[ILLUSTRATION OMITTED] Across the nation, teachers gather regularly to examine student work and uncover how their students are thinking about science. These teacher groups aim to improve both their instructional methods and the effect these methods have on student learning. Achieving positive outcomes, however, requires teachers to use thoughtful, intentional methods to design instruction, select student work, analyze these artifacts, and make evidence-based changes to instruction (Bray et al. 2000; NRC 1996; Wood 2007). This article presents a model of collaborative inquiry for groups of science teachers who want to systematically improve their practice through analyses of student work. The five-phase [APEX.sup.ST] (Advancing High-Leverage Practices by Examining Student Thinking) model (Figure 1) is appropriate for students of all achievement levels. It focuses on high-leverage practices (e.g., pressing for evidence-based explanations) and longitudinal learning for both students and teachers over the course of a year. Critical Friends Groups We have designed and tested the [APEX.sup.ST] model of collaborative inquiry for Critical Friends Groups (CFGs). A CFG is a learning community of 8-12 educators who gather for about two hours a month to discuss improving their practice through collaborative learning. In collegial CFG meetings, teachers engage in a cycle of inquiry, reflection, and action to promote adult growth that is directly linked to student learning (Curry 2008; NSRF 2009). The [APEX.sup.ST] model provides a structure and focus for these groups to engage in meaningful examination of student work. There are five phases of the [APEX.sup.ST] model that support the improvement of teaching and student learning. These are described in the following sections. [FIGURE 1 OMITTED] Phase 1: Defining a vision of worthwhile learning Perhaps the most important decision CFGs can make is to identify some aspect of student learning that is important enough to focus on for a full academic year (Ball et al. 2009; Curry 2008; Windschitl, Thompson and Braaten Forthcoming). We recommend teacher groups choose a core scientific practice they can develop across different topics and courses. In our most recent project, we chose students’ “construction of evidence-based explanations.” This type of scientific thinking is critical to understanding the more conceptual ideas in science and is a valued scientific practice (NRC 2000; Windschitl 2008). It is important that CFGs do not choose topics that can change from meeting to meeting–this will make it difficult to analyze any one scientific practice’s development over time. Creating lessons that press students for evidence-based explanations can be difficult work. The challenge is that lessons from standard curriculum materials are often organized around narrow topics or processes, not big ideas with underlying explanations. Big ideas have complex causal stories, composed of a web of events and concepts that help explain why observable phenomena occur. An example of one observable phenomenon is the diffusion of materials across a membrane. The causal explanation for why this occurs has to do with core concepts such as equilibrium, concentrations of solutions, and permeability of membranes. Once a big idea is selected, teachers can work together to identify a lesson or series of lessons that aim to explain observable phenomena. Lessons that ask students to gather information through investigation or already existing evidence help them understand both the development of an explanation and the use of evidence. One such example would be a pulley investigation in which students use ideas about forces, work, and energy to explain why a single person can lift a very heavy load using simple machines.