Purpose: Embedding assessment practices within instructional materials gives teachers a tool to monitor student learning, provide immediate feedback to students at key junctures, and adjust instruction accordingly. Because science curricula typically include complex, sequential sets of knowledge and skills to master, embedded formative assessment may be a useful component of science curricula. WestEd, in collaboration with the University of California at Los Angeles, is planning to test the efficacy of the Assessing Science Knowledge diagnostic formative assessment system, which was developed to be an integral part of the Full Option Science System elementary science modules published by Delta Education, Inc.

Project Activities: The research team is using a group-randomized design to study the impact of using the Assessing Science Knowledge system in conjunction with the Full Option Science System in fourth-grade classrooms. Assessing Science Knowledge comprises two kinds of assessments: embedded and benchmark. The embedded assessments are incorporated into instruction and provide continuous information about students' learning to both teachers and students. The benchmark assessments are summative measures of students' accumulated knowledge and understanding of science.

Products: Products from this project include published reports regarding the efficacy of an embedded diagnostic assessment system for elementary school science.

Structured Abstract

Purpose: Embedding assessment practices within instructional materials gives teachers a tool to monitor student learning, provide immediate feedback to students at key junctures, and adjust instruction accordingly. Because science curricula typically include complex, sequential sets of knowledge and skills to master, embedded formative assessment may be a useful component of science curricula. WestEd, in collaboration with the University of California at Los Angeles, is planning to test the efficacy of the Assessing Science Knowledge diagnostic formative assessment system, which was developed to be an integral part of the Full Option Science System elementary science modules published by Delta Education, Inc.

Setting: Elementary schools in Arizona.

Population: The study sample consists of approximately 140 fourth-grade teachers and their 3,500 students in 70 elementary schools in Arizona. The sample includes schools with high proportions of historically low-performing students to better ascertain the impact for groups of students under-represented in higher education.

Intervention: Assessing Science Knowledge is a diagnostic formative assessment system. Currently, Assessing Science Knowledge is integrated into 16 modules of the Full Option Science System, a widely used K-8 science curriculum. The diagnostic assessment tools in Assessing Science Knowledge are designed to monitor student progress at critical junctures in the curriculum, enabling teachers to adjust their instruction according to how well their students are mastering the material. Assessing Science Knowledge comprises two kinds of assessments: embedded and benchmark. The embedded assessments are incorporated into instruction and provide continuous information about students' learning to both teachers and students. The benchmark assessments are summative measures of students' accumulated knowledge and understanding of science. The content covered by the four Full Option Science System modules examined in this study (Variables, Landforms, Environments, Mixtures and Solutions) maps directly to several of the Arizona State standards for science at the elementary grades.

Research Design and Methods: The study uses a group-randomized design with two groups, treatment and control. Seventy schools are randomly assigned to treatment or control conditions with two participating teachers at each school. All schools use the Full Option Science System curriculum.

Control Condition: Teachers in schools randomly assigned to the control condition are teaching their fourth-grade students using the Full Option Science System modules without the Assessing Science Knowledge system.

Key Measures: Key measures include grade 4 science scores on the Arizona State Science Test and the Assessing Science Knowledge tests for each of the four Full Option Science System modules being examined in this study.

Data Analytic Strategy: The primary analyses involve fitting conditional mixed effects ANCOVA models (multilevel models), with an additional term to account for the nesting of students within schools.

Products and Publications

Journal article, monograph, or newsletter

Herman, J., Osmundson, E., Dai, Y., Ringstaff, C., and Timms, M. (2015). Investigating the Dynamics of Formative Assessment: Relationships between Teacher Knowledge, Assessment Practice and Learning. Assessment in Education: Principles, Policy & Practice, 22(3): 344–367.