Abstract
This study investigated the effects of a two-year long professional development (PD) program called Cognitively Guided Instruction (CGI) on teachers’ instructional practices for engaging students in problem-solving in elementary mathematics classrooms. The treatment involved sixteen days of CGI PD across two years. Forty-two and forty-seven teachers’ classrooms were observed and videotaped during first- and second-years, respectively. Videotaped classroom instruction was coded, and the resulting data were analyzed to compare students’ problem-solving opportunities between CGI and non-CGI classrooms. This study found that CGI teachers were more likely (than not), in comparison to non-CGI teachers to provide their students with various problem-solving opportunities by implementing instructional processes such as: posing high-cognitive tasks; posing word problems; implementing task at high-cognitive level; spending time on single problems; allowing students to choose their solution method; providing access to manipulatives; encouraging multiple strategies; giving students time to solve problems; asking students to share their solutions; spending time discussing problem as a whole-class; and listening to student thinking of solutions. The study findings suggest that CGI PD has a large positive effect on teacher instruction of problem-solving. Using present-day statistical methods and investigations conducted in classrooms set in present-day educational context, this study successfully updated a portion of Carpenter, Fennema, Peterson, Chiang and Loef (1989). This study’s findings also extended original findings to include: additional arithmetic topics; second-grade classrooms; additional instructional processes; and additional year of PD implementation, to provide additional opportunities for causal inferences linking CGI to student problem-solving opportunities.