Assessing Data Modeling and Statistical Reasoning (ADM)

Collaborating Institutions: 
Learning Sciences Institute, Vanderbilt University; Berkeley Evaluation & Research (BEAR) Center, UC Berkeley

Funded By: 
U.S. Department of Education, Institute of Education Sciences

Vanderbilt University Participants: 
Rich Lehrer (co-PI), Seth Jones

BEAR Participants: 

Former Staff: 
Tzur Karelitz, Kavita Seeratan, Kristen Burmester, Mary Full, Bob Schwartz, Jooyong Park

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.

Persons:

Highlights

Browse some of the instructional resources developed in the ADM project by visiting:

  1. http://modelingdata.org/