MIDDLE GRADES TEACHERS’ CHARACTERIZATIONS OF INTEGRATED MATHEMATICS AND SCIENCE INSTRUCTION

It would seem evident that school curriculum, school organization, and classroom instruction should form a coherent enterprise where each work together to produce the best learning environments for students. Educational reforms should consider these enterprises as a whole, assuming that changing one variable will have an impact on the others. Therefore educational reformers interested in implementing learning improvements by introducing one particular initiative have two options: monitor other variables and modify them as needed to enhance the overall reform effort; or, leave them unchanged but prepare for unintended consequences due to ignored synergies between each variable.

Few educational research studies try to understand how a change in one aspect of a reform is shaped by and shapes the overall learning environment regarding that reform effort. This is particularly true when the reform is nonspecific in terms of implementation. This is the case with standards-based curriculum or changing the organizational structure of middle schools. What this study seeks to understand is the cumulative impact of the organization of middle schools and the introduction of standards-based curriculum on integrated mathematics and science instruction.

In this article, we briefly discuss the rise of interest in integrated mathematics and science instruction in the nineties and the research describing models and methods of integrated instruction. From this we consider the movement towards standards-based curriculum and accountability tests, and how national standards construct connections between mathematics and science. Finally, we provide overviews of the “middle school” and preferred middle school organizations that attempted to create a whole school experience for middle school students. We have chosen to focus on the middle school grades because at these grades teachers tend to specialize by subject area but they do not teach in the departmentalized structure of many high schools. Once we have laid out this terrain, we discuss our research study based on the following questions:

1. How do middle grades teachers identify and characterize integrated mathematics and science teaching?

2. How do schools’ contexts affect teachers’ characterizations and implementation practices?

Based on the findings we provide insights into how middle grades teachers in our study characterize integrated mathematics and science instruction. We also discuss how middle grades teachers envision an integrated curriculum fitting into their teaching contexts and how their view of integrated instruction is shaped by their school’s organization and the current state of standards-based instruction.

An integrated curriculum in mathematics and science more accurately represents the historical and synergistic development of both knowledge domains. Further, the development and implementation of integrated mathematics and science curricula was a major strand of multiple national reform initiatives in the late eighties and early nineties: Project 2061: Science for All Americans (American Association for the Advancement of Science [AAAS], 1990), Everybody Counts: A Report to the Nation on the Future of Mathematics Education, (National Research Council [NRC], 1989), Curriculum and Evaluation Standards for School Mathematics (National Council of Teachers of Mathematics [NCTM], 1989), National Science Education Standards (NRC, 1996), and Principles and Standards for School Mathematics (NCTM, 2000). As stated specifically in the Benchmarks for Science Literacy (AAAS, 1993), “The basic point is that the ideas and practice of science, mathematics and technology are so closely intertwined that we do not see how education in any one of them can be undertaken well in isolation from the others” (pp. 321-322). This reform document and the others sought to develop a pathway for rigorous courses of study for all students, so that they could develop proficiency in school subjects and use this knowledge to function in a world that does not divide experiences into simple knowledge domains.

However, despite these initiatives integrated instruction in science and mathematics remains elusive (Sanders, 2009). Several authors have noted the lack of empirical research on mathematics and science integration (Berlin & White, 1992; Czerniak, Weber, Sandmann, & Ahern, 1999; Davison, Miller, & Metheny, 1996). What the available research on integrated mathematics and science instruction does tell us is that there is a great deal of uncertainty about what educators mean by integration (Czerniak, Weber, Sandmann, & Ahern, 1999; Davison, Miller, & Metheny, 1995; Roebuck & Warden, 1998). The lack of a definition of integration (including the alternate name of interdisciplinary instruction) is one reason for a minimal or confounded research base. In too many instances the definition of integration is idiosyncratic to a particular program or study. Czerniak et al. (1999) reviewed the models of mathematics and science integration commonly found in education literature. Based on this review and the work of Davison, Miller, and Metheny (1995) and Roebuck and Warden (1998), we found seven possible models of integrated mathematics and science instruction commonly described. We want to highlight these as integration because they require someone to intentionally or knowingly make connections. The models include:

1. Process integration involves using a mathematical process to complete science calculations. For instance, a chemistry teacher will review algebraic rules when students are solving limited reagent problems.

2. Pedagogical integration entails using authentic science context to practice math skills. For example a teacher uses the population growth of bacteria as an example of exponential growth.

3. Thematic integration (units) requires all involved subject areas to either teach the content of the theme or to use ideas in the theme as a pedagogical context. This typically looks like studying Antarctica as a theme; science ideas such as habitats, geography, and animal populations are involved while mathematics skills are embedded in activities.

4. Discipline specific integration concerns the linkages within a single discipline. For example, studying the relationships between geology and chemistry found in a field such as geochemistry.

5. Concept specific integration occurs when a particular concept or skill appears in both disciplines. These might include concepts such as metric measurement, graphing, and scientific notation.

6. Project-based integration asks students to work on complex problems that require them to use knowledge and/or skills from multiple domains, whether or not the teacher has explicit instructional goals in all areas.

7. Synergistic integration necessitates developing knowledge of mathematics and science as related content areas that must be used and developed simultaneously. It may or may not explicitly involve students in focusing on how both content domains are shaped by each other.

Although it is possible to determine different models of integration from the literature, there is not clear understanding of the models teachers draw upon when they report they are engaged in integrated instruction. Nor does the literature on integration discuss which models teachers are more or less likely to try to implement in school settings. We believe that the mathematics and science reform documents were focused on furthering the development of integration based around models 4, 6, and 7. However, the reformers were aware that models 1, 2, and 5 are already part of most science and mathematics curriculum, whether teachers were aware of them or not.

Since the above reform documents were the driving forces behind the standards-based reform that is currently in place in U.S. schools, it can be assumed that the professional organizations that wrote these documents considered the importance of connecting mathematics and science in useful and authentic ways. However, as states have adapted the national standards into state standards with associated accountability tests in mathematics and science, subject matter and instruction appears to be more discrete than before (Hamilton et al., 2007; Rodney, Sandholtz, Martinez-Flores, & Scribner, 2003) The crosscutting themes and ways of thinking, found in the NSES and the NCTM standards respectively, which pushed instruction towards integration are difficult to test, hard to determine whether they fit into the mathematics or science tests, and thus typically left out of instruction.

Overlaid on these policy documents, Beane and Brodhagen (2001), discuss the importance of integrated middle school curricula to address the learning needs of students in this age group. An integrated curriculum was seen to be an important learning tool for middle grades students so they learned to make explicit connections between subject matter and to provide them opportunities for learning transfer, self-directed learning, and exploration of authentic experiences with the subjects. In order to achieve these learning outcomes, Beane and Brodhagen (2001) outlined a new model for the organization for middle schools. They suggested a team-, house-, or familybased model for organizing students and teachers. Teams of students would share a set of core subject area teachers who would be content area specialists. In this way an individual student is well known to his or her core subject teachers, teachers come to know about their students in many subject areas, and teachers can plan collaboratively for a group of students. Finally, this middle school structure provides a transition for students from the selfcontained classroom with one teacher for all subjects to the open schedule and multiple teachers of the typical high school model.

One explicit goal of the middle school model was the creation of an integrated learning environment for the students. Researchers have identified a range of structures (Beane, 1996; Loepp, 1999) and strategies (Kolodner et al., 2003) designed to ensure fertile ground for content integration; however, the impact of standards-based reforms and accountability testing on integration has not been thoroughly explored (Au, 2007; Vars, 2001). In the next sections we discuss how our research to understand how the three reform ideas—integration, standard-based reform, and a middle school organization—have impacted on each other.

We followed an interpretive paradigm for this research. Educational reform is complex and many factors intersect to enhance or interrupt the process. We were interested in understanding one particular set of meanings and how they interacted. It is this “meaning in context” (Merriam, 1998, p. 1) that was of most interest to us; therefore, a qualitative approach was necessary. We were not interested in quantifying how often teachers integrated instruction, or measuring the time spent on discrete standards versus cross-cutting standards in mathematics and science, rather we wanted to understand how, in specific instructional contexts, teachers reflect on their meanings of integrated instruction. In addition, we listened for how they described the influences of school organization and curriculum in relationship to their meanings of integrated instruction.

At a first level of understanding, we were interested in describing the range of models teachers used to discuss integrated mathematics and science instruction. We constructed a survey structured to understand the implicit criteria middle grades teachers used to identify integration (Stinson, Harkness, Meyer, & Stallworth, 2009). The survey was modeled on Hewson and Hewson’s (1989) Conceptions of Teaching Biology interview protocol.

In the case of this research we carefully developed and piloted teaching scenarios as examples of mathematics and science integration to suggest, although loosely, six of the seven models described previously (model 4 was not included). Six teaching scenarios were employed (shown in Table 1) which reflected content found in the middle grades science and mathematics state standards. Each teaching scenario was followed by questions asking respondents: (a) if the scenario was integrated or not; (b) to explain why they found the scenario to be an example of integration; and/or (c) suggest ways to alter scenarios they thought were not integrated in an effort to make them more integrated in nature.

As discussed above, the scenarios were designed to loosely suggest a model of integration drawn from the literature. The teachers completing the survey had to interpret the scenario based on their own teaching contexts and experiences to form an opinion and provide a rationale. Although surveys are not typical qualitative research tools, the structure of this survey asked the teachers to make decisions based on their interpretations of integrated mathematics and science instruction.

In addition to the survey, a select group of teachers was interviewed at their schools to understand how their meanings of integration were shaped by the specifics of their practice and the contexts in which they taught. These teachers were identified from the survey; it asked if they would be interested in participating in face-to-face interviews about integrating instruction. A common interview protocol was developed and piloted; however, for each participant interview probes and follow up questions were used to elicit more detailed descriptions. The interview design focused the teachers on describing their schools’ organizations, their practices and resources, and their administrations’ support for integration. This allowed us to hear how different teachers constructed their teaching contexts and the impact these contextual and organizational factors had on their instructional choices.

One hundred twenty surveys were sent to mathematics and/or science teachers teaching in Grades 5 to 8 in local public schools located within a 50 mile radius of the university where the researchers were situated. Schools within that radius but sited in neighboring states were eliminated from consideration. The 33 survey respondents represented a wide range of schools and school districts in terms of size (i.e., small, medium, large), classifications (i.e., urban, suburban, rural), and socioeconomic levels. Six teachers were selected to be interviewed from those that returned the contact information, although only four interviews are reported here. The interviewed teachers were selected based on their schools’ locations and organizations and the teachers’ years of experience. The basic information about the interviewed teachers is shown in Table 2.

The analysis of the surveys was structured to inform us about the specific factors in a teaching and learning scenario which teachers used to determine if an activity was or was not integrated. The analysis proceeded as follows. The explanations of the “yes” responses were grouped together. The explanations of the “no” responses were similarly grouped. Within these two groups of responses key terms such as, math not explicit, or needs more discussion, were determined. These coded terms described the reason or factor that enabled the teacher to determine what made the scenario integrated or what needed to be added in order to make the scenario integrated. We compared the key terms with the models described in the literature to determine which integration models teachers responded to most frequently (Stinson et. al., 2009).

The analysis of the interviews took place in two phases. The first phase involved creating a narrative description of each participant’s teaching context. These descriptions were written to highlight the teachers’ voices and to create a full picture of the teachers’ contexts. The second phase of the analysis focused on instances where teachers mentioned opportunities that allowed for integrated teaching within their teaching contexts as well as descriptions of integration that they provided. Once the opportunities were identified, they were then referenced back to each individual teacher’s context information to understand how the opportunities developed. A similar scheme was followed for lost opportunities or examples of when integration could not occur and the contextual factors that disrupted these opportunities.

Table 3 shows the percentages of teachers who identified each scenario as either yes or no with regards to integration. In some cases a respondent answered both yes and no or in some instances no determination was made. As shown in Table 3 these responses have not been included; therefore there is not a 100% representation for each scenario.

Two teachers indicated that none of the scenarios were integrated. An additional teacher marked all but one as not integrated and the one that was marked as integrated he/she gave both a yes and a no response. Six teachers indicated that all of the scenarios were integrated and three provided no response for at least one scenario.

As we examined the responses to each scenario we searched for patterns in both the integration scenarios identified as examples of integration as well as the rationales respondents provided to justify their designations. In most instances, teachers appeared to create a rule that embodied important criteria for integration and then used it relatively consistently. Most commonly, respondents took a content identifying approach and provided very narrow or discrete framings of the content as exemplified by respondent # B8’s ideas from Scenario 5: “Scale model—math. Solar system —science.” An extension of this approach involved some respondents’ needs for a balance in the amount of mathematics and science learning that could result from the described activities. If a scenario appeared to have equal amounts of mathematics and science it was integrated. For example:

Or in the ‘no’ case below, more science was needed to be added if it was to be integrated.

Some teachers identified prerequisite content or skills that students needed to learn prior to engaging in the learning activity suggested in the scenario. For example:

Teachers concerned that integration required students to make important content connections claimed that directed classroom discussions were imperative.

Or

Two additional strategies were used by the teachers to determine if a scenario was an example of integrated instruction. The first was to determine if the activity was “authentic.” However, when the teachers used “authentic” they were not considering the development of the knowledge as being integrated—as described in models 2, 6 or 7— rather they appeared to be defining the term authentic as “real-life” contexts. For example: “Students get to experience real life situation of how ocean floor depths are found” (Scenario 4; respondent #3). Or more simply put by respondent #1, “real life application” in response to Scenario #6. The final strategy fits well with integration as detailed in model 5, concept specific integration. In these cases a teacher explained how a concept or skill was used in both domains. “Integrates the study of dinosaurs (science) with measurement (science and math) and creating charts (math)” (Scenario 3; respondent #8). This strategy was frequently used in the scenarios that involved, or could be extended to involve, data gathering, charting, and graphing. For example respondent R4 stated “Graphing is a major element in both math and science” in response to Scenario 1. However, some teachers did not consider all of the processes—data gathering, charting and graphing—to occur in both mathematics and science; rather, they divided the activities and placed them in different content domains. In this way, data gathering and charting were typically defined as activities of science, but graphing was a mathematics activity. So a scenario that included data but was not graphed was not seen as having sufficient mathematics. This division is most clearly seen with Scenario 1. “Graphing data requires the use of math” (respondent #8); or “Using graphs (math) with pulse (science). It can tell you pulse rate (science) and how to find slope (math)” (respondent #B5).

The models we delineated based on the research that most frequently resonated with the teachers’ responses were 1-process integration, 2-pedagogical integration, and 5-concept specific integration (as discussed previously). In the discussion of the scenarios in instances where the teachers could most readily identify common content and/or skill work they were most likely to determine that integration was taking place. This was closely followed by process integration as a cue to whether or not a scenario was integrated, although some teachers questioned whether just using an equation to solve a science problem constituted integration. The two quotes below demonstrate this contrast for two teachers in response to Scenario 4, the first said the scenario was an example of integration and the second said it was not because something would have to be added to make it integrated.

And

Respondent B6’s response to Scenario 4 provides a good example of what teachers were looking for when identifying the integration at the pedagogical level. When a pedagogical form of integration was suggested, it typically arose when a teacher stated that more would need to be done during the instruction than was provided in the scenario. It really depended on what the teacher (or students) did beyond the scenario to develop, connect, and extend the learning. So using the response above, just having students use an equation in science class was not sufficient for integration unless it was attached with broader related concepts.

In the design of the scenarios we included or implied key terms from the models of integration typically cited in middle grades research: themes (Scenario 2), units (Scenarios 3 & 5), and projects (Scenarios 4 & 5). However, the teachers did not use these terms to identify or explain a scenario as integrated. There were no instances of a teacher suggesting that if students were doing a project they would be using multiple skill sets and, therefore, the learning would be integrated. This is an interesting note because some of these terms did come up when the teachers were interviewed.

It appears that the models that teachers most readily identified as “integrated” are not the models of integration that the mathematics and science reform documents were seeking. Rather our teachers responded to the traditional views of integration which rely heavily on teacher led instruction rather than students using and applying knowledge from one domain to create and extend knowledge in the other or both. To put it more succinctly, integration is something teachers do to teach content. Teachers did not view integration as inherent in how students construct knowledge or in the interrelated growth of mathematics and science knowledge.

When we turn to integration as discussed by the four interviewed teachers in the context of their own schools a new model arises. The teachers interviewed all completed the survey; however, we intentionally did not track the survey responses to the interviews. We wanted to have no direct comparisons of their written and spoken views. Rather we used the interviews to elaborate and expand the models of integration that teachers identified in their own practice. In this way, we saw two models from the surveys reiterated and elaborated in the interviews and a new model introduced.

Based on the models above, concept specific integration and pedagogical integration were being used by three of the four middle school teachers. Each of the teachers discussed specific ways in which they relied on pedagogical or content specific integration in their teaching. Two examples from Eileen included:

And

Cathy discussed how she and the science teacher would integrate the subjects.

Finally, Becca acknowledged that integrated mathematics and science teaching did take place; however she was uncertain if this was a good thing.

And in terms of Becca integrating her mathematics instruction with science she stated that:

Dennis was sure that he could no longer integrate his science instruction with mathematics because the students lacked the content and skills that had allowed this to happen previously, when physical science was in eighth grade along with algebra.

In addition to this reliance on process and content integration, when the teachers discussed integration at their school sites, two teachers included identifying “projects” and “thematic units.” However, it was unclear how different these models were from process integration or content specific when the teachers explained how these activities related to mathematics and science. At Dennis’s school the teachers were required to use thematic units across all the content areas. This was mandated and coordinated by the school administration.

Cathy talked about integrated projects that she did with her science teaching colleague. In addition to the project above, Cathy described an integrated project her class completed and her commitment to integration.

In both of these cases the substance of the integration, even when in a project or thematic unit, was at the skill level at best.

What did become evident from the interviews with the teachers was that the great majority of their examples of integration were co-teaching events. It seems that within the middle school model of team-based instruction, teaming and integration were seen as linked. This is not a model referred to in the literature, although it is not excluded from the models outlined. However, when integration is seen as a part of team teaching and requiring a second teacher, it places the teachers and instruction at the center of the integration activity, rather than the students and the learning. A further result of this is that this form of integration becomes reliant on the relationships between teachers and their commitments to work together.

The teachers involved in the interviews described how their relationships with other teachers enhanced or inhibited their ability to integrate mathematics and science. Two of the teachers expressed a desire to integrate instruction but questioned whether the other content area teachers had the appropriate content knowledge. The interviewed teachers never questioned their own content knowledge but felt that content knowledge could be a problem for others. The interviewed teachers also believed time to meet and plan collaboratively created a barrier for integration.

In comparing the results from the two data sources, it became clear to us that in the school context, models of integration were characterized by a collaborative or team approach, involving a mathematics and science teacher working together. This was not the model presented to the same selection of teachers in the survey, nor was it part of the literature base that framed the research. The survey examples all implied one teacher, with sufficient content depth in both areas, integrating the subjects as a regular aspect of his or her own teaching. The model that appeared to develop in context suggests the content remains divided between different teachers who have learned either mathematics or science content.

Although there were discussions of standards in the survey responses given by teachers they appeared with less frequency than we had anticipated. In some cases teachers relied on whether or not they could “find” a mathematics and science standard in the scenario content. Or they might suggest that the content in the scenario was not at the correct gradelevel standard for the students they were currently teaching, or as with the case of the first scenario the content was a better fit somewhere else: “But not science. I would say health and math. Pulse rate and graphing” (respondent #12); “Pulse refers to health science. Graphic deals with data analysis benchmark” (respondent #6).

Scenario 3, which dealt with the dinosaur project, was determined by a few teachers to have a content/standard mismatch for the grade level and best exemplified by respondent #10, “For 7th grade, this would be very basic. A bar graph needs to be made. We could also do ratios of height to weight.” Although the teacher does not directly mention standards, the quote suggests the mismatch between the grade level where the science content might be placed and how seventh-grade students in a math class might work with this content.

This lack of discussion of standards on the survey contrasts with what the teachers said in the interviews. In the interviews teachers repeatedly equated the standards with the Ohio Achievement Tests (OAT), a high-stakes student assessment. The standards and the testing of the standards has reorganized the instruction in each of the schools and changed the teaching priorities; however, the teachers did not separate the two issues—standards and standardized tests.

In the case of the interviews all of the teachers raised issues with standards and their impact on teaching. For most of the teachers, standards were viewed negatively and as a hindrance to what they wanted to do as teachers. However, for Eileen the standards and standards-aligned textbooks provided structure that increased the students’ achievement test scores in science.

Eileen did not see these textbooks as being an impediment to her ability to integrate as she explained in response to the interviewer’s question, “What are some things that hold you back from doing math and science integration?”

For Cathy, the standards and OATs not only hindered her students’ opportunities to integrate science and mathematics, the tests ended science instruction, at least temporarily.

However, when the tests were over Cathy and her science colleague made time to do an integrated project with their students.

Becca taught in an urban school district that had established “pacing guides” for each subject in each grade level to ensure that all teachers covered the appropriate standards and benchmarks. For Becca, this constricted her freedom to attempt integration, “it’s sometimes an issue to kind of switch things around even though it may be beneficial to the students.” She also explained that the pacing guides were not written with integration in mind:

Finally, Dennis was the most negative about the role of standards and his ability to teach in an integrated fashion. Previous to the adoption of the state OATs and associated tests, Dennis felt he was better able to integrate instruction. His extended response below exemplifies his experiences.

It is clear that Dennis had a specific conception of mathematics and science integration in mind when he bemoaned these changes, but he, more than the other teachers, gave multiple examples of when he integrated mathematics and science, and how he at the time of the interview felt limited in his ability to do so.

The final aspect we considered in our work was the impact of the schools’ organizations as these related to the teachers’ ideas about integrating mathematics and science instruction. Although the larger school context for each school was different, all the interviewed teachers discussed a team organization for their students. These teams, as suggested by Beane and Brodhagen (2001) consisted of a core grade level of teachers who shared classes of students and were identified by some form of collective identity: “The upstairs team, team 5.1 as we call them, has the same set up” (Eileen). And in each of the schools there were multiple teams for each grade level. However, as suggested in Eileen’s quote, teams were separated, sometimes physically and in other cases pedagogically as shared by Cathy.

The team concept promoted a high degree of within team communication for each of the teachers. In all the schools the teams had at least one common planning time; however, for three of four teachers interviewed the focus of the common time was on student issues. Becca’s explanation of what happened in team planning time was typical.

Later Becca detailed team conversations as, “But there are also the informal, in the hallways and afterschool [meetings]. Especially when this past year, there was [sic] three of us right here and we often had informal conversations about our students.”

Only in Dennis’s case were there explicit discussions of curriculum and curriculum integration.

Although his team, too, had their share of discussions about students and administrative tasks.

Time to meet and plan was a frequent reason given for why teachers did not integrate instruction; however, all of these teachers had common time built into their weekly schedules. Often teachers used this common time to complete various administrative tasks. This time was also used for the benefit of students although in almost all cases the students discussed were struggling or in trouble. Only Dennis talked about a more generalized student discussion when his team planned birthday or other celebrations. Additionally, the team time was used for addressing accountability issues and planning for achievement tests.

Finally, we would like to return to the idea that the teams were creating separate teaching communities. Only Cathy spoke about working with her math counter-part on the other team, and when she did she used had highly negative overtones. The lack of discussion between two mathematics or science teachers about planning, instruction, or possible learning activities seems worrisome. It creates an impression of content completeness within the teams that is not tested or extended by conversation with other content teachers. Conversations between the mathematics and science teachers on the same team hint at a knowledge ownership that pushes the view of integrated instruction towards a co-teaching vision. This co-teaching allows the mathematics teacher to check off the math content and the science teacher to check off the science content, but the ‘intertwining’ of the content as discussed in the AAAS documents is lost.

The survey and interview responses highlighted differences in how the middle grades teachers in this study understood mathematics and science integration in two research contexts. The implication of these contrasting views needs to be carefully considered when teachers are asked to develop or implement integrated learning activities. We see curriculum integration as one way to achieve deeper mathematics and science instruction that is more aligned with what the early reform documents were asking for in terms of curriculum and learning. However, as these reform standards met with new curricular demands imposed by the states and federal accountability, attempts by teachers to connect the subjects were driven out.

The impact of standards-based reforms on teachers in their schools and classrooms seems to drive a wedge between a natural linkage of mathematics and science, linkages that draw on the multiple models for integration. Teachers have become more narrowly focused on achieving the content in each area. This may include being able to “name” the appropriate math and/or science as teachers did in the scenarios, or feeling the need to delay projects until after the high stakes tests.

Finally, although the structure of the middle school team model was in place for each of the interviewed teachers, the team structure did not seem to support integration as suggested by the reform documents. All teams were comprised of subject specialists, one for each area, with shared teachers in “specials” areas. This seemed to push the teachers to seeing the mathematics teacher or science teacher as the one responsible for teaching that subject and not the role of someone else. If teachers wanted to integrate instruction, they then needed to work in a coteaching situation with the other subject specialist. It is not that we are against co-teaching, but in these cases it does little to break down the disciplinary boundaries between the subjects that we believe is essential to an authentic model of integration.

Additionally, we were also concerned with the lack of mention of students and student learning throughout the teachers’ responses. In the surveys no teacher mentioned anything about student learning as a criterion for a scenario being integrated or not integrated. Although the survey was not specifically designed to generate student information, it was surprising that all of the responses focused on either content or teacher construction of the lesson. We were more surprised that there were no significant mentions of students and learning during the interviews. Integration was not considered to be a form of instructional differentiation to achieve learning. We were disappointed to find no discussion of how integration can support learning for students and were left to wonder if this was an artifact of the study design or whether teachers do not consider integration as an important learning process. In either case more research is necessary to understand this finding.

So as we reflect back to our initial premise, that educational reform should be seen as a coherent process that is enhanced by the structure of the curriculum, school organization and pedagogical initiatives, what does this study tell us? This study suggests that the team model for middle school organization does not lead to more integrated mathematics and science experiences for students, it may, in fact, limit integration to a singular model based on a concept of individual ownership of content knowledge, rather than the codevelopment of knowledge for each individual. Furthermore, the imposition of our current standards-based curriculum and testing systems has subverted the intent of the reforms they arose from and driven instruction into discrete content components, owned by each group and not typically shared across content areas when curriculum is set forth. So rather than making the middle school experience more coherent for students, perhaps conflicting reform strategies make learning more fragmented than previously.