Examining Dreams and Considering Possibilities: Reflections on a Formative Assessment Intervention Study

• In Event: Formative Assessment for Mathematics and Science Standard Test

Abstract

Purpose
With the advent of Common Core State Standards in Mathematics (CCSS-M), schools have sought to infuse more complex thinking and reasoning into mathematics instruction. This poster explores the development of a FA tool for middle school math that gives teachers and students an opportunity to engage in the process using high-quality materials. This poster provides a) a brief overview of tool development, b) findings from an intervention study, and c) remaining questions.
Theoretical framework
FA or the “process used by teachers and students during instruction that provides feedback to adjust ongoing teaching and learning to improve students’ achievement of intended instructional outcomes” (Council of Chief State School Officers (CCSSO), 2008) offers a venue to promote more complex mathematical thinking. Black & Wiliam (1998) presented substantial evidence supporting the notion that FA can improve students’ learning outcomes, raise standards, and support lower achievers. Literature also suggests that FA can increase students’ motivation and self-esteem (Miller & Lavin, 2007).
Methods
Based on previous phase case study findings, the research team developed five FA tasks, associated rubrics, support materials, and PD activities (including webinar discussions to review student work). The tasks incorporate key attributes of the FA process and use an engaging structure or template. Current tasks focus on proportional reasoning concepts (CCSS 8EE5-6) and emphasize multiple representations to explain one’s reasoning. Task development included a review of learning progressions and multiple rounds of review/revision by expert mathematics teachers as well as pilots.
Results
Tasks were implemented as part of a randomized control study involving two medium school districts. Schools were randomly assigned to intervention and control conditions. In total, 11 teachers, 36 classes, and more than 600 students participated. Student demographics across the two groups were similar, with substantial proportions of the students eligible for free lunch and sizeable proportions of Latino students in both.
ANCOVA analyses revealed significant albeit small differences between conditions, with treatment students scoring higher in relation to control students on the post-test overall (F = 11.01, p = .00, η2 = .03) and attitudinal post-survey (F = 6.86, p = .01, η2 = .02). Effect sizes (ES) were relatively modest in nature (partial eta squared = .02 - .03). Albeit a small sample, treatment teachers agreed that the tasks fit well with their unit of instruction and revealed students’ misconceptions.
Significance
Many questions remain that may guide future FA projects. Most teachers reported they had used information gathered to adjust their instruction. Yet, we were not fully able to document how this information impacted teachers’ feedback and next steps are given constraints in observing classrooms. How did activities spur extended student talk, particularly in instances with more than one “correct” answer? Researchers worked to create easy-to-use resources given teacher time constraints. Yet, we also recognize the need for more in-depth examination of student work (why students do what they do) and foundational content knowledge.

Authors

• Joan L. Herman, University of California - Los Angeles

• Christine Ong, University of California - Los Angeles