THE INTERPLAY BETWEEN INSTRUCTION AND THE DEVELOPMENT OF MIDDLE SCHOOL STUDENTS’ ALGEBRAIC THINKING

 

Cynthia W. Langrall

Illinois State University

langrall@ilstu.edu

 

John K. Lannin

Illinois State University

lannin@ilstu.edu

 

Abstract:  This study examined the interplay between instruction and the development of middle school students’ algebraic thinking.  Interviews were conducted with six students over a two-year period during their 7th- and 8th-grade years.  In grade 7, all students received algebra instruction using units from the Connected Mathematics Project (CMP) materials.  In grade 8, three of the students took a traditional algebra 1 course while the other three students remained in the regular mathematics class following a typical pre-algebra curriculum supplemented with two CMP units.  Our findings indicate that although the algebra 1 students obtained a level of proficiency in algebraic symbol manipulation, their reasoning and sense-making abilities were less pronounced than in grade 7 and generally not as robust as those of the students who received the regular grade 8 mathematics instruction.

 

For more than a decade, the National Council of Teachers of Mathematics (NCTM) has recommended a broad-based, integrated curriculum for the middle grades that includes topics in probability, statistics, measurement, geometry, and algebra (NCTM, 1989).  In support of this recommendation, recently published Standards-based curriculum materials and even basal textbooks have included these topics as part of the middle grades curriculum (Reys, Robinson, Sconiers, & Mark, 1999).  More recently, the NCTM (2000) has called for “significant amounts of algebra and geometry throughout grades 6, 7, and 8” (p. 212) and has recommended that algebra and geometry instruction emphasize the interconnections between these topics and among other mathematics topics.  Nevertheless, a trend adopted by many school districts is to shift mathematics content from the high school to the middle school in the form of a traditional algebra1 course in grade 8.  According to Silver (1995) “mandated algebra instruction in grade 8 can undermine those reform efforts directed precisely at broadening and integrating the curriculum of middle grades” (p. 32).  There is also the concern that the content and teaching methods used in traditional algebra courses do not promote conceptual understanding or the development of students’ algebraic reasoning (Kaput, 1999).

The purpose of this study was to investigate the interplay between instruction and the development of middle school students’ algebraic thinking.  The first phase of this two-year study sought to describe the algebraic thinking of students whose mathematics instruction in grade 7 included the study of algebra using materials from the Connected Mathematics Project (CMP) curriculum.  In a departure from traditional algebra instruction that focuses on manipulating expressions and solving symbolic equations, the Standards-based CMP curriculum uses contextual problem situations to develop students’ concepts of variable and their understandings of the relationships among variables (Phillips, 1998).  The second phase of the study followed the development of the students’ algebraic thinking through grade 8.  However, in grade 8, the students split into two groups with one group receiving instruction that continued to incorporate some of the CMP materials and the other group receiving instruction based solely on a traditional Algebra 1 textbook.  The development of students’ thinking across grade levels and between the regular 8th-grade mathematics and traditional algebra 1 students was examined.

Theoretical Considerations

The theoretical perspective guiding our work assumes that mathematics should be taught for understanding.  We refer to the definition of Hiebert et al. (1997) who indicated that understanding something involves seeing how it is related or connected to other things that are known.  Understanding is crucial to learning mathematics “because things learned with understanding can be used flexibly, adapted to new situations, and used to learn new things” (Hiebert et al., p. 10).  According to Romberg and Kaput (1999), learning for understanding “cannot be viewed as a mechanical performance or an activity that individuals engage in solely by following predetermined rules” (p. 6).  Yet it is this mechanistic view of teaching and learning mathematics that characterizes traditional school mathematics (Romberg & Kaput).  Within this theoretical perspective, we examined the development of students’ understandings of algebra from two curricular orientations: one traditional and the other more Standards based.

Method

Participants

Grade 7 students from a Midwestern school formed the population for this study.  Based on the results of a researcher-generated algebra problem-solving assessment, six students were purposefully selected from this population to represent contrasting levels of performance.  These six students participated in the study during their 7th- and 8th-grade years of schooling.

The classroom teacher in this study taught all mathematics classes for grades 7 and 8.  He had 15 years of experience at the middle school level.  His teaching style was traditional in the sense that instruction was typically teacher directed during whole-class activities but he frequently engaged students in small-group interactions that allowed them to communicate and justify their thinking.

Year-One Instruction

In grade 7, approximately 9 weeks of instruction focused on the study of algebra.  During this time, we observed the mathematics class on at least two consecutive days each week to document the nature of algebra instruction.  We also noted the classroom interactions of the students, asked them to explain their reasoning as they worked on in-class assignments, and obtained copies of their written work when it was collected by the teacher.  The CMP texts Variables and Patterns (Lappan, Fey, Fitzgerald, Friel, & Phillips, 1998d) and Moving Straight Ahead (Lappan et al., 1998b) were the basis for algebra instruction.

Year-Two Instruction

In grade 8, three of the students were selected for participation in a traditional Algebra 1 class based on their overall 7th-grade mathematics performance and their performance on a standardized algebra readiness test.  The textbook used in this class was Algebra 1: An Integrated Approach (Benson et al., 1991).  The other three students remained in the “regular” mathematics class.  The curriculum for this class incorporated the text Pre-Algebra: An Accelerated Course (Dolciani, Sorgegnfrey, & Graham 1985), the CMP texts Looking for Pythagoras (Lappan et al., 1998a) and Thinking With Mathematical Models (Lappan et al., 1998c), and was supplemented with instructional units in geometry and statistics.  Classroom observations similar to those in Year 1 were conducted in Year 2.

Procedure

In Year 1, students were individually interviewed on seven occasions to investigate the interplay between students’ algebraic thinking and instruction.  Using a semi-structured interview script, we posed problems similar to the ones students were working on in class.  We also asked probing questions to assess levels of students’ conceptual understanding, and attempted to identify any cognitive obstacles the students were encountering.  Interview topics included linear and nonlinear relationships, tabular and graphic representations, concept of slope, and solving linear equations.  In Year 2, five interviews were conducted.  All of the students were given the same interview items that represented a balanced distribution of topics from the regular mathematics and algebra 1 curricula.  As in Year 1, we asked probing questions to assess students’ conceptual understanding and attempted to identify students’ misconceptions.  Interview topics included linear and non-linear relationships, interpretation of multiple representations, concepts of slope and variable, generating expressions, and writing equations.  All interviews were audio taped and later transcribed.

Data Sources and Analysis

Data sources consisted of the transcribed interviews, students’ written work, researcher field notes, and data displays and summaries generated during the analysis.  These data were analyzed using a double coding strategy and data reduction approach (Miles & Huberman, 1994) to discern key thinking patterns for each student.  Transcripts were analyzed to characterize each student’s reasoning or solution strategy for each problem presented in the interviews.  The coding rules used in this process were modified and refined throughout the analysis.  They pertained to students’ understandings of concepts (e.g., slope, variable), as well as their abilities to describe patterns or functional relationships, generalize relationships either verbally or with symbols, and use and interpret tabular, graphical, and symbolic representations.  Codes were reconciled and organized in a Task Coding Matrix that allowed us to examine within-case and across-case trends.  Narrative summaries were constructed for each student describing his or her reasoning, solution strategies, and cognitive difficulties across all interview problems.  Also, cross-case summaries were developed for each problem.

Results

In Year 1, students examined the relationships among graphic, tabular, and symbolic representations of linear situations.  In an attempt to uphold the philosophy of the CMP materials, the teacher engaged the students in small-group problem-solving activities and encouraged students’ explanation and justification of solution strategies.  However, the teacher occasionally supplemented the CMP lessons with worksheets designed to introduce and practice procedures for symbolically solving linear equations.  By the end of grade 7, three trends in students’ thinking were identified: (1) all students could represent linear situations in symbolic, graphic, and tabular forms; (2) students demonstrated an emerging understanding of the concept of variable in the sense that they could use a variable when it was defined for them but sometimes had difficulty identifying what a variable meant in a problem context; and (3) half of the students (two who later participated in regular 8th-grade instruction and one who took the traditional algebra class) consistently referred to the problem context to make sense of a problem situation while the other three students generally ignored the problem context and focused on superficial patterns.

In Year 2, the traditional algebra 1 class focused on manipulating symbols in the context of finding unknowns, using properties of generalized arithmetic, and solving problems by representing situations symbolically.  The typical lesson followed a mechanistic approach (Romberg & Kaput, 1999) whereby the teacher demonstrated a procedure, provided a few examples for students to work individually, and assigned practice exercises.  The algebra instruction in the regular mathematics class continued to build on the algebraic understandings introduced in grade 7.  Students informally investigated the graphic, tabular, and symbolic representations involving both linear and nonlinear relationships.  It should be noted, however, that only one of the four designated grade 8 algebra units from the CMP materials was used.

By the end of grade 8, differences in the students’ understanding and reasoning emerged between the two instructional groups.  The algebra students were more proficient at solving equations and simplifying expressions.  Although all students could represent situations in various forms when asked to do so, the students in the regular mathematics class accessed a variety of representations to solve problems, while the algebra 1 students focused exclusively on symbolic representations.  The regular mathematics students referenced alternative representations to verify their work, whereas the algebra 1 students rarely sought other representations to confirm the validity of their work.  Although all students struggled to understand the different uses of variables, students from the algebra group often referred to, and used, variables as unknowns while ignoring the underlying meaning of their symbolic form.  In contrast, the regular mathematics students maintained a dynamic view of variables and frequently questioned what variables represented.  Justifying one’s thinking was another difference that emerged between the two groups.  The algebra 1 students often stated “I just know this” rather than offering any justification of their reasoning.  For them, providing an explanation did not seem to be an integral part of doing mathematics.  The regular mathematics students justified their reasoning when asked or provided an explanation as they talked their way through solving the problem.

Conclusions

Our results indicate that although students in the traditional algebra course acquired some level of proficiency in algebraic symbol manipulation, their reasoning and sense-making abilities were generally lacking.  In many respects, the reasoning and sense-making of the algebra 1 students was less pronounced than it had been in grade 7 and was not as robust as that of the students who remained in the “regular” grade 8 mathematics class.  The findings of this study underscore the concern that a traditional algebra 1 course in grade 8 may promote a mechanistic view of algebra and fail to develop concepts and understandings that would allow students to reason algebraically in new situations.

References

Benson, J., Dodge, S., Dodge, W., Hamberg, C., Milauskas, G., & Rukin, R. (1991).  Algebra 1: An integrated approach.  Evanston, IL: McDougal, Littell.

Dolciani, M. P., Sorgegnfrey, R. H., & Graham, J. A. (1985).  Pre-algebra: An accelerated course.  Boston, MA: Houghton Mifflin.

Hiebert, J., Carpenter, T. P., Fennema, E., Fuson, K. C., Wearne, D., Murray, H., Olivier, A., & Human, P. (1997).  Making sense: Teaching and learning mathematics with understanding.  Portsmouth, NH: Heinemann.

Kaput, J. J. (1999). Teaching and learning a new algebra.  In E. Fennema & T. A. Romberg (Eds.), Mathematics classrooms that promote understanding (pp. 133-155).  Mahwah, NJ: Earlbaum.

Lappan, G., Fey, J. T., Fitzgerald, W. M., Friel, S. N., & Phillips, E. D. (1998a).  Looking for Pythagoras.  Menlo Park, CA: Dale Seymour.

Lappan, G., Fey, J. T., Fitzgerald, W. M., Friel, S. N., & Phillips, E. D. (1998b).  Moving straight ahead: Linear relationships.  Menlo Park, CA: Dale Seymour.

Lappan, G., Fey, J. T., Fitzgerald, W. M., Friel, S. N., & Phillips, E. D. (1998c).  Thinking with mathematical models:  Representing relationships.  Menlo Park, CA: Dale Seymour.

Lappan, G., Fey, J. T., Fitzgerald, W. M., Friel, S. N., & Phillips, E. D. (1998d).  Variables and patterns:  Introducting algebra.  Menlo Park, CA: Dale Seymour.

Miles, M. B. & Huberman, A. M. (1994).  Qualitative data analysis: An expanded sourcebook (3rd ed.).  Newbury Park: Sage.

National Council of Teacher of Mathematics. (1989).  Curriculum and Evaluation Standards for School Mathematics.  Reston, VA: NCTM.

National Council of Teacher of Mathematics. (2000).  Principles and Standards for School Mathematics.  Reston, VA: NCTM.

Phillips, E. (1998), Developing a coherent and focused K-12 algebra curriculum.  In National Research Council (Ed.), The nature and role of algebra in the K-14 curriculum: Proceedings of a national symposium, (pp. 27-29).  Washington, D.C.: National Academy Press.

Reys, B., Robinson, E., Sconiers, S., & Mark, J. (1999) Mathematics curricula based on rigorous national standards:  What, why, and how? Phi Delta Kappan, 80, 454-456.

Romberg, T.J. & Kaput, J.J. (1999).  Mathematics worth teaching, mathematics worth understanding.  In E. Fennema & T.A. Romberg (Eds.), Mathematics classrooms that promote understanding, (pp.3-18). Mahwah, NJ: Earlbaum.

Silver, E. A. (1995).  Rethinking “algebra for all.”  Educational Leadership, 52(6), 30-33.