Middle school students and teachers making sense of the modeling practice in their classrooms

Middle school students and teachers making sense of the modeling practice in their classrooms Andrés Acher & Brian Reiser Northwestern University Paper Present at the 83 rd NARST Annual International Conference Research into Practice: Practice Informing Research Philadelphia Downtown Marriott Philadelphia, PA, USA March 21 24, 2010 Andres Acher, Learning Sciences School of Education and Social Policy, Northwestern University 2120 Campus Drive, Evanston, IL 60208-2610 a-acher@northwestern.edu This research is funded by an instructional materials development grant #ESI- 0628199 from the National Science Foundation. Any opinions, findings, and conclusions or recommendations expressed here are those of the authors. 1

Abstract The aim of this research is to investigate how teachers and students construct their own meaning of the modeling practices in their classrooms. We plan to do that by looking at both emerging tasks and criteria to perform these tasks teachers and students negotiate, and how they contribute to these negotiations. Our data come from three 6 th grade classrooms enacting a project-based chemistry unit (How can I smell things from the distance) that involves students in creating and revising models to account for different molecular phenomena. We videotaped a total of 39 modeling lessons, and start analyzing both how the teacher frames the activity and how students and teacher actively joint to negotiate this frame. Preliminary findings show students-teachers interactions in which modeling tasks like comparing models and judging models are reinterpreted by a chain of tasks that include clarifying the models and changing these models. We also found interactions that show that criteria guiding the usage of models to explain a phenomenon are also reinterpreted by including mechanistic explanations, and refined by extended the explanatory power of a model to different phenomena. We expect that by identifying patterns of interactions reflecting these reinterpretations of tasks and criteria we will offer clues to teachers and curriculum designers to better support students in performing a more sophisticated modeling practice. Introduction Engaging teachers and students in inquiring scientific practices such as modeling continue to be a priority for school reform movements in the USA (Duschl, Schweingruber & Shouse, 2007) and in Europe (e.g. Henzel, van Driel & Verloop, 2007). The science education community has been working from different perspectives to make this happen in science classrooms. For example, researchers have been working on identifying aspects of modeling that could be understood by students, organizing coherent frameworks that could inform teachers and curriculum designers about possible learning goals to guide student s involvement in sophisticated understandings of modeling (e.g. Schwarz et al 2009). Others have designed and 2

tested pre-service teaching programs aiming to teach future teachers how to bring modeling to their classrooms (e.g. Windschitl & Thompson, 2006). Professionals responsible for writing standards work for developing assessment items that could support this priority (e.g. AAAS, Project 2061, 2008), and official reports continue communicating the need of including scientific practices in the classroom in always more clear and persuasive ways to convince policy makers with renovated arguments and evidence (e.g. Duschl et al 2007). We argue that all these investments, while offer an invaluable support, will be greater benefited by gaining a better understanding of how teachers and students make sense to the modeling practice in their classrooms. New practices continue to be unfamiliar for both teachers and students, and their attempts to make sense to new ways of knowing, doing and reflecting entail challenges arising from the influence of the traditional culture of school science (Berland & Reiser, 2009; Jimenenez-Aleixandre, Rodriguez Bugallo, & Duschl, 2000). Both teachers and students need to move from previous practices to the new practice and, although their responsibilities and some of these movements may be different, many of the challenges they encounter may be the same. Communicating each other how they understand their movements when aspects of a new practice are introduced in the classroom can be critical for them to identify the new challenges. For example, teachers and students both may need to interact to check how each other is understanding the benefits of portraying new tasks; they both may need to agree on the kind of questions they are engaging while sustaining new aspects of the practice; they may also need to check each other s understandings about aspects of the practice or about the new roles they are developing during the new activities. For us, science educators and researchers, to study how teachers and students generate these mutual understandings in their classrooms opens the possibility to identify ways to articulate aspects of the new practice that make more sense to them when designed modeling activities are introduced in the classroom. In our study, we recognize that teachers and students, in order to make sense to a new practice such as modeling, construct mutual understandings upon a basic dual need: negotiating understandings of the kind of task required by modeling, and negotiating understandings of the kind criteria the new tasks require for being accomplished. While the first need refers to 3

negotiations associated with the kinds of activity teachers and students do as they engage in elements of the modeling practice such as practicing to judging models, describing models to others or clarifying models to gain better understanding of the used representations, the second one refers to sharing understandings about the more or less explicit ideas about modelling that guide their activity such as negotiating the clarity of the nature of a model component, the relations between model components to explain a phenomenon or the power of the model to explaining different phenomena. Both factors are sides of the same coin, and contributions teachers and students make to connect them would allow us to see how they are making sense of modeling in the classroom. For instance, evaluating models is a specific modeling task. Sharing understandings about this task requires negotiating understandings both about differences and similarities with other tasks such as comparing or clarifying models, and about the kind of criteria with which to perform model evaluations such as understanding the differences between judging a model through the absence or presence of model components or judging models through relations among these components that indicate a mechanism represented in the model. The aim of this research is to examine middle school classrooms developing modeling curriculum activities to investigate how teachers and students construct their own meaning of the modeling practice. We plan to do that by looking at both tasks and criteria to perform these tasks teachers and students negotiate, and how do they contribute to these negotiations. We expect that by identifying patterns of interactions reflecting these negotiations we will provide evidence to understand where the emerging understandings of modeling come from in the classroom, offering clues to teachers and curriculum designers to better support students in performing a more sophisticated modeling practice. We conduct our study upon the following research question: How do students and teachers make sense of modeling in the classroom? Context, Design and Procedures We analyze videos of middle school classrooms developing modeling science curriculum activities designed to support learners in scientific practices, including scientific modeling. The activities are part of a curriculum material called IQWST (Investigating and Questioning our World Through Science and Technology- Shwartz et al, 2008), a project aimed at developing a 4

Middle school students and teachers making sense of the modeling practice in their classrooms three year coordinated series of middle school science materials that engage learners in projectbased investigations that address a focused set of national science content standards ( Krajcik, McNeill & Reiser, 2008). We focused on the IQWST 6th grade chemistry unit, in which scientific modeling is the central practice for teaching the atomic-molecular theory of matter. This six-week unit, organized around a driving question (How can I smell things from the distance), provides a context to motivate and apply the science students learn, involving students in creating models to explain phenomena and then revising models to account for these phenomena. In order to construct a model of matter that will account for all presented phenomena, the students gradually move from a continuous or mixed view of matter to a particulate one. They start by constructing intuitive models of how air looks like when an smell from an open jar with paper mint oil travels through a room (see figure 1 below). Figure 1: initial students intutive models for how an smell travels through their classrom. In the analyzed curriculum unit, students develop their understanding of matter in conjunction with their understanding of models through a designed series of modeling activities and experiments (see figure 2 fro examples). Adding and removing air to a flask. 5

Middle school students and teachers making sense of the modeling practice in their classrooms Air compression and expansion with a syringe. Constructing a consensus model of what air is made of. Figure 2: examples of experiments and modeling activities during the Chemist Curriculum Unit analyzed. Later in the unit, they apply their knowledge in explaining an additional series of phenomena, such as phase changes. The unit draws on activities articulated upon the following sequence: anchoring phenomena to cultivate a question, involving students in constructing, testing, evaluating and revising their models, and using models to predict and explain. A particular focus in this unit is incrementally introducing new phenomena that uncover limitations in students current models. Another focus is on the process of comparing models and designing a class consensus model to best fit the empirical evidence and scientific principles developed in the investigations. We recorded more than 39 videos from three different 6th grade classrooms in a Midwest suburban school with a total of 75 students and one teacher who developed such a kind of curriculum material for her second year. The school was ethnically and linguistically diverse within middle class suburban district. Students represented a range of achievement levels by sampling from high achieving students, medium achieving students, and lower achieving students, as assessed and determined by the school. We transcribed the videos in full length, and uploaded the whole data set to Nvivo software for qualitative analysis. We started to analyze the videos by, first, identifying the general structure that dominates the classroom communication during modeling activities. We identified two main communicative events of this structure: 16

The instructional guidelines framed by the teacher to set the modeling activities. Through this frame, we observed the teacher communicating, sometimes with active students participation, the agenda for the day, and preparing the stage to develop this agenda; 2- The follow up of this instructional guidelines by both teachers and students. In this event, we ve seen teachers and students interacting in different ways to develop the agenda for the day and negotiate the teacher s frame. In a second step, we complete the analytical framework by identifying the tasks and criteria teachers and students emphasize in these two communicative events. Tasks and criteria were identified by an emergent coding process from our data that is still ongoing. Coding for tasks are making upon the following question: What job to be done is communicated?/ What job is being done? Coding for criteria are making upon the following question: What ideas about modeling is guiding the job? Upon which criteria the job is being done? A sample of the code scheme developed so far is shown in the table below. We precede the analysis by systematically comparing teacher s initial frames with teachers-students and students-students interactions that follow these frames to discover how tasks and criteria are continued, ignored or refined. We also start analysing teacher s and students moves that make classroom participants focus on certain tasks and criteria more than on others in order to complete the picture of classroom interactions supporting negotiations of the meaning students and teachers are giving the modeling practice. Coding scheme for tasks and criteria (sample) Tasks What kinds of activity teachers and students do as they engage in elements of the modeling practice Describing a model: teachers or students communicate their models without aiming to perform any other task such as evaluating, clarifying or extending the model to other phenomena. Clarifying models: teacher or students communicate failure of understanding. Their interest is neutral in the sense that there is not a problem with the model perceived or communicated. Changing a model: teacher or students get involved in changing an existed model. Judging a model: teachers or students communicate a perceived problem with their own model or with the understanding of others model. It could be by formulating a problem or proposing a solution to a problem. Comparing models: teachers or students compare more than one model. It could be as a central task or as a complementary task to perform. 7

Criteria What ideas about modeling guide teacher and students activity? Social criterion: the importance of what others say is emphasized as a criterion to perform a task. Model - Phenomenon criterion: a task is performed by correlating a model with a phenomenon showing that a model is not assessed in isolation. Mechanism criterion: a task is performed throughout focusing on the internal organization of the model. Relationships among model components are emphasized to perform the task. Process criterion: a task is performed by stressing temporal dimensions of a phenomenon. Model component criterion: a task is performed by focusing on model components without establishing any kind of relationships among them. It could be, for instance, by stressing the absence or presence of a model component. Preliminary findings We report here what emerges from the first few videos examined so far by giving examples that illustrate our preliminary findings. Our analytical procedure start to reveal what modeling tasks get reinterpreted upon original frames set by the teacher. For instance, judging models upon comparing them are two of the main tasks initially demanded by the teacher. Those seem to require the same pattern of re-interpretation: teachers and students engage first in describing and clarifying models to then change these models. While these emerging tasks are always prompted by the teacher, students seem to always make it work when they judge and compare their own models. The following excerpt, where students present on the whiteboard their first models of odor travelling through their classroom (see figure 1), illustrates this kind of interactions. In the example, a student shows her model to the rest of the classroom by compelling two tasks: judging negatively her model ( I disagree with myself and my model ), and comparing it with other students models ( So then when I saw Tina's and Kiana's [models], I saw that they have air ). When compelling these two tasks, changes are encouraged by the teacher ( And where would you put air in now if you could? How would you do that? ), and followed it by the student ( Should I draw it? I'd just draw it around. There's like air everywhere. ). The student complete that by giving a description of her model, and both students and teachers engage in 8

clarifying aspects of this model description ( Student: There's like air everywhere.; Teacher: So what's in-between a squiggly and a dot?; Student: Nothing.) Teacher: Okay, tell me what you're doing. Student: Well, first of all, I disagree with myself and my model. Teacher: Oh, no. What happened? Student: Mine [model] doesn't have air so I discovered that the molecules won't be able to move around and get to your nose. So then when I saw Tina's and Kiana's [models], I saw that they have air. So then I remembered that the molecules need air to travel around to get to your nose. Teacher: And where would you put air in now if you could? How would you do that? Student: I would put it around, like, the white space. Teacher: Show me what you would do. Student: Should I draw it? Teacher: Yeah. Student: I'd just draw it around. There's like air everywhere. Teacher: So what's in-between a squiggly and a dot? Student: Nothing. Teacher: And what does that mean? Student: I haven't thought of that. Figure 1: students drawing their first models of smell traveling through the room Another aspect that seems to emerge as a pattern during classroom interactions is the tendency of maintaining criteria framed in teacher s initial instructional guidelines to perform the task of using models to explain a phenomenon. We found students and teachers develop two criteria: they engage in clarifying the nature of model components as a necessary step to include these components in more mechanistic explanations. For instance, it seems that they need to construct common understanding of what is the white space between the molecules a model component (see figure 3, and excerpt bellow: Teacher: What was the space that wasn't the tiny particles? Student: Nothing; Teacher: Is nothing matter..?; Students: nothing is not matter ), to then relate this component with others such as molecules to build more mechanistic explanations (e.g. Student: when they re compressed [the molecules], they go together and then there s like less of that space, but when they expand, they spread apart and there s more of the nothing ). The students and teacher engagement on these criteria usually comes as a continuation of the teacher initial instructional guidelines for the day where she implicitly communicates these criteria (e.g. Teacher: Why is nothing there [in between the molecules] help me to realize how particles compress? or What was the space that wasn't the tiny particles?). This kind of continuations, although reached quite implicitly among teachers and 9

students, gives us insights about how the development of criteria looks like in the classroom when a teacher work with her students to improve the understanding of the explanatory power of models. We have also seen students re-interpreting original criteria demanded by the teacher by bringing new criteria to perform the task of using models to explain phenomena. In particular, we ve found students going beyond teacher s demands by generalizing their model over similar phenomena they have experienced during their meddling classroom activities. For instance, as shown in the excerpt below, although the teacher hadn t explicitly asked for it, the student communicates that her model can be used to give a mechanistic explanation of how air compress ( when they re compressed [the molecules], they go together and then there s like less of that space, but when they expand, they spread apart and there s more of the nothing ) but also to explain air compression, expansion, adding air or removing air [to a flask or a syringe -as they d been experienced in previous activities]. Teacher: Yesterday you put some of your ideas on the board [the consensus model shown in figure 4 on the side] What was the space in the middle that was not the tiny particles? What is this white space? Is nothing matter?; Why is nothing there [pointing to the space between the molecules on the consensus model] help me to realize how particles compresses? Student 2: when they re compressed [the molecules], they go together and then there s like less of that space, but when they expand, they spread apart and there s more of the nothing so you can show compressing, expanding, removing and adding Figure 4. Students consensus model of what air if made of. As a final remark, although our analysis is too preliminary at this moment to report any certain finding, we want to mention something that starts emerging as a significant contribution from teachers and students sustaining their negotiations about modeling. Our videos show that there are some students and teachers moves that make them focusing more in certain tasks and criteria than others. These moves are, for instance, the teacher sending students to the board at different moments or students continuously focusing on one particular model than on others. We 10

want to include these observations in our analysis to support the answer to our research questions. Discussion and Conclusion Base on the preliminary findings illustrated above, we believe that we are collecting strong evidence to picturing students and teachers interactions that sustain sophistications of the modeling practice from classroom participants perspectives. We ve seen students, sustained by their teacher, comparing models to judge these models; we ve seen them bringing their ideas with their models but clarifying these ideas for consensus; and we ve seen students getting familiar with the flexibility of their models to change them for further improvement. The detailed re-interpretation of tasks and criteria t we started to examine will contribute to better design modeling curriculum activities that keep the potential to support transitions in developing a more sophisticated understanding and doing of the modeling practice. Specifically, those activities offering alternatives to extend the criteria linked to improving the explanatory power of models or those generating criteria to generalize the explanatory power of models to a wide range of phenomena, and the re-interpretation of particular tasks such as model evaluations, will be of the substantial value to design more productive modeling activities sustaining growth to level 3 of understanding of our proposed learning progressions for modeling practices (Swartz et. Al 2009; Swartz, et al in press). Our preliminary findings also raise some challenges, mainly associated with the implicit character of the reinterpretation of tasks and criteria during the development of modeling classroom activities. For example, we haven t seen the teacher making of the criteria for accomplishing different tasks an explicit learning goal. We know that the practice of modeling is new for teachers and students, and we want to understand what are the most effective ways to smoothly introduce this new practice in the classroom, therefore, questions emerging from this preliminary findings would focus on understanding the extent to which making these criteria explicit help students to improve their articulation when they perform different modeling tasks. By completing the analysis shown in this paper, we intend to achieve a more comprehensive picture of modeling tasks and criteria that emerge from the classroom participants perspective and working for improving further research questions like this one. 11

References American Association for the Advancement of Science. (2008). Benchmarks for science literacy. New York: Oxford University Press. AAAS, Project 2061. Berland, L. K., & Reiser, B. J. (2009). Making Sense of Argumentation and Explanation. Science Education, 93(1), 26-55. Duschl, R. A., Schweingruber, H. A., & Shouse, A. W. (Eds.). (2007). Taking science to school: Learning and teaching science in grades K-8. Washington, DC: National Academies Press. Fortus, D., Krajcik, J., Dershimer, R. C., Marx, R. W., & Mamlok-Naaman, R. (2005). Design- Based Science and Real-World Problem-Solving. International Journal of Science Education, 27(7), 855-879. Henze, I., Van Driel, J., & Verloop, N. (2007). The change of science teachers' personal knowledge about teaching models and modelling in the context of science education reform. International Journal of Science Education, 29(15), 1819-1846. Jimenenez-Aleixandre, M. P., Rodriguez, A. B., & Duschl, R. A. (2000). Doing the lesson or doing science : Argument in high school genetics. Science Education, 84, 757 792. Krajcik, J., McNeill, K. L., & Reiser, B. J. (2008). Learning-goals-driven design model: Developing curriculum materials that align with national standards and incorporate projectbased pedagogy. Science Education, 92(1), 1-32. Shwartz, Y; Weizman, A.; Fortus, D. ; Krajcik, J & Reiser, B. (2008). The IQWST Experience: Using Coherence as a Design Principle for a Middle School Science Curriculum. The Elementary School Journal, Volume 109, Number 2. Schwarz, C. V., Reiser, B. J., Davis, E. A., Kenyon, L., Acher, A., Fortus, D., et al. (2009). Developing a learning progression for scientific modeling: Making scientific modeling accessible and meaningful for learners. Journal of Research in Science Teaching, 46(6), 632-654. 12

Schwarz, C., Reiser, B., Fortus, D., Kenyon, L., & Acher, A. (in press). MoDeLS: Issues and challenges in defining a learning progression for scientific modeling. In A. Alonzo & A. Gotwals (Eds.) Learning Progressions in Science (LeaPS) Windschitl, M., & Thompson, J. (2006). Transcending simple forms of school science investigation: The impact of pre-service instruction on teachers understandings of modelbased inquiry. American Educational Research Journal, 43(4), 783 835. 13