Assessing Data Modeling and Statistical Reasoning (ADM)
The Assessing Data Modeling and Statistical Reasoning (ADMSR) project aims to develop an assessment system to evaluate elementary and middle school students’ skills and understandings related to data modeling and statistical reasoning. Data modeling is conceived as the application of statistical concepts and methods to investigate questions about the world. The ADMSR project will provide teachers with innovative and useful ways to assess students’ coordination of data with inquiry and their use of statistics. Our work builds on prior research and development under the Constructing Data, Modeling Worlds project (CDMW,see completed project section on the BEAR website). The assessment system is designed to accompany the CDMW curriculum, but can be used independently to support teaching of data modeling and statistics in other curriculums.
The BEAR Center collaborates with Vanderbilt University researchers to develop construct maps that tap 7 strands related to data modeling and statistical reasoning: (a) Theory of Measurement, (b) Modeling Measurement, (c) Data Display, (d) Meta Representational Competence, (e) Conceptions of Statistics, (f) Informal Inference, and (g) Chance. The team develops an embedded assessment system to track student progression on these constructs. Specific emphasis is given to the appropriate use of the assessment system in the classroom. The system is designed to be embedded seamlessly in the curriculum, facilitate classroom activities and discussion, and support formative and summative decisions. Teachers in middle schools in Nashville, TN and Phoenix, AZ, collect evidence (student responses, think-aloud, classroom videos) to support revisions of the constructs, assessment items, scoring guides, and instructional strategies. In collaboration with a second project, the BEAR Center will develop an electronic item database of items cross-indexed with constructs. Hence, the project will produce an assessment system that will increase both diagnostic and instructional capacity in an area vital to education in both mathematics and science.
- Triangulating Evidence to Investigate the Validity of Measures: Evidence from Discussion during Instruction, Cognitive Interviews, and Written Assessments.
- Analytical Framework for Conjecture-based Whole-class Mathematical Discourse.
- Challenges and Opportunities of Learning Progressions for the Psychometric Community
- Triangulating Validity Evidence from Classroom Discussions, Written Assessments, and Cognitive Interviews.” In J. Matthews-Lopez (Chair), Dimensions of Test Validation.
- “Investigation of the Validity of Evidence Obtained from Classroom Discussions.” In I. Grabovsky (Chair), Innovations in Measurement.
- Exploring the Contexts of Assessment: Comparing Evidence of Learning from Within the Classroom.
- Collaboration at the Boundaries: Brokering Learning and Assessment Improves the Quality of Education.
- "Exploring Contexts of Assessment." In R. Lehrer (Chair), Constructing A Multidimensional Learning Progression of Data Modeling: Design Studies, Psychometric Modeling and Brokering Professional Development.
- Making Sense of Student Responses to Assessment Items Using Scoring Exemplars.