8Synthesis and Reflections

Publication Details

This chapter opens with a synthesis of the evidence of areas of intersection between science education and the development of 21st century skills. It then summarizes participants’ reflections on the workshop and final comments from the planning committee members.


In the following discussion, which is organized around the five 21st century skills, the preliminary definition of each skill is followed by a description of how the different presenters interpreted the relationship between the skill and the curriculum models they reviewed. The section concludes by identifying common themes that emerged across the curriculum models as well as in other workshop presentations.


Adaptability is defined as the ability and willingness to cope with uncertain, new, and rapidly changing conditions on the job, including responding effectively to emergencies or crisis situations and learning new tasks, technologies, and procedures. Adaptability also includes handling work stress; adapting to different personalities, communication styles, and cultures; and physical adaptability to various indoor or outdoor work environments (Houston, 2007; Pulakos et al., 2000).

The 5E Model

Bybee noted that although the integrated sequence of instructional activities in the 5E model may support adaptability, he did not find evidence of development of this skill in the research on the 5E model.

Online Learning Environments for Argumentation

Clark reviewed four online learning environments, each of which engages students in developing, warranting, and communicating a persuasive argument and in critiquing arguments developed by others. He proposed that these learning activities may support development of adaptability in three ways. First, the environments may help students adapt their everyday communication skills to align more closely with the skills used in scientific argumentation. Research on implementation of all four environments yields evidence that students improved either in scientific argumentation (Clark, 2004; Clark and Sampson, 2005; Clark et al., 2008; Cuthbert, Clark, and Linn, 2002) or in similar forms of argumentation (Janssen, Erkens, and Kansellar, 2007; Janssen et al., 2007; Marttunen and Laurinen, 2006; Salminen, Marttunen, and Laurinen, 2007; Stegmann et al., 2007; Stegmann, Weinberger, and Fischer, 2007).

Second, Clark proposed that adaptability develops as an offshoot of gains in argumentation, as students learn how to adapt to changing information or changing contexts. Research on some of the environments provides evidence of such development. For example, studies of the Dialogical Reasoning Educational Web Tool (DREW) indicate that it helps students learn how to identify and evaluate the arguments for and against a particular position when investigating an unfamiliar topic (Marttunen and Laurinen, 2006; Salminen, Marttunen, and Laurinen, 2007). Research on the Computer-supported Argumentation Supported by Scripts-experimental Implementation System (CASSIS) indicates that it is effective in improving students’ ability to generate persuasive and convincing arguments and counterarguments (Stegmann et al., 2007; Stegmann, Weinberger, and Fischer, 2007).

Third, he suggested that these online environments develop adaptability by distributing and redistributing roles and activities to individual group members, so that they must take on new perspectives that may differ from their personal views and respond accordingly. For example, CASSIS uses scripts that guide learners to take on and rotate the roles of case analyst and constructive critic. Research indicates that the use of the scripts increased students’ ability to elaborate arguments and counterarguments and share their knowledge and perspectives in the discussions (Weinberger, 2008; Weinberger et al., 2005).

These three strands of evidence suggest that the online argumentation environments develop students’ adaptability to uncertain, new, and rapidly changing conditions.

Learning by Design

Kolodner noted that, in Learning by Design (LBD), students work on a variety of different teams throughout the school year, requiring them to adapt to fellow students with different working styles, strengths, and weaknesses. In addition, their work on multiple large challenges, each requiring different knowledge and skills, may develop adaptability.

Kolodner identified some evidence of development of adaptability in her comparison studies of LBD students and matched comparison classrooms, with students with similar levels of science achievement and socioeconomic status (Kolodner, Gray, and Fasse, 2003). The authors designed a performance task to assess the ability of student groups to design an experiment and gather and analyze experimental data. They also used written pre- and posttests, consisting mostly of multiple-choice items, to assess the students’ content learning. They found that the LBD students immediately got down to work on the performance task, while the students in the comparison classroom took more time to get into groups and begin addressing the task. In comparison to the non-LBD students, the LBD students displayed significantly higher levels of negotiations during collaboration and distribution of the task among group members (Kolodner et al., 2003; Kolodner, Gray, and Fasse, 2003). These findings suggest that LBD students developed adaptability to new performance tasks and different personalities, communication styles, and cultures.

Investigating and Questioning our World through Science and Technology

Krajcik and Sutherland (2009) observed that the Investigating and Questioning our World through Science and Technology (IQWST) curriculum challenges students to build and revise models throughout the middle grades, based on evidence related to scientific phenomena. He argued that students’ realization that models can change when new evidence is presented represents development of adaptability. In addition, he suggested that adaptability is supported by another IQWST learning goal—the expectation that, as learners become more sophisticated in constructing explanations, they will rule out other possible explanations. Krajcik observed that students need to consider if they have sufficient and appropriate evidence to support their claims and, if they do not, adapt by writing new claims that are supported by their evidence.

Research on implementation of IQWST provides evidence that students improve in modifying and revising models when presented with new evidence (Merritt, Shwartz, and Krajcik, 2008; Schwarz et al., 2009; Shwartz et al., 2008). The research also indicates that students improve in using evidence and reasoning to support their claims (McNeill and Krajcik, 2008a, 2008b; McNeill et al., 2006).

Complex Communication/Social Skills

Complex communication skills are defined as the ability to process and interpret both verbal and nonverbal information from others in order to respond appropriately. A skilled communicator is able to select key pieces of a complex idea to express in words, sounds, and images, in order to build shared understanding (Levy and Murnane, 2004). Skilled communicators negotiate positive outcomes with customers, subordinates, and superiors through social perceptiveness, persuasion, negotiation, instructing, and service orientation (Peterson et al., 1999).

The 5E Model

Bybee proposed that science curricula based on the 5E model support development of communication skills by engaging students in scientific argumentation.

He found evidence for development of argumentation in a study comparing an inquiry science curriculum based on the 5E instructional model with “commonplace teaching” of the same material, as defined by national surveys of science teachers (Wilson et al., 2009). The authors used a randomized controlled trial research design, assigning a total of 58 students ages 14–16 to either a group that was instructed based on the 5E model or to a group that received commonplace teaching. Students in the 5E group reached significantly higher levels of achievement compared with the other group in terms of three different learning goals—knowledge, scientific reasoning, and argumentation. The finding held for testing immediately following instruction and four weeks later.

Bybee also indicated that evidence of development of complex communication skills was provided by a study that found that the 5E model increased levels of higher order thinking among a small group of science students (Boddy, Watson, and Aubusson, 2003).

Online Learning Environments for Argumentation

Clark proposed that the online argumentation environments may support complex communication skills in at least two ways. First, all of the environments require students to develop, warrant, and communicate a persuasive argument, based on evidence. Because the goal of argumentation is to persuade and build shared understanding, “argumentation skills are therefore an integral component of complex communication skills” (Clark et al., 2009, p. 18). From this perspective, studies showing that students engaged with all four of these environments advance in argumentation may be seen as evidence of development of complex communication skills (e.g., Clark, D’Angelo, and Menekse, in press; Marttunen and Laurinen, 2006; Stegmann, Weinberger, and Fischer, 2007).

One way in which these environments support complex communication skills, Clark said, is through the use of scripts that orchestrate and control students’ interactions with each other and the learning environment. He cited a study demonstrating that the use of scripts in CASSIS improved the quality of argumentation (Stegmann, Weinberger, and Fischer, 2007).

Learning by Design

Kolodner observed that the LBD curriculum integrates learning activities with opportunities for guided reflection that involve communication. Students present design ideas to each other, along with the scientific principles and experimental results that support these ideas. Teachers lead whole-class discussions after these sessions to help students focus on the ways in which science concepts are applied in the designs. Students also present their experimental procedures and results to each other in poster sessions, and the teacher leads several types of whole-class discussions. All of these activities have the potential to develop complex communication skills. In addition, an explicit goal of the curriculum is to enact a set of values and expectations, one of which is collaboration (Kolodner, Gray, and Fasse, 2003).

The comparison study described above found that the LBD students displayed significantly higher levels of negotiations during collaboration and distribution of the task among group members (Kolodner et al., 2003; Kolodner, Gray, and Fasse, 2003). Negotiation is a dimension of complex communication skills.

Investigating and Questioning our World through Science and Technology

Krajcik noted that several learning activities in IQWST may support development of complex communication skills. Students use evidence and reasoning to support claims (scientific explanations), both verbally and in writing; they also evaluate and critique claims made by others, both verbally and in writing. In addition, students construct models, present their models to class members, and justify their models based on evidence. In constructing models and writing scientific explanations, they need to consider that all the evidence is accounted for. Often, students construct explanations and build and revise models in small groups, and the groups present their explanations and models to other students for critique and feedback. These small group discussions, too, may support development of complex communication skills (Krajcik and Sutherland, 2009).

Several recent studies of implementation of IQWST indicate that students improve in these activities. These studies provide evidence that engagement with IQWST improves students’ ability to support claims using evidence (McNeill and Krajcik, 2008a, 2008b; McNeill et al., 2006) and their ability to construct and communicate scientific explanations (Krajcik et al., 2008; Krajcik, McNeill, and Reiser, 2008). In addition, studies show that IQWST students improve in accounting for all evidence when constructing models (Merritt, Shwartz, and Krajcik, 2008; Shwartz et al., 2008). These findings may represent development of students’ ability to select key pieces of a complex idea to express in words, sounds, and images, in order to build shared understanding.

Nonroutine Problem Solving

The ability to solve nonroutine problems is defined as follows: A skilled problem solver uses expert thinking to examine a broad span of information, recognize patterns, and narrow the information to reach a diagnosis of the problem. Moving beyond diagnosis to a solution requires knowledge of how the information is linked conceptually and involves metacognition—the ability to reflect on whether a problem-solving strategy is working and to switch to another strategy if the current strategy isn’t working (Levy and Murnane, 2004). It includes creativity to generate new and innovative solutions, integrating seemingly unrelated information, and entertaining possibilities others may miss (Houston, 2007).

The 5E Model

Bybee identified three sources of evidence that engagement with the 5E model supports development of nonroutine problem solving. First, the comparative study discussed above (Wilson et al., 2009) measured student progress toward three goals of inquiry-based instruction, one of which was scientific reasoning. The study provides evidence that engagement with the 5E model increases students’ scientific reasoning ability in comparison to the reasoning ability of students receiving more typical forms of science instruction. Bybee noted “a linkage between scientific reasoning and problem solving” (Bybee, 2009, p. 15). Second, he mentioned a study that found increases in students’ higher order thinking following instruction based on the 5E model (Boddy, Watson, and Aubusson, 2003). Finally, he mentioned a study by Taylor, Van Scotter, and Coulson (2007), which found that students whose teachers fully implemented the 5E model were more able to apply their understanding to new situations than students whose teachers did not fully implement the model.

Online Learning Environments for Argumentation

Clark discussed the role of online argumentation environments in development of nonroutine problem solving. In WISE, students negotiate consensus and critique novel ideas rapidly introduced by other students, requiring them to examine and use a broad span of information from the initial laboratory activities, simulations, and everyday experiences. Research on students using this environment suggests that these activities increase the conceptual and structural quality of students’ argumentation (Clark, D’Angelo, and Menekse, in press). This research suggests that students improve in such elements of nonroutine problem solving as integrating seemingly unrelated information and entertaining possibilities others may miss.

The CASSIS environment encourages learners to apply attribution theory to solve authentic problems, with support from argumentative collaboration scripts. In recent studies, CASSIS learners supported with an epistemic script were better able than other CASSIS learners to focus on the core aspects of a problem case and also pursued additional information and explored multiple perspectives (Mäkitalo et al., 2005; Weinberger, 2008; Weinberger et al., 2007). These findings suggest that engagement with CASSIS supports development of such dimensions of nonroutine problem solving as learning to analyze large amounts of information, recognize patterns, and determine whether or not a claim is well supported by available evidence.

The argument diagram tool in the DREW environment can also promote nonroutine problem-solving skills. The results on DREW thus far have shown that students deepen and broaden their knowledge of a given topic when diagrams are used across three sequential phases of students’ work (Marttunen and Laurinen, 2006). The DREW diagrams have been demonstrated to support students in reflecting on their previous debate and earlier knowledge (Marttunen and Laurinen, 2007), providing evidence of development of metacognition, a dimension of nonroutine problem solving.

Learning by Design

Kolodner said that, in LBD, students work on a variety of design challenges over the course of the year, using repeated activities. The goal of these activities is to support development of science practices, such as how to design an experiment and how to interpret evidence, that students can apply not only to the immediate design challenge but also to new or nonroutine problems (Kolodner, Gray, and Fasse, 2003). Research on implementation of LBD has identified many anecdotal examples of students applying science practices developed over the course of addressing one design challenge to a new design challenge, as well as spontaneously applying science practices to their science fair projects, without coaching or prompting by the teacher.

The matched comparison study of LBD and non-LBD students described above (Kolodner, Gray, and Fasse, 2003) found that the LBD students scored significantly higher than the comparison group in designing and carrying out an experiment and analyzing the resulting data. These findings suggest that LBD students’ skills to solve nonroutine scientific problems are greater than those of non-LBD students.

The LBD curriculum aims to help students develop metacognitive strategies, such as conducting self-checks of their progress when designing an experiment, running an experiment, and analyzing the data. A comparison of average-achieving LBD students with average-achieving students taught using a traditional science curriculum, conducted only two months into the 2000–2001 school year, found that the LBD students scored significantly higher than the comparison group in conducting self-checks. These findings indicate that the LBD curriculum helps students develop metacognitive strategies, an element of nonroutine problem solving, to a greater extent than more traditional science instruction.

Investigating and Questioning our World through Science and Technology

Krajcik observed that the IQWST curriculum engages middle school learners in using evidence and reasoning to build models that describe and explain a host of different phenomena. Studies of implementation of IQWST show that learners improve in their reasoning—specifically, in taking into account sufficient and necessary evidence to support an explanatory model (McNeill and Krajcik, 2008a, 2008b; McNeill et al., 2006). Because scientific reasoning is similar to nonroutine problem solving, these findings suggest that engagement with IQWST increases students’ nonroutine problem-solving skills.


Self-management/self-development is defined as the ability to work remotely, in virtual teams; to work autonomously; and to be self-motivating and self-monitoring. One aspect of self-management is the willingness and ability to acquire new information and skills related to work (Houston, 2007).

The 5E Model

Bybee noted that the phases of the 5E instructional model, including the initial “engagement” phase and also the “exploration” phase, in which students explore natural phenomena, are designed to motivate students and increase their interest in science and science learning. He said that studies by Akar (2005), Tinnin (2000), and Von Secker (2002) provide evidence that the 5E model develops student interest in science and positive attitudes toward science learning. Increased interest and positive attitudes may represent development of self-management and self-development.

Online Learning Environments for Argumentation

Clark noted that some of the online environments include participant awareness tools that help students monitor their own contributions and the contributions of other group members, which may encourage self-development/self-management. For example, the Virtual Collaborative Research Institute (VCRI) includes a “shared space,” which analyzes chat messages and shows the extent to which group members are conducting shallow online discussions or are engaged in critical exploratory discussion. The tool also visualizes whether group members are agreeing or disagreeing about a topic during online discussion. One study (Jannssen et al., 2007) found that, in comparison to students using VCRI without the shared space tool, students with access to the tool perceived their group’s norms and behaviors more positively and their group’s strategies as more effective. In addition, students with access to the shared space tool engaged in different collaborative activities, and performed better on one part of a research task in the domain of history. Such participant awareness tools develop students’ ability to work in virtual teams, to work autonomously, and to be self-motivating and self-monitoring.

Research suggests that the inclusion of scripts in the CASSIS and WISE environments can promote self-management and self-development (Weinberger et al., 2007). In one study (Wecker and Fischer, 2007), the scripts supporting students in classifying the components of the argumentation of their learning partners and in formulating counterarguments were gradually reduced, and students began to carry out argumentation on their own. These studies indicate that CASSIS develops students’ ability to monitor and manage their own performance as well as the performance of others. Research on an early version of WISE (Davis, 2003; Davis and Linn, 2000) showed that generic prompts that ask students to “stop and think” encourage greater reflection in comparison to directed prompts that provide hints indicating potentially productive directions for their reflection. Prompts can support students’ self-monitoring, a dimension of self-management/self-development.

Learning by Design

Kolodner noted that the design challenges in LBD require students to identify what skills and concepts they need to learn, carry out investigations to learn what they need to know, and apply their learning. These activities are designed to support students in learning how to monitor and manage their own learning.

The comparison study described above (Kolodner, Gray, and Fasse, 2003) found that LBD students consistently performed significantly better than non-LBD students at conducting self-checks during experiment design, running experiments, and analysis. These findings suggest that engagement with LBD develops students’ skills in self-management/self-development of their own learning.

Investigating and Questioning our World through Science and Technology

Krajcik indicated that several different learning activities in IQWST may support development of self-management/self-development. First, students evaluate and critique their own models and scientific explanations as well as those created by others. Studies of IQWST indicate that students improved in these activities (Merritt, Shwartz, and Krajcik, 2008; Shwartz et al., 2008). Second, students use criteria to make judgments, and there is evidence that the curriculum supports improvement in this skill. Third, students consider if they have sufficient and appropriate evidence to support claims, requiring them to monitor and manage their own learning. The studies of IQWST also yield evidence of improvement in these skills (McNeill and Krajcik, 2008a). Finally, adhering to project guidelines and timelines during various projects supports growth in students’ self-management and self-development.

Systems Thinking

Systems thinking is defined as the ability to understand how an entire system works; how an action, change, or malfunction in one part of the system affects the rest of the system; and adopting a big-picture perspective on work (Houston, 2007). It includes judgment and decision making, systems analysis, and systems evaluation as well as abstract reasoning about how the different elements of a work process interact (Peterson et al., 1999).

The 5E Model

Bybee observed that understanding of systems thinking may be viewed as a necessary foundation for development and application of systems thinking as a skill. Based on this interpretation, he drew the inference that evidence of the 5E model’s effectiveness in enhancing students’ mastery of scientific subjects represents evidence of development of systems thinking. Several studies provide “strong” evidence that the model enhances mastery of scientific subject matter, Bybee said (Akar, 2005; Bybee et al., 2006; Coulson, 2002; Taylor et al., 2007; Wilson et al., 2009).

Online Learning Environments for Argumentation

Clark proposed that scientific argumentation and development of systems thinking are related, because arguments are systems and chains of claims, warrants, backings, and data that can involve substantial complexity as they evolve through discussion. In order to participate productively in these discussions, students must learn how to evaluate information, make well-reasoned decisions, and examine how the various components of an argument or counterargument fit together with one another. They must also develop criteria for evaluating what counts as warranted knowledge and how to determine if information is relevant to the phenomenon under discussion, or if there is sufficient information to make a decision. Through these activities, students learn to adopt a big-picture perspective on their work. From this perspective, all of the research showing improvement in argumentation among students engaged with these environments provides evidence of development of systems thinking.

Two studies of DREW, focusing more specifically on systems thinking, suggest that engagement with this environment develops improved understanding of a complex phenomenon or system and supports engagement in systems analysis and evaluation (Marttunen and Laurinen, 2006, 2007).

Learning by Design

Kolodner observed that working toward the design challenges in LBD requires students to develop understanding of a system or set of systems. She proposed that solving such challenges requires judgment and decision making, systems analysis, systems evaluation, and reasoning about how the different elements of a system interact. The curriculum, she said, is designed to support students in developing these dimensions of systems thinking. However, no studies to date have examined development of these dimensions of systems thinking among LBD students.1

Investigating and Questioning our World through Science and Technology

Krajcik observed that IQWST supports students’ development of systems thinking, but no assessments have been conducted specifically to measure systems thinking. IQWST is designed to enhance student understanding of complex scientific systems, and such understanding may develop systems thinking. The effectiveness of the curriculum in supporting learning of complex scientific content has been studied in large urban areas, suburban areas, and rural areas, including areas with populations of students eligible for free and reduced-price meals. These studies, using pre and posttests, have all shown statistically significant gains in the learning of science concepts (McNeill and Krajcik, 2008a; McNeill et al., 2006; Merritt, Shwartz, and Krajcik, 2008).

Common Themes

Several common themes appear across the promising curriculum models and in other workshop presentations. First, the paper authors’ interpretations of the five skills are often quite similar. Clark and Krajcik both view the construction and revision of scientific models, arguments, and explanations as processes that develop adaptability. All of the authors viewed the creation and communication of scientific arguments and explanations—verbally and in writing—as activities that develop complex communication skills. In addition, Clark, Kolodner, and Krajcik identified student work in small groups as supporting development of complex communication skills.

Similarly, all of the authors viewed engagement of students in inquiry and development of scientific explanations as supporting development of nonroutine problem solving, and most viewed improvement in students’ understanding of complex scientific systems as evidence of development of systems thinking. These authors’ shared interpretations of the skills reinforce Schunn’s finding that there are many areas of overlap between the science education goals embodied in state and national standards and the five skills.

Second, underlying these common interpretations of the five skills are similar instructional design approaches in the curriculum models. All of the curricula take a problem-based or project-based approach. These curriculum models embed learning of science content (or content in other domains) in investigations and discussions focusing on real-world phenomena or challenges. They are designed to motivate students to learn by engaging them around a driving question, problem, or challenge.

Reinforcing this theme, Anderman and Sinatra recommended that science teachers promote active engagement in inquiry and problem solving, based on connections to adolescent students’ personal interests and career goals (see Chapter 3). Windschitl also suggested that the five skills can be developed through problem- or project-based learning, including through scientific inquiry, and proposed that teachers should reconstruct curriculum around a few big ideas in science. Taken together, these presentations suggest that science instruction that embeds student learning in the investigation of real-world problems or phenomena and focuses on a few selected driving questions related to these problems or phenomena is most likely to support development of 21st century skills.

A third theme is that advancing such forms of science instruction would require not only new curriculum designs, but also increased capacity to support teachers. Windschitl called for a continuous improvement system to support teachers in cultivating students’ 21st century skills, cautioning that this would require major reforms in science teacher preparation, induction of new teachers, and ongoing professional development. Anderman and Sinatra indicated that adolescents’ cognitive capacity to develop the five 21st century skills can be tapped if teachers can motivate them with new teaching and assessment strategies. However, they said that teachers need support from administrators for shared lesson planning and other forms of professional development, as well as training in adolescent development, to design and implement these new strategies. Several other workshop participants highlighted the importance of building capacity to support science teachers in fostering students’ 21st century skills.


William Sandoval invited all participants to return to the small discussion groups they had been assigned to on the first day. He asked each group to think about and discuss policy options that might support development of 21st century skills in science education and to suggest at least one short-term step that could be taken immediately and one longer term policy option.

Christine Massey (University of Pennsylvania) invited a reporter from each group to share these suggestions. Hanna Doerr (National Commission on Teaching and America’s Future) reported that her group’s suggestion that, in the short term, the National Council for Accreditation of Teacher Education, the National Association for the Education of Young Children, the Teacher Education Accreditation Council, and other organizations concerned with teacher preparation begin a discussion about integrating 21st century skills into their standards for teacher education and certification. The group’s longer term goal is to reform teacher education in order to help future teachers learn 21st century skills and prepare to teach these skills. For example, schools of education could engage future teachers in project-based learning and collaborative study of student assessment results and other student materials.

David Vanier (National Institutes of Health Office of Science Education) reported that, in the short term, his group proposes using some of the money Congress provided to the U.S. Department of Education through the American Recovery and Reinvestment Act of 2009 (commonly known as the stimulus package) to infuse 21st century skills into K-12 education. For example, the funds could be used to support development of education school curriculum materials focused on 21st century skills, to provide grants for teacher education related to 21st century skills, or to forgive student loans to teachers who incorporate 21st century skills. Over the longer term, the group calls for reforming national education standards, both in science and in other school subjects, to incorporate 21st century skills.

The next reporter said that the group proposes, in the short term, to clarify the terms and definitions of 21st century skills, as a common basis for a possible future workshop on assessment of 21st century skills. This group agreed that the skills fall into four categories, including problem solving and critical thinking, flexibility and adaptability, communication skills, and some form of self-direction. Over the longer term, the group advocated development of a research and development agenda for science education that would show how 21st century skills are incorporated into the teaching of science content and provide concrete examples of what these skills look like in curriculum, instruction, and assessment.

Susan Albertine said that her group discussed how the five 21st century skills compared with the goals of science education reform. The members agreed that two of the skills demanded by business—complex communication and nonroutine problem solving—were well aligned with approaches that are gaining support in reform of K-12 and college-level science education, such as problem-based learning, inquiry, and engaging students in design. Albertine said the group proposes, in the short term, to increase clarity about the alignment between 21st century skills and science education reform goals. Over the longer term, this group suggests carrying out a longitudinal study of children to understand what happens over time when they participate in learning environments that emphasize 21st century skills.

Gina Schatteman described her group’s short-term goal: to use technology as a leverage point for changing the education system, specifically by using online tools to monitor both student and teacher progress and providing both online and face-to-face mentoring of teachers. The group’s long-term goal is to align improved science standards and curriculum to convey 21st century skills, incorporating a progression of learning across grade levels.

The final reporter said the group’s immediate action step would be to define the 21st century skills more clearly and operationally. Over the long term, this group thought that universities receiving science research grants from the National Science Foundation should be required to provide expanded research experiences to undergraduates. This suggestion is based on research indicating that such experiences build students’ appreciation for both the content and process of science, which may support self-management/self-development (Kardash, 2000).


Massey observed that several of the groups asked for clearer definitions of 21st century skills, including their relationship to the goals of science education. She also noted common themes in the areas of incorporating 21st century skills into education standards and assessments and connections between K-12 and higher education, as groups called for changes in colleges of education and in undergraduate science courses, in order to develop teachers’ 21st century skills.

Jacob Foster (Massachusetts Department of Education) observed that engaging students in environmentally focused design projects appears to support learning of 21st century skills (Krajcik and Sutherland, 2009; Schunn, 2009). Currently, he said, he must work with three different sets of state education standards, addressing science, technology, and engineering, which can be challenging and confusing. He suggested that, over the long term, it would be valuable to bring together the societies from these three disciplines to develop more integrated standards. As part of this effort, the societies could analyze how the standards relate to, and promote learning of, 21st century skills.

Patricia Harvey (National Research Council) said that members of her group had discussed development of 21st century skills in informal learning environments, such as science museums and field trips. Massey responded that learning outside formal school settings had been mentioned several times and suggested that it would be valuable to involve experts in this field in future activities focusing on 21st century skills.

Ken Kay asked why, when several groups identified selected skills as most relevant to science education, they had dropped systems thinking, a skill he views as an important component of science. Christian Schunn responded that his group left it out because it was unsure whether systems thinking at work differed from scientific analysis of systems. He said this group viewed adaptability and self-management/self-development as elements of high-quality science education, but not as direct goals of science education reform. Based on this view, the group suggested focusing on nonroutine problem solving and complex communication/social skills.

Janet Kolodner said that only one group had narrowed the list of five skills to two (complex communication/social skills and nonroutine problem solving) and that this might be a minority view. She reminded Schunn that he had earlier suggested engaging students in large team projects to develop self-management, indicating that this skill should be a goal of science education. She went on to say that Joseph Krajcik had described students engaged in systems thinking while learning science and that that systems thinking is an essential component of understanding scientific concepts. Although many groups asked for improved definitions of the skills, especially in order to develop assessments, she said, “it would be self-defeating” to assume that all five skills can’t be developed in the context of science education.


In the final workshop session, members of the planning committee offered individual reflections on the intersection of science education and 21st century skills. William Bonvillian began by thanking all participants for sharing their thoughts and insights throughout the workshop. He expressed his view that the commissioned papers had sparked coherent discussions that, in turn, had begun to translate abstract ideas about workforce skills into real actions that could be taken to infuse these skills into science education. He called for further thinking about how best to bring insights from the workshop into science curriculum development, standard setting, teacher professional development, and the creation of technology-enhanced materials.

Marcia Linn offered special thanks to the industry representatives for helping her and the other science education experts to understand their goals and for listening patiently to the science education experts’ different vocabulary to describe skills. She said she was especially impressed by the industry representatives’ comments about growing demand for 21st century skills, driven by flat organizations, globalization, and especially by “the fact that everyone changes jobs and responsibilities all the time.” Recalling warnings decades earlier about the increasing skill demands of work, Linn said that these warnings have now become reality. She observed that her adult children, like most young adults, change jobs frequently, so they need different skills from those of previous generations.

Linn commented that the theme of systems thinking had emerged frequently, as the workshop participants considered how to meet the increasing demand for 21st century skills in the context of a large, complex education system that is slow to change. She said that technology may offer a leverage point to meet this challenge. She found it puzzling that some education policy leaders argue that schools do not require advanced technology, given the reality that 90 percent of children use technology every day at home, and over 60 percent of middle school and high school students have their own personal websites (Lenhart, Madden, and Hitlin, 2005). She argued that technology skills are important for the 21st century, and students should be able to use and further develop these skills for learning at school.

Linn said that there are many excellent examples of technology-supported science learning environments that encourage students to reflect on and assess their own learning and also allow teachers to view these reflections and respond in real time to student ideas (Linn, 2006; Linn and Eylon, 2006), including the examples discussed at the workshop. Because these systems not only enhance student learning but also provide feedback to teachers for use in changing their instructional practices, Linn said, they offer an important avenue for large-scale change in the education system.

Massey thanked workshop sponsors Ken Kay and Bruce Fuchs as well as the Board on Science Education staff, describing the workshop as “an amazing experience.” She said that the workshop had reminded her of the progress being made in science education, including development of good pedagogical models, curriculum materials, and technological learning tools. Research has yielded new insights into design of effective teacher preparation and professional development programs and innovative assessment methods, and these developments can be linked together in more comprehensive science education reform. Massey observed that support is growing for this type of comprehensive reform of science education at all levels, from preschool through the undergraduate level. She suggested that, in thinking about how science education intersects with 21st century skills, it would be valuable to select from, and combine, this array of promising new developments in science education.

Noting that the workshop was designed with an open mind about the extent to which science education might intersect with 21st century skills, Massey said that the presentations and discussions had illuminated many promising intersections. She suggested that communication, collaboration, and systems thinking would be required to further understand and develop these intersections. It is important, she said, to clarify and articulate a synthesized message about the strongest and most promising areas of intersection, framed so that different constituencies (science educators, the business community) can support it and recognize their own interests.

Reflecting on conversations about leverage points in the education system, Massey recalled Mark Windschitl’s graphic illustration of the challenge of changing many interrelated components of K-12 and higher education at once, in order to support teaching of 21st century skills (see Figure 6-1). She also recalled Christian Schunn’s depiction of many system components, including state and national science standards, state tests, and national organizations, all influencing classroom science teaching (see Figure 2-1). Massey said that workshop participants had begun to identify potential leverage points in the education system, including assessments, science standards, teacher certification requirements, and changes in individual schools. Although there is still a great deal to think about, the workshop provided some depth in moving forward to tackle problems in science education, she concluded.

Carlo Parravano suggested that a consensus study might be valuable to illuminate what good science teaching would look like if it incorporated standards-based science content intertwined with 21st century skills. Such a study could be modeled on Taking Science to School: Learning and Teaching Science in Grades K-8 (National Research Council, 2007a), which is based on research, and its companion volume, Ready, Set, Science: Putting Research to Work in K-8 Science Classrooms (National Research Council, 2008c), which translates that research for practitioners. Parravano said that these two reports have moved the field of science education a great deal in the year and a half since they became available, and a similar effort on science education and 21st century skills would provide a common vision, language, and framework. It would provide guidance for teacher professional development, curriculum and assessment development, and future research needs related to 21st century skills and science education.

In terms of leverage points, Parravano said that parents may be particularly effective. He relayed a favorite observation of a colleague who is closely tied to the policy world—”evidence doesn’t move legislators, the public does.” To the extent that this is true, he said, further work in this area would help to build a strong vision among the parents and the public about 21st century skills that science education aims to develop. Finally, Parravano said it is important to remember that “the 21st century skills are really a tool and not an end point.” A problem with science standards, he said, is that they are now seen as an end point, rather than as a tool to improve science teaching and learning. As a result, they are rewritten in a way that makes their original goals no longer recognizable, and they lose meaning. He suggested working hard to protect the shared understanding of 21st century skills that had developed over the course of the workshop.

William Sandoval added his thanks to all participants and thanked the Board on Science Education staff for giving him the opportunity to work with, and learn from, the other members of the workshop planning committee. He agreed with other committee members that it was valuable to bring people representing different constituencies together to build shared understanding. Sandoval said he believed that the workshop had provided good answers to the six guiding questions developed by the planning committee (see Chapter 1).

Sandoval said he thinks, based on the workshop, that there are extensive and strong areas of overlap between models of high-quality science education and 21st century skills, and that these overlaps represent a positive development. At the same time, however, he cautioned that there are many barriers that stand in the way of systemic change in science education. He encouraged the audience to focus on the promising models of science curriculum, teacher education, and assessment (discussed at the workshop) and promote the expansion of these models in their home communities. He repeated his concern about several important groups that were not represented at the workshop, including parents, scientists who teach undergraduates, and teachers, and suggested including these groups in future conversations. While acknowledging the value of common definitions and shared understandings of 21st century skills, he suggested that these future conversations would also benefit from the “healthy pluralism” that is a hallmark of democracy. Finally, Sandoval explained that he was initially skeptical of the workshop’s focus on the skill demands of the economy, including the preliminary definitions of the five skills in job contexts (see Box 1-1). He called for extending the purpose and definitions of these 21st century skills to include civic dispositions and other skills needed to participate effectively in a complex, technologically sophisticated democracy.

Arthur Eisenkraft said that, although he was enriched by the workshop papers and discussions, he does not yet fully understand the meaning of the five 21st century skills. He recalled his search several years ago to identify the top 20 high school physics students in America in order to engage in an international competition. After several rounds of testing, he sent out a very difficult physics problem by mail, but he deliberately left out one crucial ingredient. He did not explicitly mention that the particle was moving at right angles to the field, although the picture he included seemed to show this.

Eisenkraft said he received three different types of written responses on the students’ test booklets. Students in one group wrote that, because they did not have complete information, they would not solve the problem. Students in a second group wrote that, lacking complete information, they assumed from the picture that the particle was moving at a 90 degree angle, and they solved the problem. A third, much smaller group, wrote that, although they did not have complete information, they had assumed the 90 degree angle and solved the problem. However, students in this group did not stop there. They explained that, if the particle was not moving at a 90 degree angle, they could not solve the problem, but they nevertheless offered their best guess about whether the solution would change. These were the students he wanted for the competition, Eisenkraft, said, because they had 21st century skills.

He suggested that educators want students to understand when to be adaptable and how adaptable to be. He asked for improved definitions of 21st century skills, with examples of how individuals deploy these skills when carrying out specific tasks at work. It would be helpful, he said, to have a detailed description of the ways in which a janitor with 21st century skills mops a hospital floor to compare with a similarly detailed description of the ways in which a janitor lacking these skills would perform the same task. Such descriptions, he said, could help educators discuss and more clearly define, the skills. They could also help to clarify when 21st century skills should be deployed. Eisenkraft noted the example of Chuck Yeager, a well-known test pilot in the U.S. Air Force. Yeager, he said, rejected the use of instruction manuals describing routine flight procedures and preferred to learn through the actual process of flying. Turning to the field of surgery, Eisenkraft said most surgeries are routine, and the patient prefers the surgeon to follow well-established, successful procedures, rather than being adaptable. Occasionally, however, when something goes wrong, the surgeon’s ability to adapt and improvise solutions to nonroutine problems suddenly becomes critical.

Eisenkraft concluded that the workshop was only the first step in a continuing process, with many questions related to the intersection of science education and 21st century skills yet to be answered. He said that the papers had enriched the discussions, and that the active participation of individuals from a variety of constituencies had encouraged all participants to think from different perspectives.

In closing, Ken Kay noted that state education agencies and school districts are at different stages in their efforts to infuse 21st century skills into science and other subjects. Reflecting that the Partnership for 21st Century Skills has established a vision (2003, 2009), he observed that West Virginia is beginning to implement this vision, and other states still have many questions. Some school districts, he said have developed rubrics to guide teaching and learning of 21st century skills, while others are just beginning to develop such rubrics. Similarly, states and school districts are at different stages of incorporating 21st century skills into teacher professional development. In the context of this continuum of different stages of movement toward 21st century skills, he said, the workshop papers and discussions are very helpful.



Other workshop presenters viewed student gains in understanding of complex scientific systems as evidence of development of systems thinking. Comparative studies of LBD and non-LBD classrooms indicate that LBD students consistently learn science content as well as or better than comparison students, with the largest gains among economically disadvantaged student and students who tested lowest on the pretest (Kolodner et al., 2003).