Exploring the Intersection of Science Education and 21st Century Skills: A Workshop Summary (2010)

This chapter continues the focus on promising new models for developing 21st century skills through science education begun in the previous chapter. Research conducted on both of the curriculum models described in this chapter provides support for a key premise of the workshop—that deep understanding of science content and development of skills in science processes (e.g., designing an investigation, formulating a scientific explanation based on evidence) are closely connected and mutually reinforcing (National Research Council, 2007a). The science process skills developed by students exposed to the models include facets of the five 21st century skills listed in Box 1-1.

Janet Kolodner (Georgia Institute of Technology) explained that Learning by Design (LBD) is a project-based inquiry approach to learning science and scientific reasoning in the context of design challenges, developed by her research group in the late 1990s and early 2000s (Kolodner, 2009). Middle school students work in small groups on design challenges that require targeted science, scientific reasoning, collaboration, communication, and planning.¹ The curriculum includes a set of units that cover a half-year each of earth and physical sciences. The design challenge is one that can be achieved in the physical or virtual world, for example, designing and building a vehicle that can navigate a certain terrain (to learn about motion

and forces) or designing and modeling an erosion control system (to learn about the earth’s ground processes).

Although the design challenge creates an authentic need for learning the targeted science content and skills, Kolodner emphasized that design challenges are not required in order to support the learning of complex skills. She said she is currently applying the overall approach, in collaboration with others, to develop middle school science curriculum units as part of the Project-Based Inquiry Science project.²

Kolodner outlined several learning goals targeted in the LBD curriculum. First, the team hoped to develop students’ knowledge and skills in design, including understanding a project design challenge, planning and managing time, aiming for shared solutions with understanding, developing specifications and criteria, managing trade-offs, understanding and working with real-world constraints, and gaining experience in the iterative process of design. Second, the curriculum was designed to develop knowledge and skills in science practices, including identifying what needs to be investigated and carrying out an investigation well. One of the science practices targeted for development is informed decision making, which includes reporting on and justifying conclusions and judging the trustworthiness of experimental results in order to use evidence appropriately to inform decisions. In addition, the LBD curriculum aims to help students learn to develop and articulate scientific explanations.

Kolodner said that enhancing learners’ collaboration skills is another important goal of the LBD curriculum. Components of this broad goal include supporting the development of teamwork, collaboration across teams, and giving credit to individuals and teams. Finally, the curriculum aims to develop science content knowledge consistent with middle school objectives. Kolodner explained that the curriculum was originally developed with a focus on technology education, but it was revised to place greater emphasis on science content knowledge, in order to align with state and national science standards.

The LBD curriculum is based on a constructivist model of learning called case-based reasoning (Kolodner, 1993; Schank, 1982, 1999), which suggests several principles for promoting learning of complex skills, such as the five 21st century skills. First, the model suggests that learning is

enhanced when learners have goals that they want to pursue, because they will reflect on their progress towards achieving those goals and seek explanations when their progress is not as expected. Second, it suggests that learners should have experiences that allow them to try out targeted skills in the context of working towards their goals, analyze whether they are achieving them through those skills, identify what they need to do better, and have the opportunity to try again. Third, it suggests that learners need multiple opportunities to try out each of the skills they are learning. Fourth, the model suggests that, in order to track their progress toward their goals, learners need to be able to easily identify the effects of what they are doing. Fifth, because identifying these effects may be difficult, the model suggests that learners be helped to analyze feedback, identify what they are doing well and not as well, and generate ideas about how to perform more productively. Finally, so that skills are learned in a way that is durable and transferable to new situations, learners should practice the targeted skills in varied contexts that are representative of the kinds of situations they will encounter outside the formal learning environment.

Kolodner noted that these principles for learning complex skills are consistent with the cognitive literature on skills learning and transfer (National Research Council, 2000) and on learning through shared reflection on practice (Lave and Wenger, 1991; Wenger, 1998).

In addition to reflecting the research on case-based reasoning, the LBD design was informed by problem-based learning (Barrows, 1985; Koschmann et al., 1994), an approach that has been used and studied extensively in medical schools. Problem-based learning integrates coaching, scaffolding, and reflection into learners’ problem-solving experiences in order to develop both targeted content and the reasoning needed to solve problems.

All of these bodies of literature informed the creation of the LBD curriculum, designed to encourage learning from experience. Kolodner presented an illustration of the two related cycles of experiences incorporated in the curriculum (see Figure 5-1).

The first iterative cycle involves design and redesign. It begins with “understanding the challenge,” or knowing what needs to be accomplished. An understanding of the design challenge leads students to develop ideas for what they “need to do” and “need to know” in order to meet the challenge. These ideas lead the learners into the second cycle, “investigate-explore,” in which they obtain and share information that they need to know to inform construction and testing of the design. If the test indicates that the design does work well enough, this leads to new ideas about what the students need to do and know, and the cycles continue.

FIGURE 5-1 Design and investigation cycles in Learning by Design.

SOURCE: Kolodner (2009).

Kolodner said that comparisons of learning among students using LBD and matched classes of science students (in terms of science achievement level, teacher understanding of the material, and socioeconomic status) indicate that the curriculum develops all five of the skills used as a framework for the workshop (Kolodner, Gray, and Fasse, 2003). She argued that students learn adaptability as they work on multiple large challenges throughout the year and over several years and join different small groups. They learn complex communication skills and social interactions through presentations and discussing and writing down their ideas on a project white board. Because the challenges are complex, students need each other’s ideas in small groups, and each group needs the results of the other groups, further developing communication and social skills.

Nonroutine problem solving develops through engagement with the curriculum, Kolodner said, because the students work on a variety of problems that may have many good answers. Self-management is a key goal. Students are supported in learning self-management through “launcher units,” which gradually introduce the practices of scientists and engineers in contexts in which the value and purpose of these practices are clear. Self-management is also enhanced through scaffolding that helps learners be successful in engaging in design/redesign and investigation activities and by emphasizing practices and reflecting on them. Finally, students develop systems thinking in the course of working to achieve design challenges that require systems thinking, with appropriate supports.

Reflecting on lessons learned in the design and testing of this curriculum model, Kolodner emphasized several features that support learning

of complex skills. First, emphasizing practices pushes learners to think about how they are doing things, what works and what does not. Second, introducing science practices through the launcher units helps students understand the importance of these practices while also providing an initial opportunity to use and learn about them. These introductory units promote creation of a learning community and a positive classroom culture. Third, repeated public presentations help learners develop internal scripts (Schank and Abelson, 1977) that make both presenting and listening and asking questions feel automatic. These public presentations help learners reflect on and improve their approaches to investigations and design/redesign, in a context of authentic need. Overall, she said, the curriculum promotes a shift in the roles of teachers and students toward increased initiative by the students.

Joseph Krajcik (University of Michigan) opened the presentation of his paper (Krajcik and Sutherland, 2009) by explaining that Investigating and Questioning our World through Science and Technology (IQWST) is a large, multi-institution curriculum research and development project under way since 2000. The primary goal is to help middle school students develop integrated understanding of core ideas of science through coherent curriculum materials (Shwartz et al., 2008). He noted that the project team has implemented and studied elements of the curriculum in both affluent suburban schools and inner-city schools.

Krajcik outlined the goals of the IQWST project:

• Design, develop, and assess the next generation of middle school science materials;

• Enable teachers to effectively teach students with a variety of backgrounds;

• Explore core ideas from each scientific discipline each year; and

• Support students in building sophisticated and systematic understanding of scientific ideas and practices.

The key features of IQWST include coherence, development of curriculum driven by learning goals, a focus on learning big ideas of science over time, and project-based learning.

Krajcik said that curriculum coherence is valuable, because research

has shown that it leads to integrated understanding in learners (Kali, Linn, and Roseman, 2008; Linn and Eylon, 2006; Schmidt, Wang, and McKnight, 2005). The developers of IQWST aim for coherence in learning goals, by selecting key goals that build on each other. They also strive for coherence within each 8-10-week project-based curriculum unit, coordinating among content learning goals, scientific practices, and curricular activities (Krajcik and Blumenfeld, 2006). In addition, the curriculum developers seek coherence across the separate units, coordinating the units to support how big ideas in science connect with each other.

IQWST curriculum materials are built around the big ideas of science that can help students understand the natural world—such as the particulate nature of matter. The focus on big ideas includes development of both science content and scientific practices and allows designers to revisit ideas throughout the curriculum so that student understanding becomes progressively more refined, developed, and elaborated across different science disciplines.

Krajcik described the IQWST approach to integrating goals for content and skills through “learning performances.” He noted that science standards often call for students to “know” or “understand” a science concept, while also including separate goals for science process skills. In IQWST, however, learning performances describe not only what it means for a learner to understand a concept, but also how the student should apply the concept, using scientific reasoning and other skills. Learning performances are based on the research team’s view that students cannot learn science content without practice, and they cannot learn science practice without content.

Krajcik presented an example learning performance that integrates the content and practice standards for grades 5-8 that are included in the National Science Education Standards (National Research Council, 1996). The content standard focuses on understanding that the properties of a substance are independent of the amount of the sample, and the practice standard focuses on using evidence to develop explanations. The performance standards call on students to construct a scientific explanation that includes a claim, evidence, and reasoning to support the concept that different substances have different properties.

Krajcik observed that the focus of IQWST on learning performances and scientific practices supports the development of adaptability, although

he did not find evidence of this (see Chapter 8). IQWST also supports development of complex communication skills, Krajcik said, noting that research on IQWST provides evidence of improvement in students’ construction of written explanations. Krajcik argued that three features of IQWST—its emphasis on coherence, its focus on big ideas in science, and its emphasis on constructing explanatory models—support students in solving nonroutine problems (see Chapter 8). He added that two features of IQWST—its inclusion of scientific practices and its focus on evaluating and revising models—support growth in self-development. Reflecting on one’s understanding is critical to self-development. In IQWST, learners need to consider if the model they have constructed accurately represents the phenomenon being studied. If not, then learners need to revise their models to more accurately represent it. There is evidence that students’ evaluation and revision of models improves during and across units.

In addition, monitoring an investigation is critical to self-management and self-development. For example, in the eighth grade chemistry unit, How do I get the energy to do things?, students design and carry out a long-term experiment in which they have to study one variable that influences plant growth. They manage data collection over a 5-week period, requiring them to monitor their investigations several times a week. Krajcik repeated that IQWST is built around big ideas in science, and that tracking how one part of a scientific system affects the rest of it is a critical aspect of developing systems thinking. For example, students in sixth grade biology track the flow of energy in an ecosystem; students in seventh grade chemistry consider the mass changes in closed and open systems; and students in eighth grade chemistry investigate how matter and energy move between organisms.

Krajcik concluded that various studies support the claim that middle school students can learn both scientific concepts and practices through engagement with the IQWST curriculum materials. He noted that, if constructing scientific explanations, building and revising models, designing investigations, and building products reflect understanding of the five 21st century skills, then there is evidence that students can learn these skills. The greatest challenge is encouraging students to use the reasoning component when constructing scientific explanations.

Krajcik cautioned that, although the IQWST coherent curriculum materials have the potential to produce a populace that is scientifically literate and prepared for the new skill demands of the 21st century, this hypothesis requires further empirical support. He noted that, although the national field test currently under way will provide some data, the IQWST materials

need to be tested using a more careful experimental design in which teachers are randomly assigned to using the IQWST materials or some other materials, and then teachers using the IQWST materials must be tracked to see if they are implementing the unit according to the designers’ intent. Observing that teacher implementation of materials affects student learning (McNeill and Krajcik, 2008b), he said it is important to provide intense professional development to help teachers use the materials as intended.

In closing, Krajcik thanked his colleagues in the IQWST project, including Brian Reiser, Northwestern University; David Fortus, Weizmann Institute of Science in Israel; his coauthor LeeAnn Sutherland of the University of Michigan; and many graduate student contributors. He thanked the teachers who have been willing to test the materials in their classrooms, and National Science Foundation Program Officer Gerhard Salinger for his support of the project through the years.

Following the two presentations, moderator Carlo Parravano invited the audience members to write down their reflections and discuss them with a neighbor. After a few minutes, he called for questions. The first questioner asked how to bring these promising models into classrooms, given the data showing their effectiveness. Kolodner responded that the LBD approach has been incorporated into a comprehensive middle school curriculum called Project-Based Inquiry Science (PBIS). The new curriculum includes units lasting 8 to 10 weeks that are carefully sequenced, so the teacher is not required to decide when to introduce topics or learning activities. She added that publication of the materials has encouraged more teachers to use them.³

This exchange led to a discussion about adoption of published curriculum materials. Krajcik said that the IQWST materials are being used in the Lubbock, Texas, school district, which has launched an initiative on writing across the curriculum. The IQWST focus on writing scientific explanations fits well with this initiative. He noted that, although it is difficult to respond to critics who say that the IQWST materials do not meet all of the Texas science standards for grades 6 through 8, it is valuable to emphasize the broader goals of the materials, such as helping students learn to construct arguments and write scientifically. The teachers and principals in the district who are using the IQWST materials are very enthusiastic about them, he said. They are delivering presentations about the curriculum materials and promoting their use to parents and community leaders. Kolodner said that

the same process is happening among principals and teachers who have adopted the PBIS materials.

Douglas Clark asked how the approaches to assessment used in LBD and IQWST could inform design of large-scale assessment of science process skills and 21st century skills. He observed that widely used multiple-choice examinations cost only about ten cents per student to administer, whereas the Organisation for Economic Co-operation and Development’s Program of International Student Assessment costs about $42 per student to administer.

Kolodner said she was not sure that the approaches to performance assessment used by her research team could be scaled up, because they require an entire class period. She explained that, in these assessments, students are asked to solve a novel problem, working in small groups. The research team makes video and audio recordings of the discussion of the problem and also asks students to write down how they would design an investigation, gather data, and formulate an explanation to solve the problem. The team analyzes the recordings and written documents to assess individual and group performance.

Krajcik said that an outside evaluator is conducting a study that tracks the performance of IQWST students and a matched comparison group of similar students over three years, from sixth through eighth grade. The evaluator originally designed multiple-choice assessments matched to the benchmarks for these grades included in the National Science Education Standards. For four years, Krajcik said, the IQWST advisory board expressed its disagreement with this assessment instrument. In the fifth year, the evaluator finally agreed to include items in the assessment focusing on the use of evidence and the construction and revision of explanatory models. He noted that it is challenging to create such test items, but they will be included in the assessments administered to the IQWST group and the comparison group when the two groups are in the seventh and eighth grades. Krajcik noted that the IQWST team has just begun collaborating with some of the best assessment experts in the nation to develop new approaches to tracking development over time in students’ understanding of how matter interacts and changes.

In response to a question about integrating learning of science content knowledge and skills with learning of other subjects, Kolodner said that she has observed this in some schools. As a practical matter, she said, when researchers or curriculum developers are creating science curriculum materials, they cannot assume that the science teacher will use the materials in collaboration with teachers of other subjects. Krajcik said that many teachers using IQWST in different parts of the country connect the curriculum’s focus on explanations with other subjects. These teachers, he argued, recognize that argumentation can be useful in English and history classes.

A participant observed that the LBD curriculum model emphasizes students’ need to know certain science concepts and processes and asked Krajcik whether the need to know also plays a role in the IQWST materials. Krajcik responded that each IQWST curriculum unit begins with a large driving question that builds not only coherence of learning activities but also students’ motivation. Typically, he said, the unit begins by engaging students in activities related to a phenomenon in order to see the importance of the driving question. For example, in one unit, the teacher asks students to close their eyes and then releases an odor into the classroom. In this unit, the students return several times to this opening encounter with the phenomenon of odor, building models to explain why something that is a source can reach their noses. Through this process, they gain understanding of the particulate nature of matter and the process of evaporation.

A participant asked whether students become more aware of their own acquisition of scientific processes through engagement with the curriculum models. Krajcik responded that the IQWST materials are extremely explicit about this. For example, when introducing the concept of scientific explanations, students are given a problem and a proposed explanation and invited to comment on the quality of the explanation. The materials explicitly describe what a scientific claim is, what constitutes evidence, how reasoning is used, and the role of each component in building a scientific explanation. Kolodner said that the LBD curriculum model uses the same approach, with explicit description of what a claim is and what constitutes evidence to support a claim. Students share their explanations with the class, discuss what makes one explanation better than another, and develop a whole-class explanation that they can all agree on. Krajcik said that the IQWST materials also engage students in publicly sharing their explanations and obtaining feedback from other students in order to improve their explanations. They encourage students to provide helpful feedback on other students’ explanations, such as noting if an explanation lacks evidence or reasoning.

Reflecting on the session and the previous day’s session on promising models (see Chapter 4), Parravano said that “there is very, very good indirect evidence … that these materials really are able to develop 21st century skills.” He noted that all of the presenters had emphasized the importance of fidelity in delivery of the curriculum models, commenting that this finding underlined the importance of the upcoming workshop session on teacher readiness for 21st century skills, discussed in Chapter 6.