Developing an Instrument to Assess Informal and Formal Statistical Inference: An Argument-Based Approach to Validity

• In Event: Poster Session 3
  In Poster Session: Test Validity Research and Evaluation

Abstract

Concepts of formal statistical inference are often misunderstood. There has been much recent interest in informal statistical reasoning, which is defined as “the way in which students use their informal statistical knowledge to make arguments to support inferences about unknown populations based on observed samples” (Zieffler, Garfield, delMas, & Reading, 2008). This paper describes the development and validation of an instrument for assessing informal and formal statistical reasoning. As validity is a fundamental assessment of how well an assessment supports particular inferences related to specific uses, this study subsequently collects and integrates a diverse set of validity evidence. An argument-based approach to validity was used as a theoretical framework to validate a measure of college students’ statistical inference.

Authors

• Jiyoon Park, Federation of State Boards of Physical Therapy

• Robert C. Delmas, University of Minnesota