Instruction followed by problem solving

This section of the guide focuses on peer-led team learning (PLTL) and worked examples plus practice. We included these pedagogies because they are well-defined examples of instruction followed by problem solving and are either widely used in undergraduate science education or have a strong literature base.

Peer-led Team Learning (PLTL)
  • PLTL Approach and Benefits
    • In the PLTL approach, the instruction phase takes place in the traditional classroom, often in the form of lecture, and the problem-solving phase takes place as students work in collaborative groups (typically ranging from 6-10 students) facilitated by a trained undergraduate peer leader for 90−120 minutes each week.
    • This approach provides facilitated help to students in their courses, improves students’ problem-solving skills, enhances students’ communication abilities, and provides an active-learning experience for students.
    • The peer leader should not help solve the problems with the students in the group, but guides them to discuss their reasoning by asking probing questions and to equally participate by using different collaborative learning strategies (such as round robin, scribe, and pairs). Students decide as a group whether the answer is correct or not, which encourages the students to consider the problem more deeply.
    • PLTL is used in many STEM undergraduate disciplines including biology, chemistry, mathematics, physics, psychology, and computer science, and in all types of institutions. It has been used at all different levels of undergraduate courses. If done well, PLTL can improve course grades, course and series retention, standardized and course exam performance, and DWF rates.
    • PLTL can also be beneficial to peer leaders, self-reporting greater content learning, improved study skills, improved interpersonal skills, increased leadership skills, and confidence.
  • Social constructivism provides the theoretical framework.
    • In the constructivist framework, teaching is not the transmission of knowledge from the instructor to the student. The instructor is a facilitator or guide, giving structure to the learning process.
    • Students construct meaning (e.g., develop concepts and models) through active involvement with the material and by making sense of their experiences.
    • Social constructivism assumes that students’ understanding and sense-making are developed jointly in collaboration with other students.
    • Social constructivism emphasizes the role of Vygotsky’s Zone of Proximal Development (ZPD). Vygotsky stated that learning moves from one’s current intellectual level to a higher level that is most close to one’s potential to learn. That is, ZPD is defined as the difference between what a student can do without assistance and what a student can do in collaboration with a more capable peer.
      • The peer leader is considered to be an effective guide because they are in the ZPD of the students in their PLTL group.
  • Optimal benefits from PLTL require the use of several design principles:
    • The PLTL program is integral to the course and integrated with other course components.
    • Course instructors are closely involved with organizing the program and selecting, training, and supervising the peer leaders. Training should include:
      • how to effectively create a community of practice within their group such that students will make joint decisions while solving the problems, discuss multiple approaches to solve the problems, and practice professional social and communication skills;
      • practice with questioning strategies to support students in deepening their discussion to include explanations for their ideas and problem-solving processes;
      • learning about how students learn based on psychology and education research, and how to apply this information while facilitating their group.
    • Materials are challenging at an appropriate level, written to encourage collaboration and active learning. Materials should:
      • require students to work collaboratively to solve problems;
      • encourage students to engage deeply with the content (i.e., include prompts asking them to explain their reasoning or process), disciplinary vocabulary (i.e., include prompts asking them to define terms in their own words), and essential skills;
      • become more complex throughout the problem set while ensuring that the students are always within their Zone of Proximal Development.
    • Organizational arrangements promote active learning via focus on group size, room space, length of session, and low noise level.
    • The institution and department encourage and support innovative teaching.
Frey, R. F., Fink, A., Cahill, M. J., McDaniel, M. A., & Solomon, E. D. (2018). Peer-led team learning in general chemistry I: Interactions with identity, academic preparation, and a course-based intervention. Journal of Chemical Education, 95(12), 2103-2113. This research article describes the findings of a quasi-experimental study using 5 years of data (2012-2016) involving N = 1254 first-year students in first-semester general chemistry. The authors analyzed exam performance of students who participated in the peer-led team learning program (PLTL) (N=847) versus those who did not participate in the PLTL program (N=407), disaggregated by demographics, academic preparation including college-preparatory coursework, and participation in a growth-mindset intervention. The PLTL implementation followed the standard model, which incorporates collaborative group work that is facilitated by a trained undergraduate peer leader using collaborative learning strategies, except in this implementation the students self-selected into the program, which has mandatory attendance (see supplemental material for this article for program details). For this study, it is important to note that all students, whether they were in PLTL or not, had access to the weekly PLTL problem set after the groups met. The PLTL effect was 5 percentage points across demographic identity groups. PLTL gave a larger benefit to students with less college-preparatory coursework (8 points for students with no AP; 2 points for students with 4 AP scores) but was not correlated with their math or chemistry knowledge. Instructors should note that five years of performance data from General Chemistry I revealed the robustness of the PLTL effect on exam performance. The findings suggest that PLTL may assist in students’ self-management or scientific-reasoning skills, as opposed to differences in their basic knowledge. Hence, PLTL may foster equity when situated within an array of evidence-based resources targeting students’ unique academic and personal backgrounds.
Preszler, R. W. (2009). Replacing lecture with peer-led workshops improves student learning. CBE—Life Sciences Education, 8(3), 182-192. PLTL in first-semester introductory biology at an institution predominantly serving Hispanic and Native American students was examined by comparing 7 semesters before implementing PLTL with 3 semesters after implementing PLTL. Specifically, one of three 50-minute lectures was replaced with a 65-minute required PLTL session focused on case studies. The implementation differed from the standard PLTL model in three ways: the PLTL sessions contained 19 students instead of the normal 6-10 students; the PLTL sessions were 65 minutes versus the recommended 90-120 minutes; peer leaders graded the PLTL problem sheets where in the PLTL model peer leaders do not grade. The authors studied course grades and student performance on exam questions in pre-PLTL and PLTL semesters. They found that students in PLTL semesters scored higher on paired exam questions and that course grades were higher in PLTL semesters. Notably, student scores on exam questions that required higher-level thinking increased from pre-PLTL to PLTL semesters. Females had a larger improvement in grades and increased retention than males, and African Amercian, Latino, and Native American students had a larger improvement in grades than Asian American and Caucasian students. Students in all declared majors benefited from the PLTL sessions, but undeclared students did not improve from pre-PLTL to PLTL semesters. It should be noted that this study did not use any covariates of student preparation, such as ACT/SAT or pre-semester knowledge assessment. Instructors should note that this study shows that PLTL can be implemented in an introductory biology course in which the problems are case-based or model based and not quantitative. It also shows that implementing PLTL can improve student learning by replacing one of the lectures. It is important to keep the following key elements: sessions are mandatory, peer leaders are trained, course instructors write the PLTL problems, and students are prompted via questions to solve the problems together, not shown the solutions.
Snyder, J. J., Sloane, J. D., Dunk, R. D., & Wiles, J. R. (2016). Peer-led team learning helps minority students succeed. PLoS Biology, 14(3), e1002398. This study describes the effect of PLTL in a large second-semester introductory biology course where enrollment in a concurrent lab course was optional. The DFW rates of underrepresented minorities (URM) versus non-URM students were compared dependent on PLTL participation and concurrent lab enrollment. The standard PLTL model, which incorporates collaborative group work that is facilitated by a trained undergraduate peer leader using collaborative learning strategies, was followed except that PLTL participation was optional and considered supplemental to the lecture sessions. The authors found that course retention was higher for students in PLTL than not in PLTL, independent of concurrent laboratory enrollment. There was a decrease from 40% to 15% in the DFW rates of URM students in PLTL compared to URM students not in PLTL, and PLTL participation led to a small but significant decrease in DFW rates among non-URM students as well.  No difference emerged in the DFW rates between URM and non-URM students in PLTL, meaning that PLTL closed the DFW gap between racial/ethnic groups. Non-lab students in PLTL earned average grades equivalent to lab students. For non-lab URM students, 50% who did not engage in PLTL withdrew from the course; those who did engage in PLTL completed with at least a C grade. Instructors should note that this study shows that PLTL benefits all students but has a larger impact for URM students and for students not enrolled in a laboratory component.
Repice, M. D., Sawyer, R. K., Hogrebe, M. C., Brown, P. L., Luesse, S. B., Gealy, D. J., & Frey, R. F. (2016). Talking through the problems: A study of discourse in peer-led small groups. Chemistry Education Research and Practice, 17(3), 555-568. This paper describes the discourse students use while working together to solve problems in a peer-led small-group setting. Interactions of students solving three different problem types (calculation, data analysis, and model building) were studied across one semester. Importantly, prompts did not contain explicit information about progressive steps to take or about what information was needed for problem solving. The authors found that the most common discourse structures observed included True Dialogue (leaders or students ask questions without knowing or seeking the ‘‘correct” answer), Cross-Discussion (conversation between students moderated by the peer leader), Groupwork (small groups work on shared tasks), confirming results from Lemke (1990). According to Lemke, these discourse structures give students more practice in ‘‘talking science’’ than the typical instructor-student discourse, the Triadic Dialogue, in which the instructor controls the interaction. Study results revealed students: 1) used regulative language to promote discussion, exchange information, and manage their and group-members’ learning; 2) used instructional discourse to practice “talking science” and develop shared understanding of chemistry knowledge and vocabulary; 3) communicated by focusing on the process of complex problem solving to move through the problems together; 4) engaged in little deeper-meaning-making discourse unless prompted. This paper shows that communication is a crucial aspect of learning in small-group settings and the facilitator should encourage equal participation within the group. Instructors should note that “talking science’’ in a small group creates a community of practice around the subject, in which students talk through problems to make joint decisions, solve problems in multiple ways, discuss ideas, learn about their own learning, and learn professional social and communicative skills. However, students do not, on their own, often engage in open questioning, deeper conceptual explanations, and self-monitoring of their learning. Hence, instructors need to have carefully constructed activities and well-trained peer leaders who prompt students to engage in effective discourse.
Knight, J. K., Wise, S. B., Rentsch, J., & Furtak, E. M. (2015). Cues matter: Learning assistants influence introductory biology student interactions during clicker-question discussions. CBE—Life Sciences Education, 14(4), ar41.  All instructional practices that involve active problem solving are based on constructivism, in which students jointly engage in sensemaking through discussion and collaboration. Hence it is essential to understand how peer leaders can facilitate students to   discuss ideas and justify reasoning during problem solving. This paper describes how students in small groups engage in discussion, examining the interplay between student and peer leader statements. In particular, it focuses on student reasoning and questioning and on the effect peer leaders (in this case, learning assistants (LAs)) had on stopping or encouraging discussion in a first-year introductory molecular and cell biology course. Groups of 4 students were recorded (6/23 groups), and student discussions and the effect that LAs’ interactions had on these discussions were examined. Results revealed that 1) when prompted, students explained their reasoning to others, but infrequently used claims logically connected to evidence; 2) students spent a short time (averaged ~1 minute) discussing ideas, and spent ~1.5 minute with LAs present; 3) students articulated reasoning for multiple answers if the clicker question was at a lower-order Bloom’s level rather than a more complex question; 4) the presence of LAs shifted student questioning from requesting information to requesting feedback; 5) LAs giving explanations to students as feedback stopped further student discussion, while LAs using prompting questions encouraged further student discussion and encouraged students to collaboratively discuss their reasoning and understand the material in more depth. This paper shows that peer leaders can increase students’ ability to use reasoning and participate in discussions that encourage deeper thinking in large introductory courses. Instructors should note that to increase the use of reasoning in student discussion in small groups, instructors should add cues to small-group questions that state how students should discuss together (e.g., Explain your reasoning in your groups and explain your reasoning for not selecting other answers), and then ask for these explanations during the whole-class discussion. LAs or other peer leaders (e.g., PLTL leaders) should learn about and have the chance to practice using questioning strategies that encourage student discussion that includes reasoning.
Eberlein, T., Kampmeier, J., Minderhout, V., Moog, R. S., Platt, T., Varma‐Nelson, P., & White, H. B. (2008). Pedagogies of engagement in science. Biochemistry and molecular biology education, 36(4), 262-273. This article compares and contrasts the key features of three collaborative pedagogies: PLTL (peer-led team learning), POGIL (process-oriented, guided-inquiry learning), and PBL (problem-based learning). The paper describes each pedagogy in general as well as the theoretical framework, classroom characteristics, problem characteristics, scalability, assessment, and faculty/student acceptance. In the standard PLTL model, six critical criteria have been identified to ensure the long-term success of implementation: integration with the course, course instructors actively involved with the PLTL component, trained peer leaders, appropriately challenging problems that require group interaction, rooms arranged to facilitate group work, and department and institutional support. Two clear distinctions between PLTL and the other two (i.e., POGIL and PBL) are that 1) PLTL is supplemental to the lecture although integrally incorporated into the course, and 2) PLTL is developed around the importance and use of trained peer leaders, who are upper-level undergraduates and have successfully completed the course of interest. This reliance on peer leaders is based on the social-constructivist ideas of Vygotsky that focus on the zone of proximal development (ZPD) in which students are solving challenging problems that they can solve only with interaction with their PLTL group facilitated by the peer leader. Instructors should note that this paper provides an excellent summary of the key features of PLTL and the related pedagogies, POGIL and PBL. This allows an instructor new to these pedagogies to decide which one would best fit into their course learning objectives and departmental environment.


Wilson, S. B., & Varma-Nelson, P. (2016). Small groups, significant impact: A review of peer-led team learning research with implications for STEM education researchers and faculty. Journal of Chemical Education, 93(10), 1686-1702. This review describes key features of PLTL, the theoretical framework, and the current literature grouped into the following 5 categories: student success measures, student perceptions, reasoning and critical thinking skills, research on peer leaders, and variations of the traditional PLTL model. In the standard PLTL model, groups of approximately eight students are facilitated by a trained peer leader and collaboratively solve problems for 90−120 minutes each week. Problem sets are written by the course instructors and are active-learning activities designed to be collaborative. The review discusses PLTL’s basis in social constructivism and the implications it has for implementation. For example, in PLTL peer leaders provide support for effective collaboration by encouraging students to discuss their ideas and decide together how to solve problems. Peer leaders decrease support as the students become independent learners who can work collaboratively. PLTL is used in many STEM undergraduate disciplines including biology, chemistry, mathematics, physics, psychology, and computer science, and in all types of institutions. Performance outcomes that have been evaluated include course grades, course and series retention, standardized and course exam performance, and DWF rates. Discourse studies have been performed to better understand the types of language that students use while they are discussing the problems, their process, and their solutions. Research on peer leaders examines how leaders translate their training into practice; their effect on student discourse; and the benefits of being a peer leader. Last, the review discusses different variations on the traditional PLTL approach. Instructors should note that this review provides an excellent overview of the key features and philosophy of the PLTL approach, as well as the variations that have been implemented in the PLTL model. In addition, it has a robust discussion about the effect that PLTL has on multiple student performance outcomes in a variety of the disciplines.
This is the website for the PLTL International Society. It is an open community of educators who are fostering student learning via PLTL and are interested assisting others in implementing PLTL at their institutions. This community has an annual conference in June, and the proceedings of these conferences are posted on this website. In addition, the website contains a newsletter, resources for starting a PLTL program and training peer leaders, and a list of publications about PLTL.



Worked examples
  • In worked examples, the instruction takes the form of example problems. These problems include a problem statement and a step-by-step procedure for solving the problem, intended to show how an expert might solve this type of problem. After this explicit instruction, students complete practice problems like the worked examples.
  • Worked examples plus practice problems have been found to be beneficial when compared to instruction followed by problem solving alone. This benefit is observed for novices learning to solve complex problems but is lost as learners become more expert in the domain and is not observed for simple problems.
  • Worked examples provide guidance that can help students learn to do analogous problems (near transfer) and may have similar benefits to productive failure and scaffolded guided inquiry for near transfer.
  • Cognitive Load Theory (see a description in the Underpinnings section) is used to explain the benefit of worked examples, which reduce the burden on working memory when students are in early stages of learning to solve problems in a domain. By focusing students’ attention and reducing extraneous demands on working memory, worked examples can facilitate students’ recognition of important problem and solution features. As students move to later stages of skill acquisition that focus on increasing speed and accuracy, worked examples can become redundant with their knowledge and thus provide unneeded burden on working memory.
    • Worked examples help students in early-to-intermediate stages of cognitive skill development as they are learning to abstract general principles for solving a given type of problem.
    • Comparing worked examples that focus on different types of problems can also help students identify deep features and abstract general principles that help them know when to use a given problem solving approach.
    • Problem-solving practice that incorporates strategies like retrieval and interleaving become more effective as students seek to become faster and more accurate.
  • Optimal benefits from using worked examples in instruction require the use of several design principles:
    • Individual worked examples should be designed to place minimal demands on working memory by
      • Integrating sources of information (e.g., images integrated with explanatory text or auditory explanations),
      • Including visual cues to help students readily follow the explanation,
      • Fostering identification of subgoals within a problem, either by labeling or visually separating chunks of a problem solution corresponding to a subgoal.
    • Lessons should include at least two worked examples for a type of problem.
    • Worked examples should be accompanied by practice problems and should be interspersed throughout a lesson rather than combined in one section of the lesson.
    • Different problem types should use similar cover stories to emphasize deep problem features.
    • Instruction should foster student self-explanation. Both training students to self-explain and prompting them to do so (even by simple means, such as choosing from a menu of potentially relevant principles) can help students self-explain. Students should be able to:
      • Relate solutions to abstract principles
      • Compare different examples and self-explain key similarities and differences.
    • If multiple worked examples for a given type of problem are used, it can be beneficial to remove guidance in stages (also known as fading). Backwards fading (leaving blanks later in the problems first, then earlier and earlier) has been found to be more beneficial than forward fading.


Tuovinen, J.E., and Sweller, J. (1999). A comparison of cognitive load associated with discovery learning and worked examples. Journal of Education Psychology, 91 (2), 334-341.  This study compares the effects of discovery/exploration learning to learning with worked examples. The use of worked examples as instructional tools is derived from cognitive load theory. This theory predicts that structured learning approaches are more effective for novices because they put lower demand on learners’ limited working memory.  In this study, the authors compared unstructured (“exploratory”) problem solving practice to worked examples + problem solving in college students (N = 32). Initial instruction was common for all students: a lecture on a database program accompanied by associated tasks. Students were then randomly assigned to a worked examples or exploratory practice group for two additional lessons, each of which consisted of brief direct instruction (10-15 minutes) followed by practice (40 minutes). The exploratory practice group received instructions to “Try out the functions in each of the lessons in situations you create yourself,” while the worked examples group was given pairs of problems: one with a worked-out solution, and one that they were directed to do themselves. All students took a common test with problems like those used in instruction and also rated their mental effort. Students who had seldom or never used databases earned significantly higher test scores in the worked examples group than in the exploration group. Students who had previous experience with databases, however, earned higher test scores when in the exploration group.  Both students with and without database experience indicated that exploration required greater mental effort, with the difference being much larger for novice students. Based on this study, instructors should note that providing guidance in the form of worked examples can benefit novice learners learning new problem-solving approaches.
Kalyuga, S., Chandler, P., and Sweller, J. (2001) Learner experience and efficiency of instructional guidance. Educational Psychology, 21(1), 5-23.  This study asks whether task complexity and learners’ prior experience impact the learning from two instructional methods, worked examples and exploratory problem solving. The authors test the hypothesis that problem solving guided by worked examples is more effective for inexperienced learners but not for more experienced learners. They report two experiments with trade apprentices. In the first experiment, students learned to use a diameter-circumference conversion chart. Approximately half (n = 8) of participants received instruction with solved example problems while the other half (n = 9) received an introduction to the chart and were then encouraged to explore it as a tool to solve problems. There was no significant difference in the performance of the two groups on a subsequent test of this relatively simple task, although the worked examples group did report lower mental effort. In the second experiment, the authors compared the two instructional approaches for a more complex task (writing Boolean switching equations) after each of two training sessions. They found that students in the worked examples group performed better after the first training session (when they were more novice), but that this benefit was lost for the second training session, concluding that the guidance provided by the worked examples became unimportant as the learners gained experience. Based on these results, instructors should note that guidance in the form of worked examples can benefit novices learning a complex task, but that exploratory problem solving is equally effective when the task is simple or as learners become more expert.
Renkl, A., and Atkinson, R.K. (2003). Structuring the transition from example study to problem solving in cognitive skill acquisition: A cognitive load perspective. Educational Psychologist, 38 (1), 15-22.  This brief review provides guidance on structuring the transition from example-based learning in early stages of skill acquisition to problem solving in later stages. The authors focus on cognitive skill acquisition in well-structured domains, such as mathematics, physics, and programming. They describe the phases of skill acquisition: early phase, in which learners gain basic understanding; intermediate phase, in which learners focus on applying abstract principles to solve problems; late stage, in which speed and accuracy increase through practice. They also describe cognitive load theory, emphasizing that the limited capacity of working memory is a primary factor in instructional methods’ effectiveness. They differentiate between cognitive load that is intrinsic, extraneous, or germane (i.e., important for learning the target skill). The authors describe how worked examples reduce extraneous load during intermediate phase skill acquisition and note that self-explanation of examples can increase germane load. They also observe, however, that worked examples and self-explanation can be redundant with learners’ own knowledge in late-stage skill acquisition; these activities can thus impose extraneous load at this stage. Therefore at this stage, problem solving practice becomes more effective. Based on these considerations, the authors suggest instruction should involve a gradual increase in problem-solving demands, such as: presentation of a complete example, followed by example problems with increasing numbers of steps left to the learner, culminating in independent problem solving. The authors describe previously published results comparing this fading procedure to worked example-problem pairs. These results indicate that 1) fading is beneficial for near transfer (i.e., students being able to solve similar problems); 2) backwards fading (leaving blanks later in the problems first, then earlier and earlier) is more beneficial than forward fading; 3) prompting for self-explanation can further enhance near transfer and can promote far transfer (i.e., modification of learned solution methods). Based on these observations, instructors should note that removing guidance in stages and fostering self-explanation enhances learning from worked examples.
Atkinson, R.K., Derry, S.J., Renkl, A., and Wortham, D. (2000). Learning from examples: Instructional principles from the worked examples research. Review of Educational Research, 70 (2), 181-214.  Worked examples consist of a problem statement and a step-by-step procedure for solving the problem. They can be used as an instructional device to reduce the burden on working memory when students are in early stages of learning to solve problems in a domain, thus facilitating students’ recognition of important problem and solution features. This review identifies instructional principles for 1) formatting worked examples and 2) creating lessons that maximize their value. Beneficial formatting includes diagrams integrated with explanatory text as well as visual indicators that accentuate problem subgoals. Lesson structures that are beneficial include using at least two examples for each type of problem as well as the pairing of worked examples with practice problems. In addition, a finite set of cover stories that are used across problem types can help learners identify deep structural features rather than focusing on surface features. Finally, the value of worked examples is significantly enhanced when students engage in self-explanation, which can be stimulated by training students to self-explain or through structural manipulations of the worked example (i.e., labeling subgoals or leaving blanks for learners to fill). After identifying these instructional principles, the authors offer a framework explaining the causal relationships among these principles, suggesting that some design features directly impact students’ acquisition of problem-solving skills while others do so indirectly by enhancing the quality of student self-explanations. Based on this review, instructors should recognize key design elements for worked example instruction: individual worked examples should integrate sources of information and foster identification of subgoals within a problem; lessons should include at least two examples for a type of problem, and worked examples should be interspersed with practice problems; different problem types should use similar cover stories to emphasize deep problem features; instruction should foster student self-explanation.  
Renkl, A. (2014). Learning from worked examples: how to prepare students for meaningful problem solving. In V. A. Benassi, C. E. Overson, & C. M. Hakala (Eds.), Applying science of learning in education: Infusing psychological science into the curriculum (p. 118–130). Society for the Teaching of Psychology.  This teacher-focused review briefly describes the worked examples approach and the types of problem solving for which it has been shown to be useful. The author argues that worked examples can be valuable not only for learning algorithmic solution procedures in mathematics and sciences, but also for learning heuristic methods for a variety of purposes (e.g., to guide productive cooperation or analysis of legal cases). Ten principles that are important for effective learning from worked examples are identified, five of which are summarized here. 1) Learner self-explanations are key. Within a given worked example, students need to relate solutions to abstract principles. In addition, they need to compare examples to identify key similarities and differences.  Both training students to self-explain and prompting them to do so can help students achieve this goal. 2) If students are unable to self-explain correctly, expert explanations should be provided. This can be accomplished with automated worked examples or by think-pair-share or clicker-type questions followed by instructor explanation in class. 3) Prompting students to compare the example sets that are varied in intentional ways can help students identify deep features and develop flexible problem-solving abilities. 4) Within a given worked example, textual or auditory explanations should be integrated with visual prompts, and visual cues should be used to make it easy for students to follow the explanation. 5) Sub-goals within a solution should be presented as meaningful blocks and should be labeled or visually isolated. This approach can help students use parts of solution procedures in a variety of problems. Each principle is presented with explicit statements about its limitations, allowing instructors to determine whether the principle is relevant for their context. Instructors should use this review to consider the evidence for design elements associated with worked example instruction, determining how these elements apply in their teaching context.
Halmo, S.A., Sensibaugh, C.A., Reinhart, P. Stogniy, O., Fiorella, L., and Lemons, P.P. (2020). Advancing the guidance debate: Lessons from educational psychology and implications for biochemistry learning. CBE—Life Sciences Education, 19: ar41, 1-14.  This study compares the effectiveness of four instructional methods for helping students learn about the physical basis of noncovalent interactions, a challenging concept in biochemistry. Specifically, the authors compared the effects of worked examples that included explicit instructor explanation of problems followed by small group problem solving; productive failure, which included unguided small group exploration of a problem followed by explicit instructor explanation that draws on student ideas; unscaffolded guided inquiry, which took the form of student group problem solving supported by just-in-time aid from an instructional team; and scaffolded guided inquiry, where the problems that student groups solved were broken into small chunks to provide scaffolded knowledge building. The authors measured the effects of these conditions on basic knowledge, near transfer problem solving, and far transfer problem solving. Students were recruited from courses that serve as the prerequisite for biochemistry and completed a pretest two weeks before the instruction, which took place in a mock classroom. The authors found that student performance on a basic knowledge posttest was equal in all four instructional conditions. However, student performance on near transfer problems exhibited differences between conditions: scores for the unscaffolded guided inquiry condition were lower than for scaffolded guided inquiry, productive failure, and worked examples + practice, although there were no differences in these three conditions. Finally, the authors found no difference for productive failure and scaffolded guided inquiry for performance on far transfer questions. Based on this study, instructors should note that explicit guidance helps student performance on near transfer problems, but that the form of that guidance can vary.
Nievelstein, F., van Gog, T., van Dijck, and Boshuizen, H.P.A. (2013) The worked example and expertise reversal effect in less structured tasks: Learning to reason about legal cases. Contemporary Educational Psychology, 38, 118-125.  [Author abstract with modifications] The worked example effect indicates that learning by studying worked examples is more effective than learning by solving the equivalent problems. The expertise reversal effect indicates that this is only the case for novice learners; once prior knowledge of the task is available problem solving becomes more effective for learning. These effects, however, have mainly been studied using highly structured tasks. This study investigated whether they also occur on less structured tasks, in this case, learning to reason about legal cases. Less structured tasks take longer to master, and hence, examples may remain effective for a longer period of time. First-year (n = 75) and third-year (n = 36) law students at a Dutch university studied civil law cases in a mock classroom setting, working individually. They received either a description of general process steps they should take; worked examples; worked examples including the process steps; or no instructional support for reasoning. Results show that worked examples were more effective for learning than problem solving, both for first- and third-year students, even though the latter had significantly more prior knowledge. Further, the inclusion of process steps with the worked examples did not appear to have an effect; student performance did not differ in the +/- process step conditions. Thus a benefit from worked examples was observed for both novice and advanced students. The expertise-reversal effect that has been reported as learners become more expert was not observed for these less structured tasks. Based on this finding, instructors should note that worked examples can benefit both novice and more expert students learning to solve unstructured problems.

Return to Map

Cite this guide: Frey RF, Brame CJ, Fink A, and Lemons PP. (2022) Evidence Based Teaching Guide: Problem Solving. CBE Life Science Education. Retrieved from
  • Your feedback helps improve the site