JOURNAL OF RESEARCH IN SCIENCE TEACHING VOL. 45, NO. 1, PP. 53���78 (2008) Scientific Explanations: Characterizing and Evaluating the Effects of Teachers��� Instructional Practices on Student Learning Katherine L. McNeill,1 Joseph Krajcik2 1 Lynch School of Education, Boston College, 140 Commonwealth Avenue, Chestnut Hill, Massachusetts 02467 2 School of Education, University of Michigan, Ann Arbor, Michigan Received 13 June 2006 Accepted 16 January 2007 Abstract: Teacher practices are essential for supporting students in scientific inquiry practices, such as the construction of scientific explanations. In this study, we examine what instructional practices teachers engage in when they introduce scientific explanation and whether these practices influence students��� ability to construct scientific explanations during a middle school chemistry unit. Thirteen teachers enacted a project-based chemistry unit, How can I make new stuff from old stuff?, with 1197 seventh grade students. We videotaped each teacher���s enactment of the focal lesson on scientific explanation and then coded the videotape for four different instructional practices: modeling scientific explanation, making the rationale of scientific explanation explicit, defining scientific explanation, and connecting scientific explanation to everyday explanation. Our results suggest that when teachers introduce scientific explanation, they vary in the practices they engage in as well as the quality of their use of these practices. We also found that teachers��� use of instructional practices can influence student learning of scientific explanation and that the effect of these instructional practices depends on the context in terms of what other instructional practices the teacher uses. �� 2007 Wiley Periodicals, Inc. J Res Sci Teach 45: 53���78, 2008 Keywords: chemistry inquiry middle school science classroom research Classrooms are complex systems where many factors influence student learning, including teachers, peers, and other resources (Lampert, 2002). Recent research (Reiser et al., 2001) and reform documents (American Association for the Advancement of Science, 1993 National Research Council, 1996) argue that the role of the teacher is essential in structuring and guiding students��� understanding of scientific inquiry, a key learning goal in recent science education reform efforts. Teachers need to support students in making sense of these scientific inquiry practices (Driver, Asoko, Leach, Mortimer, & Scott, 1994). We are interested in how different teacher instructional practices during the enactment of the same instructional unit influence Contract grant sponsor: National Science Foundation Contract grant number: ESI 0101780, ESI 0227557. Correspondence to: K.L. McNeill E-mail: kmcneill@bc.edu DOI 10.1002/tea.20201 Published online 31 July 2007 in Wiley InterScience (www.interscience.wiley.com). �� 2007 Wiley Periodicals, Inc.

students��� ability to engage in one important scientific inquiry practice, the construction of scientific explanations. Role of Teachers in Inquiry It is not enough to acknowledge that teachers play a critical role. We need to know what their role is in order to help support them in the difficult task of creating an inquiry-oriented classroom. Teachers have difficulty helping students with scientific inquiry practices, such as asking thoughtful questions, designing experiments, and drawing conclusions from data (Marx, Blumenfeld, Krajcik, & Soloway, 1997). Many science teachers may not have the appropriate expertise to create an inquiry-based learning environment (Krajcik, Mamlok, & Hug, 2001). Teachers need to learn new ways of teaching to promote scientific inquiry, which may differ from their own earlier socialization into school science as students (Lee, 2004 Metz, 2000). Although teachers often have difficulty supporting students, there is little research that provides guidance on what types of teacher practices may help students with scientific inquiry. Research literature about inquiry classrooms often does not describe the classroom practices, rather classroom inquiry is summarized as ������doing science,������ ������hands-on science,������ or ������real-world science������ (Crawford, 2000). Furthermore, researchers often label a classroom as inquiry-oriented based on the nature of the curriculum materials used by the teacher and not by what the teacher and students are actually doing (Flick, 1995). Because teachers��� beliefs about the nature of science, student learning, and the role of the teacher substantially affect their enactment of inquiry curriculum (Keys & Bryan, 2001), this raises the question of how using inquiry materials actually translates into inquiry-oriented classrooms. There is probably a range of inquiry occurring in these research studies labeled as exploring inquiry-oriented classrooms. Like other researchers (Flick, 2000 Keys & Bryan, 2001), we argue that there are few research studies that actually examine teachers��� instructional practices in inquiry classrooms. Scientific Explanations One prominent scientific inquiry practice in both the standards documents (AAAS, 1993 NRC, 1996) and recent research literature in science education is the construction of scientific explanations or arguments (e.g., Bell & Linn, 2000 Driver, Newton, & Osborne, 2000 Jimenez-Aleixandre, �� Rodr��guez, �� & Duschl, 2000 Kelly & Takao, 2002 Sandoval, 2003 Zohar & Nemet, 2002). Explanations refer to how or why a phenomenon occurs (Chin & Brown, 2000). An argument is an assertion with a justification (Kuhn, 1991) or a standpoint that is justified or defended for a particular audience (Van Eemeren et al., 1996). In our work, we use the word ������explanation������ to align with the national and state science standards that our teachers need to address, but our work builds on literature for both explanation and argumentation. Our goal is to help students construct scientific explanations about phenomena where they justify their claims using appropriate evidence and scientific principles. Engaging students in scientific explanation and argumentation is a fundamental aspect of scientific inquiry (Duschl & Osborne, 2002). A key goal for science education is to help students seek evidence and reasons for the ideas or knowledge claims that we draw in science (Driver et al., 2000). Helping students engage in this practice may help shift their view of science away from science as a static set of facts to science as a social process where knowledge is constructed. Bell and Linn (2000) found that there is a correlation between students��� views about science and the arguments that they construct. They suggested that engaging students in this practice may help refine their image of science. Furthermore, engaging in scientific explanation may help students construct a deeper understanding of the content knowledge. For example, Zohar and Nemet (2002) 54 McNEILL AND KRAJCIK Journal of Research in Science Teaching. DOI 10.1002/tea

found that students who were engaged in a unit on argumentation skills through dilemmas in human genetics learned greater biological content knowledge than a comparison group who learned genetics in a more traditional manner. Although engaging in scientific explanation is an important learning goal for students, students often have difficulty articulating and defending their knowledge claims (Sadler, 2004). Kuhn (1991) investigated both children and adults��� ability to construct arguments and found that this practice often did not come naturally to them. They often had difficulty coordinating their claims and evidence. Even in a classroom setting where scientific explanation is an explicit goal, students still have many difficulties. Students can have difficulty using appropriate evidence (Sandoval, 2003) and providing sufficient evidence for their claims (Sandoval & Millwood, 2005). Students also have difficulty justifying why they chose their evidence to support their claims (Bell & Linn, 2000). In our previous work, we found that students had the most difficulty using scientific principles to justify why their evidence supports their claim (McNeill et al., 2006). To help middle school students and teachers with this difficult scientific inquiry practice, we developed an instructional model for scientific explanation by adapting Toulmin���s (1958) model of argumentation. The scientific explanation framework includes three components: a claim (a conclusion about a problem) evidence (data that supports the claim) and reasoning (a justification, built from scientific principles, for why the evidence supports the claim). In other work, we discussed the development of our framework as an instructional model (McNeill, Lizotte, Krajcik, & Marx, 2006 Moje et al., 2004) and as an assessment tool (McNeill & Krajcik, 2007). In this study, we explore how teachers���different uses of the explanation framework in their classrooms influenced student learning. Teacher Instructional Practices Supporting Scientific Explanation Few research studies have explored teacher instructional practices and their influence on students��� construction of scientific explanation or argument. Previous research on students��� construction of explanations in science has focused on scaffolds provided in the student materials or software programs (e.g., Bell & Linn, 2000 Lee & Songer, 2004 Sandoval, 2003 Zembal- Saul, Munford, Crawford, Friedrichsen, & Land, 2002) or on students��� discussions in order to characterize their explanations (Jimenez-Aleixandre �� et al., 2000 Meyer & Woodruff, 1997). Tabak (2004) looked at the role of the teacher in helping students construct evidence-based explanations. She argued that the teacher plays an important role in distributed scaffolding where many aspects of the learning environment, including software and other tools, come together synergistically to support student learning. Osborne, Erduran, & Simon, (2004) recently began exploring pedagogical practices that support students in argumentation. They argued that argumentation does not come naturally to students and that pedagogical practices are important for enhancing the quality of students��� arguments. One of their initial findings is that teacher differences in their emphasis on components of argument may be a result of their different understandings of what counts as an argument (Erduran, Simon, & Osborne, 2004). To further understand the role of teachers in supporting scientific explanation, we examined the literature for instructional practices that may support student learning of scientific explanation, but also other scientific inquiry practices, such as asking questions and designing experiments. From this literature, as well as a preliminary study we conducted on teacher practices (Lizotte, McNeill, & Krajcik, 2004), we decided to examine how teachers used four instructional practices during their introduction of scientific explanation: defining scientific explanation, making the rationale of scientific explanation explicit, modeling scientific explanation, and connecting TEACHER PRACTICES FOR SCIENTIFIC EXPLANATION 55 Journal of Research in Science Teaching. DOI 10.1002/tea

scientific explanation to everyday explanation. We describe each of these instructional practices and provide examples of how they may support students��� successful engagement in scientific explanations. Defining Scientific Explanation What is meant by various inquiry practices, such as designing experiments, asking questions, or constructing explanations, is not necessarily understood by students. One instructional practice a teacher may use to help students with these inquiry practices is to explicitly make the definition of these practices clear to students. Making scientific thinking strategies explicit to students can help facilitate their understanding and use of the strategies (Herrenkohl, Palincsar, DeWater, & Kawasaki, 1999). For example, Metz (2000) found that being explicit about scientific inquiry practices was important for helping children with the inquiry practice of formulating and refining questions. Explicit instruction may benefit diverse learners who are more likely to be unfamiliar with the participation rules and practices that are an essential part of scientific inquiry (Fradd & Lee, 1999). Consequently, this type of explicitness may allow students with impoverished experiences in science education to more effectively participate in classroom instruction as well as be beneficial to all students. In terms of scientific explanation, students may create stronger explanations if teachers explicitly define what is meant by a scientific explanation and define the three components, claim, evidence, and reasoning. In a preliminary study (Lizotte et al., 2004), we found that when teachers explicitly defined scientific explanation, particularly the reasoning component, their students constructed stronger explanations. Making the Rationale of Scientific Explanation Explicit Instruction should both facilitate students��� ability to perform inquiry practices and their understanding of the logic behind the practice (Kuhn, Black, Keselman, & Kaplan, 2000). Helping students understand the rationale behind why a particular scientific inquiry practice is important in science may result in students being better able to complete a performance. Chen and Klahr (1999) found that providing students with the rationale behind controlling variables in science experiments resulted in greater learning of this inquiry practice relative to students who did not receive the explicit instruction. Discussing why it is important to control variables to conduct a ������fair������ experiment helped students when they had to conduct their own experiments. For scientific explanations, it may help students to construct stronger explanations if they understand why an individual may want to construct a scientific explanation and why providing evidence and reasoning results in a stronger, more convincing explanation. Students may need help understanding why someone would argue for a claim. Furthermore, it might be unclear why providing evidence and reasoning provides greater support than just providing an opinion. Modeling Scientific Explanation Modeling various inquiry practices is another instructional practice teachers can use to support student inquiry. Crawford (2000) argued that one of the key characteristics of a teacher establishing an inquiry-based learning environment is modeling the behaviors of a scientist. For example, the teacher Crawford researched in her case study frequently modeled how to grapple with data���specifically, through the extensive questioning of both the methods and results of data collection. Tabak and Reiser (1997) also found that student learning through collaboration in inquiry settings is more effectivewhen teachers model strategies. For example, a teacher modeling 56 McNEILL AND KRAJCIK Journal of Research in Science Teaching. DOI 10.1002/tea

how to reason from biological data can help students complete this same process of analyzing data on their own (Tabak, 2004). Modeling how to include evidence and reasons for claims can help students in their own practice (Crawford, Kelly, & Brown, 2000). This can also help students learn how to use the general scientific explanation framework in a domain-specific context. Teachers can model explanations either through writing or speaking to provide students with concrete examples. Providing students with examples of strong and weak arguments can help them develop an understanding of what counts as a good argument (Osborne et al., 2004). Connecting Scientific Explanation to Everyday Explanation Connecting scientific discourse and inquiry practices to students��� everyday discourse can help support students��� learning of scientific inquiry. Lee and Fradd (1998) proposed ������the notion of instructional congruence to indicate the process of mediating the nature of academic content with students��� language and cultural experiences to make such content (e.g., science) accessible, meaningful, and relevant for diverse students������ (p. 12). Moje, Collazo, Carrillo, and Marx, (2001) built on this concept of instructional congruence. The way students use scientific discourse is shaped by the everyday discourses that they bring to the classroom. To help students develop scientific discourse, teachers need to develop students��� awareness of different discourses and make connections between students��� everyday discourse and science discourse (Moje et al., 2001). Focusing on science as a discoursewith distinct language forms and ways of knowing, such as building theories, analyzing data, and communicating their findings, can help language-minority students learn to think and talk scientifically (Rosebery, Warren, & Conant, 1992). Students need to understand how constructing an explanation in science or supporting a claim in science looks different than in everyday life. Teachers also need to draw from students��� everyday discourse (Moje et al., 2001) and make connections about the similarities between scientific discourse and everyday discourse. For example, a teacher may want to discuss how ������using evidence������ or ������constructing an explanation������ is similar and different in science compared with students��� everyday lives. Method Instructional Context This study occurred during a middle school chemistry unit, How can I make new stuff from old stuff? (Stuff) (McNeill et al., 2004), which we developed using a learning-goals-driven design model (Reiser, Krajcik, Moje, & Marx, 2003). The unit is contextualized in two everyday substances, soap and lard, with the students ultimately investigating how to make soap from lard. During the instructional sequence, students experience other phenomena as well, but they cycle back to soap and lard as they delve deeper into the different content learning goals. The learning- goals-driven design model emphasizes the alignment of the materials with national standards (AAAS, 1993 NRC, 1996). During the 8-week chemistry unit, students learn about substances and properties, chemical reactions, and conservation of mass, both at the phenomena level and the particulate level. Besides content learning goals, the unit also focuses on scientific inquiry practices. During the unit, students design investigations, conduct investigations, analyze data, create models, and construct scientific explanations. Frequently, the construction of scientific explanations is the culminating event in a lesson and supports the meaning-making by TEACHER PRACTICES FOR SCIENTIFIC EXPLANATION 57 Journal of Research in Science Teaching. DOI 10.1002/tea