One of the most vexing problems in education is the persistence of an achievement gap between middle class students of European American descent and students from historically underserved communities. This gap is reflected in performance on a range of achievement tests, from NAEP to state measures being instituted under NCLB. Despite the increased use of standardized test results to make consequential decisions in education, little is actually known about the kinds of knowledge, reasoning, and interpretive practices that students - both those who do well on such tests and those who do not - use in responding to them.
The proposed research is a first step in addressing this gap in the field's knowledge. The main line of inquiry will investigate how children from diverse ethnic and linguistic communities make sense of science items on the Grade 4 Massachusetts Comprehensive Assessment System (MCAS), a high stakes achievement test viewed as a model in meeting NCLB requirements. The outcome will be a deeper understanding of a) the conceptual, linguistic, textual and representational demands of the science items on the MCAS, and b) the forms of knowledge, reasoning and interpretive practices that children from diverse backgrounds use in responding to such items. A second, more exploratory line of inquiry will investigate what, if anything, teachers and administrators may learn from close examination of children's reasoning about achievement tests. It will focus on what they learn about a) children's performance and competence, b) how tests like the MCAS work, and c) possible relationships of curriculum and instruction to test performance. An important outcome of both lines of inquiry will be the specification of guidelines for test-makers and practitioners to use in evaluating the meaning of students' test performance.
Intellectual Merit:
The aim of this project is to contribute to the development of a fuller science of assessment that serves the goal of equity. In pursuit of this goal, this project will bring together sociocultural, sociolinguistic, cognitive, and psychometric perspectives on human performance to develop a theoretically and empirically grounded understanding of the interpretive work in which children engage as they respond to items on science achievement tests.
This research will shed light on heretofore unidentified interactions between children and specific features of science test items (e.g., linguistic and representational forms) that may affect performance in unintended ways.
Broader Impact:
This project will break new ground by accounting for the complexity of the sense-making interaction between children and standardized tests, thus deepening understanding of the achievement gap, our nation's most persistent educational problem. The results of this research will be translated into guidelines for test makers and practitioners. Guidelines for testmakers will address the implications of the findings of this research for the design of test items and interpretation of children's performance. Guidelines for practitioners will address how analysis and discussion of student test performance can inform their practice and support coordination among curriculum, instruction, and assessment in ways that best serve the education of all students.
