N EW D IRECTIONS FOR T EACHING AND L EARNING , no. 128, Winter 2011 © Wiley Periodicals, Inc.
Published online in Wiley Online Library (wileyonlinelibrary .com) • DOI: 10.1002/tl.4653Problem-Bad Learning
Deborah E. Allen, Richard S. Donham, Stephen A. Bernhardt
Problem-bad learning (PBL) has wide currency on many college and uni-versity campus, including our own, the University of Delaware. Although we would like to be able to claim clear evidence for PBL in terms of student learning outcomes, bad on our review of the literature, we cannot state that rearch strongly favors a PBL approach, at least not if the primary evidence is subject matter learning.
There is some evidence of PBL effectiveness in medical school ttings where it began, and there are numerous accounts of PBL implementation in various undergraduate contexts, replete with persuasively positive data from cour evaluations (Duch, Groh, and Allen, 2001). However, evidence for learning outcomes is still needed. In this chapter, we review the origins of PBL, outline its characteristic methods, and suggest why we believe PBL has a persistent and growing infl uence among educators.
Origins of PBL in Medical Schools
PBL was formalized by medical educators in the 1950s and 1960s to address the exponential expansion of medical knowledge while better aligning traditional classroom problem-solving approaches with tho ud in clinical practice (Barrows and Tamblyn, 1980; Boud, 1985). Traditional approaches were bad on the bucket theory (Wood, 1994): If medical stu-dents were fi lled with the requisite foundational knowledge, they would be able to strategically retrieve and direct just the right subts of it toward problems of clinical practice. PBL was designed to address the underlying fl aws of the bucket theory , especially leaky , overfl owing, or inappropriately 21
In problem-bad learning, students working in collaborative groups
learn by resolving complex, realistic problems under the guidance of faculty. In this chapter , we examine the evidence for effectiveness of
the method to achieve its goals of fostering deep understandings
of content and discuss the potential for developing process skills:
rearch, negotiation and teamwork, writing, and verbal
communication.
22E VIDENCE-B ASED T EACHING
fi lled buckets. By prenting complex ca histories typical of real patients as the pretext for learning, PBL demanded that students call on an inte-grated, multidisciplinary knowledge ba (Wood, 1994).
In the idealized learning cycle of medical school PBL (Engle, 1999), students working in teams learn by solving real or realistic problems. Stu-dents grapple with a multistage, complex medical ca history, which offers an engaging and memorable context for learning. As they defi ne the prob-lem’s scope and boundaries, student teams identify and organize relevant ideas and prior knowledge. The teams form questions bad on lf- identifi ed gaps in their knowledge, and they u the questions to guide subquent independent rearch outside the classroom, with rearch tasks parceled out among team members. When the students reconvene, they prent and discuss their fi ndings, integrating their new knowledge and skills into the problem context. As they move through the stages of a complex problem, they continue to defi ne new areas of needed learning in pursuit of a solution. In the ca of this original PBL model, a solution is an accurate diagnosis and recommendation of successful treatment of the patient.
PBL continues to be a favored method in many medical schools. What became evident in effectiveness studies was that there was no simple answer to the question “Is PBL better than traditional methods?” Several meta-analys of the data suggested that PBL has modest or no benefi cial effect on student learning of content (from the United States Medical Licensing Examination [USMLE] Step 1—basic science understanding; Albane and Mitchell, 1993; Nandi and others, 2000; Vernon and Blake, 1993). In fact, it appears that students in a traditional medical program sometimes, but not consistently, slightly outperform their PBL counterparts.
However, disaggregation of the data suggests an underlying richness that is not captured simply by looking at student achievement on content recall exams. If, for example, scores on the USMLE Step 2 (knowledge of clinical practice) or ability to apply knowledge in the clinic after graduation are considered, medical school students with PBL experience frequently outperform their traditional counterparts (Albane and Mitchell, 1993; Dochy, Segers, Van den Bossche, and Gijbels, 2003; Koh, Khoo, Wong, and Koh, 2008; Vernon and Blake, 1993). Recent meta-analys have begun to tea apart some of the relative merits of PBL and suggest that the most positive effects are en with student understanding of the organizing prin-ciples that link concepts in the knowledge domain being studied (Gijbels, Dochy, Van den Bossche, and Segers, 2005). Dochy and others (2003) report
ed a robust positive effect from PBL on the skills of students, noting that, intriguingly, students in PBL remember more acquired knowledge compared with their traditional counterparts. The early meta-analys of PBL outcomes in the medical school tting (Albane and Mitchell, 1993; V ernon and Blake, 1993) also document positive student attitudes about N EW D IRECTIONS FOR T EACHING AND L EARNING • DOI: 10.1002/tl
P ROBLEM-B ASED L EARNING 23 learning, with students frequently viewing PBL as both a challenging and a motivating approach.
Strategies for PBL Implementation
Becau PBL explicitly address some of the shortcomings of science edu-cation, it migrated into undergraduate science and engineering classrooms (Woods, 1985). It then expanded into basic as well as applied fi elds as well as into the humanities and social sciences (Duch and others, 2001). With the introduction of PBL to undergraduate cours, teachers modifi ed the method to accommodate larger class sizes, greater student diversity, timing and scheduling issues, multiple classroom groups, and lack of suitable classroom space (Allen, Duch, and Groh, 1996).
PBL requires a shift in the educational paradigm for faculty. In PBL, the role of the instructor shifts fro
m prenter of information to facilitator of a problem-solving process. Although the PBL process calls on students to become lf-directed learners, faculty facilitators guide them by monitoring discussion and intervening when appropriate, asking questions that probe accuracy, relevance, and depth of information and analys; raising new (or neglected) issues for consideration; and fostering full and even participa-tion (Mayo, Donnelly, and Schwartz, 1995).
Instead of lecturing, PBL instructors must fi nd or create good prob-lems bad on clear learning goals. Through the problems, instructors lead students to learn key concepts, facts, and process related to core cour content. PBL problems must be carefully constructed—not only to prent students with issues and dilemmas that matter to them but also to foster their development of conceptual frameworks (Hung, Jonasn, and Liu, 2007). PBL problems may intentionally po cognitive challenges by not providing all the information needed, thereby motivating a lf-directed arch for explanations. Instructors often allow students considerable lati-tude to make fal starts and wrong turns. Well-developed, peer-reviewed problems can be found at the PBL Clearinghou (University of Delaware, 2010).
Successful implementation of PBL is critically dependent on the instructor’s scaffolding of students’ active learning and knowledge con-struction (Amador, Miles, and Peters, 2006; Duch and others, 20
01). For example, PBL instructors can plan for intervals of class discussion or mini-lectures to help students navigate conceptual impass, to dig more deeply into certain topics, or to fi nd uful resources. Instructors can enter team discussions to listen and po questions (Hmelo-Silver, Duncan, and Chinn, 2007). They can also u student facilitators to extend their instruc-tional reach.
Importantly, PBL can support the development of a range of “soft” skills: rearch skills, negotiation and teamwork, reading, writing, and oral communication. Cooperative learning strategies that foster effective
N EW D IRECTIONS FOR T EACHING AND L EARNING • DOI: 10.1002/tl
24 E VIDENCE -B ASED T EACHING
N EW D IRECTIONS FOR T EACHING AND L EARNING • DOI: 10.1002/tl
teamwork become critical, as does the need for everyone to work to keep team members engaged and on track (Johnson, Johnson, and Smith, 1998). PBL classrooms are particularly well suited to the development of writing abilities. PBL instructors tend to rely on authentic asssment, with most problems leading up to a demonstration or prentation of learning, often taking the form of a written
product: a solution, a recommendation, a sum-mary of what was learned, or some other form of group or individual reporting. To encourage development of writing skills, thinking skills, and learning in general, instructors can call for students to produce specifi c genres of writing: progress reports, schedules, task lists, meeting minutes, abstracts, literature reviews, proofs, lab reports, data analys, and technical briefi ngs (Klein, 1999). Alaimo, Bean, Langenhan, and Nichols (2009) showed how to integrate writing as a core activity in an inquiry-bad chemistry cour, demonstrating strong learning outcomes in the process.
Instructors must also encourage good team communication strategies. Teams must avoid reaching premature closure or succumbing to group-think—where a group izes on a path becau a team member is forceful or persuasive. The teams that perform best are tho that generate and sus-tain consideration of multiple alternatives, engaging in and sustaining “substantive confl ict” (Burnett, 1991).
Effectiveness of PBL on Content Learning in Undergraduate Settings
Confusion and lack of specifi cation about what PBL is as it is actually prac-ticed in the classroom hampers analysis of the effect of PBL on the acquisi-tion of content learning. In particular, PBL adopt
ers in undergraduate ttings, grappling with the diffi culties of monitoring multiple classroom groups, hybridize the method in various ways to incorporate aspects of discussion and ca study method teaching (Silverman and Welty , 1990). Instructors tend to inrt highly choreographed gments of instructor-centered, whole-class discussions into the PBL cycle and to interpo PBL problems intermittently throughout the cour schedule, blended with more traditional instruction (Duch and others, 2001). As Newman (2003) noted, this hybridization of PBL makes it “diffi cult to distinguish between different types of PBL and even to distinguish between PBL and other edu-cational interventions” (p. 7).
Nevertheless, there are scattered reports of positive outcomes. In a study of over 6,500 students, Hake (1998) found that interactive engage-ment methods (broadly defined as heads-on, hands-on activities with immediate feedback) were strongly superior to lecture-centered instruction in improving performance on valid and reliable mechanics tests ud to asss students’ understanding of physics. Williams (2001) reported gains in the Force Concept Inventory for students in a PBL cour that are con-sistent with the averages in other introductory physics cours that u
P ROBLEM-B ASED L EARNING 25 interactive engagement methods. Palaez (2002) obrved that
students in a PBL biology cour with an intensive writing component outperformed students in a cour using traditional lecture-bad instruction on exams that assd conceptual understandings.
Although there is less rearch on undergraduate learning than in medical education, the data support the broad conclusion that PBL may show only modest benefi ts on recalled content knowledge, but it positively infl uences integration of new knowledge with existing knowledge. How-ever, faculty members frequently adopt PBL to help students develop life-long learning skills. The skills are exercid routinely in the natural cour of the PBL learning cycle. Given the additional but divergent stu-dent learning goals, many faculty members are satisfi ed with student con-tent learning that is similar or not signifi cantly decread when using PBL. At the very least, the fi ndings assuage any residual concerns they or oth-ers may have that spending time on the ambitious process objectives undermines the learning of esntial cour content.
Effectiveness of PBL on Process Skills
Becau PBL engages students in a range of soft skills, perhaps other posi-tive learning outcomes can be claimed for the method. A ca in point is the benefi t of using cooperative learning groups on
such general aspects of academic success as retention as well as on fostering positive student atti-tudes about learning (Springer, Stanne, and Donovan, 1999). Another is the u of writing-to-learn strategies in PBL. Incorporation of short, in-class writing assignments improves student performance on traditional concept and content-bad exams (Butler, Phillmann, and Smart, 2001; Davidson and Pearce, 1990; Drabick, Weisberg, Paul, and Bubier, 2007; Stewart, Myers, and Culley, 2010).
There is some evidence that systemic and sustained u of PBL in the classroom fosters cognitive growth. Downing and others (2009) followed two parallel cohorts of students in degree programs, one taught with PBL, the other by traditional methods, and found greater gains in metacognitive skills in the PBL group. Tiwari, Lai, So, and Yurn (2006) similarly reported signifi cant differences in the development of undergraduate nursing stu-dents’ critical thinking dispositions in a PBL versus a lecture-bad cour, as determined by comparisons of pre- and posttest scores on the California Critical Thinking Disposition Inventory.
Effectiveness of PBL on Student Engagement
Widespread agreement is emerging that at the core of effective teaching are activities that engage students by challenging them academically and involving them intenly, within supportive environments that provide multiple opportunities for interactions with faculty, peers, and members of
N EW D IRECTIONS FOR T EACHING AND L EARNING • DOI: 10.1002/tl
