Elisa Stone
Cal Teach Berkeley
UC Berkeley

Stephen Moore
BEAR Center
UC Berkeley

Kathleen Scalise
BEAR Center
UC Berkeley

Assessing Student Progress Toward College Readiness with Psychometric
and Cognitive Models of Student Learning in Mathematics

Diana Wilmot, Santa Clara County Office of Education

Current college admissions tests (a) inadequately predict college success (Geiser, & Studley, 2002), and (b) provide insufficient information for students regarding college readiness at earlier stages
of their academic careers (Kirst, 2003; Olson, 2006). This research informs the creation of a more predictive and informative test for the University of California, by investigating selected college readiness standards in mathematics. In particular, this study investigates how students develop an understanding of mathematical functions from 6th through 12th grade.

An instrument that measures student achievement in this area has been constructed following the four principles of the BEAR Assessment System (Wilson, 2005): 1) a developmental theory of learning (referred to hereafter as a construct map) in the area of mathematical functions; 2) an item design and scoring guide that aligns with current mathematics standards, curriculum and instruction; 3) evidence to support the theories and the measures using the Rasch model (Rasch, 1961,1980), and interviews with students and teachers; and 4) an effective way of communicating the results, through visual, interpretive maps, known as Wright Maps (Wright & Masters, 1982). This mixed method approach has the potential to offer a measure of college readiness that will help students make much smoother transitions from high school to college. This research provides concrete evidence of how to bring the benefits of cognitive research to admissions testing. It demonstrates the sorts of interpretation, framework, and items that may help in the development of a UC-generated admissions test. Further efforts using this research could eventually provide teachers with a framework to guide the vertical articulation of their math curriculum.

top

Feb. 16

TBA

Ruth Chung Wei and Leah Walker, Stanford University



top

TBA

Elisa Stone, Cal Teach Berkeley, UC Berkeley

top

Desired Results Developmental Profile (DRDP):
The California Department of Education's application of the BEAR assessment system in state-funded early childhood education programs

Stephen Moore, BEAR Center, UC Berkeley

Since 2001, Mark Wilson and BEAR Center researchers have led an effort on behalf of the Child Development Division (CDD) of the California Department of Education (CDE) to develop and validate the Desired Results Developmental Profile (DRDP). DRDP is now the core assessment system in every CDE-funded early childhood education program, and is mandatory for all children served in these programs state-wide. DRDP represents a realization of the benefits to classroom teachers, administrators, and policy-makers that are inherent in the BEAR Assessment System (BAS). The story of DRDP development is a classic example of the psychometric and scientific building blocks of BAS methodology in action, and of the flexibility of BAS methodology to adapt to novel applications. This presentation will describe what the DRDP is, how its results are used, the studies conducted and the models used to develop and refine DRDP instruments and calibrate its measurement scales, and the new directions for DRDP that CDE has commissioned BEAR to lead in the future. There will also be a demonstration of DRDPtech, the online version of DRDP that the BEAR Center's Information-Technology Group has developed for CDE.

top

Formative Assessment Delivery System (FADS):
The Development of Resources and Tools for Teacher Assessment of Student Learning

Kathleen Scalise, BEAR Center, UC Berkeley

The Formative Assessment Delivery System (FADS) builds on extant research in the design of assessment tasks and measurement techniques to produce tools that teachers can use in their classrooms for formative assessment. It is a computerized system to design, develop, deliver and report on assessments within an interpretive context that helps teachers accurately diagnose students' comprehension and learning needs. The interpretive context is key to the alignment of educational goals, instructional content, and classroom and large scale assessment. The project involves (1) analyzing selected mathematics and science curricula to derive appropriate standards-based constructs that are assessable within the instructional activities, (2) identifying from these curricula instructional and assessment activities that relate to these constructs and also lend themselves to computerized delivery, (3) developing approaches to deliver the activities and gather and score student responses viacomputer, and (4) producing reports of student performance that inform teachers' understanding of student learning and help teachers plan next instructional steps.

top

TBA

Kenneth Bollen, University of North Carolina