures in anchored instruction: Lessons from beyond theivory tower. In R. Glaser (Ed.), Advances in instructional psychology. Volume 5: Educational design andcognitive science (pp. 35-99). Mahwah, NJ: Lawrence Erlbaum Associates.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale, NJ: Lawrence Erlbaum Associates.

Collins, A., Brown, J.S., & Newman, S.E. (1989). Cognitive apprenticeship: Teaching the crafts of reading, writing, andmathematics. In L. Resnick (Ed.), Knowing, learning, and instruction: Essays in honor of Robert Glaser (pp. 453-494). Hillsdale, NJ: Lawrence Erlbaum Associates.

Corno, L., Cronbach, L.J., Kupermintz, H., Lohman, D.F., Mandinach, E., Porteus, A.W., et al. (2002). Remaking theconcept of aptitude: Extending the legacy of Richard E. Snow. Mahwah, NJ: Lawrence Erlbaum Associates.

De Corte, E. (1996). Instructional psychology: Overview. In E. De Corte & F.E. Weinert (Eds.), Internationalencyclopedia of developmental and instructional psychology (pp. 33-43). Oxford, UK: Elsevier Science Ltd.

De Corte, E. (2000). Marrying theory building and the improvement of school practice: A permanent challenge forinstructional psychology. Learning and Instruction, 10, 249-256.

De Corte, E. (2003). Transfer as the productive use of acquired knowledge, skills, and motivations. Current Directionsin Psychological Science, 12, 142-146.

De Corte, E. (2004). Mainstreams and perspectives in research on learning (mathematics) from instruction. AppliedPsychology: An International Review, 53, 279-310.

De Corte, E., & Verschaffel, L. (1989). Teaching word problems in the primary school. What research has to say to theteacher. In B. Greer & G. Mulhern (Eds.), New developments in teaching mathematics (pp. 85-106). London:Routledge.

De Corte, E., Greer, B., & Verschaffel, L. (1996). Mathematics teaching and learning. In D.C. Berliner & R.C. Calfee(Eds.), Handbook of educational psychology (pp. 491-549). New York: Macmillan.



Design-Based Research Collective. (2003). Design-based research: An emerging paradigm for educational inquiry.Educational Researcher, 32(1), 5-8.

European Round Table of Industrialists (ERT) (1995). Education for Europeans. Towards a learning society. Brussels,Belgium: ERT.

Gage, N.L. (1996). Confronting counsels of despair for the behavioral sciences. Educational Researcher, 25(3), 5-15, 22.

Glaser, R. (1976). Components of a psychology of instruction: Toward a science of design. Review of EducationalResearch, 46, 1-24.

Glaser, R., & Silver, E. (1994). Assessment, testing, and instruction: Retrospect and prospect. In L. Darling-Hammond(Ed.), Review of research in education (Vol. 20, pp. 393-419). Washington, DC: American Educational ResearchAssociation.

Kelly, A.E. (2003). The role of design in educational research (Theme issue). Educational Researcher, 32(1).

Levin, J.R., & O’Donnell, A.M. (1999). What to do about educational research’s credibility gap? Issues in Education:Contributions from Educational Psychology, 5, 177-229.

Masui, C. (2002). Leervaardigheid bevorderen in het hoger onderwijs: Een ontwerponderzoek bij eerstejaarsstudenten[Enhancing learning competence in higher education: A design experiment with university freshmen] (with asummary in English). Leuven: Universitaire Pers Leuven.

Masui, C, & De Corte, E. (1999). Enhancing learning and problem-solving skills: Orienting and self-judging, twopowerful and trainable tools. Learning and Instruction, 9, 517-542.

Masui, C., & De Corte, E. (in press). Learning to reflect and to attribute constructively as basic components of self-regulated learning. British Journal of Educational Psyychology.

Masui, C., Borremans, A., Van Damme, J., & Vandenberghe, R. (1986). Technieken voor het expliciteren eninterpreteren van oplossingsgedrag in het kader van studiebegeleiding [Techniques for the verbalization andinterpretation problem-solving behavior within the framework of study guidance]. In E. De Corte (Ed.),Onderwijsleerprocessen: Bevindingen van Leuvens onderwijspsychologisch onderzoek [Teaching-learningprocesses: Findings of Leuven instructional psychology research]. Leuven/Amersfoort: Acco.

Mayer, R. E. (2001). Changing conceptions of learning: A century of progress in the scientific study of education. In L. Corno (Ed.), Education across a century: The centennial volume. Hundredth Yearbook of the National Societyfor the Study of Education (pp. 34-75). Chicago, IL: National Society for the Study of Education.

Ministerie van de Vlaamse Gemeenschap (1997). Gewoon basisonderwijs: Ontwikkelingsdoelen en eindtermen. Besluitvan mei ‘97 en decreet van juli ‘97 [Educational standards for the elementary school]. Brussel: DepartementOnderwijs, Centrum voor Informatie en Documentatie.

National Research Council (2000). How people learn: Brain, mind, experience, and school. In J.D. Bransford, A.L.Brown, & R.R. Cocking (Eds.), Committee on Developments in the Science of Learning with additional materialfrom the Committee on Learning Research and Educational Practice. Washington, DC: National Academy Press.

Palincsar, A.S., & Brown, A.L. (1984). Reciprocal teaching of comprehension-fostering and monitoring activities.Cognition and Instruction, 1, 117-175.

Perkins, D.N. (1995). Outsmarting IQ: The emerging science of learnable intelligence. New York: The Free Press.

Resnick, L.B. (1983). Toward a cognitive theory of instruction. In S.G. Paris, G.M. Olson, & H.W. Stevenson (Eds.),Learning and motivation in the classroom (pp. 5-38). Hillsdale, NJ: Lawrence Erlbaum Associates.

Salomon, G., & Perkins, D.N. (1998). Individual and social aspects of learning. In P.D. Pearson & A. Iran-Nejad (Eds.),Review of research in education (Vol. 23, pp. 1-24). Washington, DC: American Educational ResearchAssociation.

Scardamalia, M., Bereiter, C., & Steinbach, R. (1984). Teachability of reflective processes in written composition.Cognitive Science, 8, 173-190.

Schoenfeld, A.H. (1985). Mathematical problem solving. New York: Academic Press.

Schunk, D.H., & Zimmerman, B.J. (Eds.). (1998). Self-regulated learning: From teaching to self-reflective practice.New York/London: The Guilford Press.

Shavelson, R.J., & Baxter, G.P. (1992). Linking assessment with instruction. In F.K. Oser, A. Dick, & J.L. Patry (Eds.),Effective and responsible teaching. The new synthesis (pp. 80-90). San Francisco, CA: Jossey-Bass Publishers.

Shepard, L.A. (2001). The role of classroom assessment in teaching and learning. In V. Richardson (Ed.), Handbook ofresearch on teaching (4th ed., pp. 1066-1101). Washington, DC: American Educational Research Association.

Shuell, T.J. (1988). The role of the student in learning from instruction. Contemporary Educational Psychology, 13,276-295.

Slavin, R.E. (2002). Evidence-based education policies: Transforming educational practice and research. EducationalResearcher, 31(7), 15-21.

Snow, R.E., & Swanson, J. (1992). Instructional psychology: Aptitude, adaptation, and assessment. Annual Review ofPsychology, 43, 583-626.

Snow, R.E, Corno, L., & Jackson, D. III. (1996). Individual differences in affective and conative functions. In D.C.Berliner & R.C. Calfee (Eds.), Handbook of educational psychology (pp. 242-310). New York: Macmillan.

Vermunt, J.D.H.M. (1995). Process-oriented instruction in learning and thinking strategies. European Journal ofPsychology of Education, 10, 325-350.

Verschaffel, L., De Corte, E., Lasure, S., Van Vaerenbergh, G., Bogaerts, H., & Ratinckx, E. (1999). Learning to solvemathematical application problems: A design experiment with fifth graders. Mathematical Thinking andLearning, 1, 195-229.

Un défi important pour l’éducation et la recherche pédagogiqueest de développer en partant de notre compréhension actuelle del’apprentissage des environnements éducationnels susceptibles depromouvoir chez les étudiants des habiletés d’apprentissageautorégulatrices et collaboratives, des connaissances transférables, etune disposition orientée vers le raisonnement et la résolution deproblèmes compétents. Tenant compte des résultats de la recherche surl’apprentissage et l’enseignement, cet article présente le modèle “CLIA(Competence, Learning, Intervention, Assessment)” comme cadre deréférence pour concevoir des environnements d’apprentissage visés àstimuler chez les étudiants des processus d’apprentissage qui facilitentl’acquisition de connaissances productives et des habiletés compétentesd’apprentissage et de raisonnement. Ensuite, deux recherchesd’intervention sont présentées qui représentent les composantesmajeures du modèle CLIA: une expérience porte sur la résolution deproblèmes mathématiques dans l’enseignement primaire, et la secondea pour objet les habiletés autorégulatrices chez des étudiants enpremière année de l’université. Ces recherches ont été réalisées enparallèle avec le développement du modèle, et ont été instrumentalespour l’identification et la spécification des différentes composantes dumodèle. Les deux investigations ont apporté du support initial pour lemodèle en montrant que des environnements d’apprentissage basés surle cadre de référence CLIA sont en effet stimulants pour faciliter chezles étudiants l’acquisition de résultats d’apprentissage d’ordresupérieur, spécialement l’acquisition et le transfert d’habiletésd’autorégulation de l’apprentissage et de la résolution de problèmes.

Key words: Intervention studies, Learning, Powerful learning environments, Problem solving,Self-regulation

Received: October 2003

Revision received: July 2004



Erik De Corte. Center for Instructional Psychology and Technology, University of Leuven,Vesaliusstraat 2, B-3000 Leuven, Belgium; E-mail: [email protected]; Web site:www.kuleuven.be

Current theme of research:

Analysis and improvement of mathematical problem solving, Mathematics-related beliefs in students and teachers,Designing powerful learning environments.

Most relevant publications in the field of Psychology of Education:

De Corte, E. (2000). Marrying theory building and the improvement of school practice: A permanent challenge forinstructional psychology. Learning and Instruction, 10, 249-266.

De Corte, E., Verschaffel, L., & Op ’t Eynde P. (2000). Self-regulation: A characteristic and a goal of mathematicseducation. In P. Pintrich, M. Boekaerts, & M. Zeidner (Eds.), Self-regulation: Theory, research, and applications(pp. 687-726). Mahwah, NJ: Lawrence Erlbaum Associates.

De Corte, E., Op ’t Eynde, P., & Verschaffel, L. (2002). Knowing what to believe: The relevance of mathematicalbeliefs for mathematics education. In B.K. Hofer & P.R. Pintrich (Eds.), Personal epistemology; The psychologyof beliefs about knowledge and knowing (pp. 297-320). Mahwah, NJ: Lawrence Erlbaum Associates.

De Corte, E. (2003). Transfer as the productive use of acquired knowledge, skills, and motivations. Current Directionsin Psychological Science, 12, 142-146.

De Corte, E. (2004). Mainstreams and perspectives in research on learning (mathematics) and instruction. AppliedPsychology: An International Review, 53, 279-310.

Lieven Verschaffel. Center for Instructional Psychology and Technology, University of Leuven,Vesaliusstraat 2, B-3000 Leuven, Belgium; E-mail: [email protected]; Website: www.kuleuven.be


Strategy choice and strategy change, Mathematical modeling and problem solving, Psychology of mathematics education.


Verschaffel, L., & De Corte, E. (1997). Teaching realistic mathematical modeling in the elementary school. A teachingexperiment with fifth graders. Journal of Research in Mathematics Education, 28, 577-601.

Verschaffel, L., De Corte, E., Lasure, S., Van Vaerenbergh, G., Bogaerts, H., & Ratinckx, E. (1999). Design andevaluation of a learning environment for mathematical modeling and problem solving in upper elementary schoolchildren. Mathematical Thinking and Learning, 1, 195-230.

Verschaffel, L., De Corte, E., & Vierstraete, H. (1999). Upper elementary school pupils’ difficulties in modeling andsolving non-standard additive word problems involving ordinal numbers. Journal for Research in MathematicsEducation, 30, 265-285.

Verschaffel, L., Greer, B., & De Corte, E. (2000). Making sense of word problems. Lisse, The Netherlands: Swets &Zeitlinger, XVII-203 pp.

Verschaffel, L., Greer, B., & De Corte, E. (2003). Everyday knowledge and mathematical modeling of school wordproblems. In K. Gravemeijer, R. Lehrer, B. Van Oers, & L. Verschaffel (Eds.), Symbolizing, modeling and tooluse in mathematics education (pp. 257-276). Dordrecht, The Netherlands: Kluwer.

Chris Masui. Department of Men, Society and Communication, Limburg University Center,Universitaire Campus, gebouw D, B-3590 Diepenbeek, Belgium; E-mail: [email protected];Web site: www.luc.ac.be


Learning and problem solving skills in higher education, Self-theories, Conceptions of learning and teaching.


Masui, C. (2002). Leervaardigheid bevorderen in het hoger onderwijs: Een ontwerponderzoek bij eerstejaarsstudenten[Enhancing learning competence in higher education: A design experiment with university freshmen] (with asummary in English). Leuven: Universitaire Pers Leuven.

Masui, C., & De Corte, E. (1999). Enhancing learning and problem solving skills: Orienting and self-judging, twopowerful and trainable tools. Learning and Instruction, 9, 517-542.

Masui, C., & De Corte, E. (in press). Learning to reflect and to attribute constructively as basic components of self-regulated learning. British Journal of Educational Psychology.


the clia-model: a framework for designing powerful learning environments for thinking and problem...

Documents