Reflections from the end:

COMMSTEP accomplished its goal of upgrading many middle school teachers' mathematics knowledge and assisting them in implementing their chosen curriculum. However, the project assumed there would be little or no teacher and/or administrator turnover. Our goal of providing 130 hours of professional development for all project teachers proved unrealistic and naive. By the conclusion of the project over 500 teachers will have benefited from professional development sponsored by the project. Despite over 50% teacher turnover, over 250 teachers will have experienced at least 100 hours of professional development. On the Horizon Research mathematics questionnaire, 88% of responding teachers report they have increased their knowledge of mathematics, are using a variety of instructional methods, and can assess student understanding using a variety of techniques.

The order of the promoters and inhibitors indicates their relative importance.

Major promoters include:

1. Highly-qualified teacher mentors. The project was fortunate to have a staff of teacher mentors who were knowledgeable about math content, effective pedagogy, assessment alternatives, and adult learners. They visited each project teacher's classroom every month. Mentors met regularly to network with each other and to participate in professional development that strengthened their roles as mentors, e.g., Cognitive Coaching.
2. Highly-qualified leadership team. The project's leadership team consisted of a former Co-Principal Investigator for the Colorado Statewide Systemic Initiative, the Project Director/Principal Investigator of TEAM 2000, the former Colorado Department of Education Senior Consultant for Mathematics, a Professor of Mathematics, and an Associate Professor of Mathematics Education. The team worked well together drawing upon their different perspectives and areas of expertise as they provided guidance in the project's design and implementation.
3. Support from district-level math coordinators. These administrators helped recruit project schools, followed through with building administrators and teachers regarding project commitments, and provided continuing education credit.
4. District-level financial support. Each district fully supported the purchase of all curriculum materials including books, manipulatives, and graphing calculators. Substitutes were provided for teachers attending workshops on school days and stipends for teachers attending Saturday workshops.
5. Additional funding that enabled increased mentoring. The project received over $500,000 from local foundations and granting agencies to enhance the $1.5 million NSF grant. This financial support allowed the project to increase the number of teacher mentors so that mentors visited project teachers' classrooms monthly.
6. Assessment project and scoring conference. At the request of a major local granting agency, the project created a pre/post-test that was used with all 6th, 7th and 8th grade students in the project to measure gains in student achievement. The test consisted of both multiple-choice and constructed-response items. Test items were selected from the National Assessment of Educational Progress (NAEP), the Third International Mathematics and Science Study (TIMSS), the New Standards Project, and from the three middle school mathematics curricula supported by the project: Connected Mathematics, Mathematics in Context, and MathScape. Teachers participated in scoring the student assessments each summer. The following fall each teacher received individual student results that they analyzed by type and content standard. Both the scoring conference and follow-up workshop proved to be powerful learning opportunities for teachers.
7. State-level math standards and assessment program. Because the Colorado Model Content Standards for Mathematics are based on the NCTM curriculum standards, the standards have driven the use of standards-based instructional materials, such as those used in COMMSTEP schools. Since the Colorado Student Assessment Program (CSAP) is based on state standards, implementing these curricula well has resulted in higher student achievement.

Major inhibitors include:

1. Teacher turnover. In most project schools at least one new teacher has been hired each year of the project. In some schools there has been a complete turnover of the entire mathematics department. Over 50% of teachers are no longer teaching mathematics in project schools. The biggest surprise was to learn that the majority of these teachers are still at their schools, but teaching a subject other than mathematics. To deal with the turnover issue, COMMSTEP has conducted additional week-long Connected Mathematics institutes prior to the start of school for the past two years. New teachers are also included in grade-level meetings with COMMSTEP mentors. However, teacher turnover will continue to be a major problem for quality curriculum implementation unless ongoing professional development is provided to address the needs of new teachers.
2. Administrator support/turnover. In addition to teacher turnover, 50% of building-level administrators are either retired or moved to a different school/district. A majority of principals indicate they do not feel comfortable with mathematics. Administrator support and consistency of leadership are critical to the success of projects such as COMMSTEP.
3. Support of three curriculum projects. When the grant proposal was first conceived, the leadership team felt it important to give participating schools a choice among three curricula (Connected Mathematics Program, Mathematics in Context, and MathScape). It has been very difficult to provide professional development on all three curricula at three grade levels. Given that over 80% of project schools are implementing CMP, Mathematics in Context and MathScape teachers sometimes feel overlooked.
4. Differing levels of experience. In a few schools the project supported teachers who had taught CMP before becoming part of COMMSTEP. A small number of teachers had attended summer workshops at Michigan State University. Even though the majority were not implementing the curriculum well, the few teachers who had attended professional development sessions at MSU were more likely to implement CMP as designed. It was difficult to meet the needs of these "experienced" teachers in workshop settings; however, mentoring sessions provided the individualization these teachers needed.
5. Differing levels of school-year professional development. COMMSTEP's original design provided monthly workshops for sixth-grade teachers and quarterly workshops for seventh- and eighth-grade teachers. The project design focused "up front" efforts on sixth-grade teachers for several reasons: traditionally less mathematics content background, entry point for middle school, and entry point into new curriculum. However, by not providing as many school-year workshops for seventh- and eighth-grade teachers, curriculum implementation was more "sluggish" at these levels when compared with sixth-grade teachers' efforts.
6. State assessment program. Although the Colorado Student Assessment Program (CSAP) is aligned with state model content standards and standards-based mathematics curriculum, some teachers, administrators, and district assessment coordinators look for a "quick fix" to address the testing issue. "Test Prep" programs are put in place that detract from the conceptual underpinnings of curriculum being implemented and focus students on isolated skill problems with no connection to current lessons. Although 25% of CSAP items are released each year, teachers continue to have a skewed view of the content of the assessment believing that computation is a major emphasis.
7. Insufficient time for teaching math. Several schools still allocate no more than 45 minutes per day for mathematics teaching. This time frame makes it extremely difficult for teachers to allow students the exploration time necessary to fully understand curriculum concepts. Lessons are hurried and usually lack closure. In addition, literacy continues to dominate most school-based professional development sessions.