Co-Principal Investigator: Williford, Amanda

Purpose: This study will explore whether and how the use of formal and informal math assessments is associated with greater child math gains. The use of assessments has expanded rapidly in recent years, driven in part by federal initiatives. However, the research basis for these assessments is limited, especially in terms of linking the use of assessments to improved student outcomes. Researchers will examine direct associations between math assessment use and student learning gains. They will also examine links among teacher math assessment, teacher use of specific instructional practices to meet individual student needs, and student math learning. Researchers will also examine teacher data literacy and math knowledge for teaching as factors that may affect how assessment data are used.

Project Activities: In Year 1, researchers will conduct a qualitative study to explore malleable factors associated with teacher use of math assessments and adjustments to instruction. Researchers will also finalize instrumentation, so that measures reflect the range of math assessment, data use, and related instructional practices in use in kindergarten classrooms. In Years 2 and 3, researchers will conduct the main observational study, using 2 independent sequential cohorts to explore associations among math assessment use, use of ability grouping, teacher data literacy, teacher math knowledge, and student learning gains. In Year 4, researchers will complete analyses and report results.

Products: Researchers will produce preliminary evidence of potentially promising practices. They will share this evidence in peer-reviewed publications and brief reports describing and interpreting findings to participating schools as well as CASTL website and email list. In addition, researchers will present findings at meetings with school administrators across the State and at practitioner-oriented conferences in Virginia such as Virginia ASCD, Virginia Council of Teachers of Mathematics, or the Virginia Association for the Education of Young Children (VAEYC).

Structured Abstract

Setting: The project will take place in Virginia.

Sample: Twenty school staff (principals and math specialists) and 20 teachers will participate in interviews during a qualitative study in Year 1. The main observational study will include 150 kindergarten classrooms (75 in Year 2, 75 in Year 3) and 600 students (4 per classroom). The demographic characteristics of the students and teachers will be diverse, with children of mixed income levels and racial and ethnic backgrounds reflecting the diversity of Virginia.

Malleable Factors: This project will focus on the following malleable factors: teacher use of formal and informal math assessments; teacher instructional practices (for example, use of ability grouping); teacher data literacy; and math knowledge for teaching.

Research Design and Methods: Researchers will interview teachers about their use of assessment and data. For the observational study, 2 sequential independent cohorts of teachers and children will participate. Researchers will survey teachers 3 times per year about their instructional practices, assess student math skills in the fall and spring using a direct assessment and teacher rating scale, and observe math instruction using a classroom-level and a child-level observation measure.

Control Condition: Due to the exploratory nature of the research design, there is no control condition.

Key Measures: Virginia's statewide student math assessment – the Early Math Assessment System (EMAS) — and the teacher-report Academic Rating Scale — Math will measure student math learning and be administered in the fall and spring. For students who are identified by teachers as Spanish-speaking dual language learners, they will be given the Bateria III Woodcock-Munos-Problemas Aplicadas to measure Spanish-language math skills. Researchers will use surveys adapted from the Teacher Data Use Survey and practices used by the ECLS-K:2011 study to measure teacher self-reported data use, use of data to inform instruction, and use of ability groups. Teachers will complete a data literacy assessment and a measure of math content for teaching. The research team will measure observed teacher instructional practices using the Individualizing Student Instruction — Math coding system and Classroom Observation of Early Math Teaching (COEMET) coding system.

Data Analytic Strategy: The research team will analyze qualitative data using consensual qualitative coding. They will analyze quantitative data using multilevel models to examine the associations between teacher data use, teacher instructional practices, and student math learning gains. The research team will use multilevel path analysis to test the extent to which the effect of formal and informal assessment on child outcomes might be mediated by math instructional practices.