Poster #45 - Comparing Information Obtained From Multiple-Choice and Constructed-Response Next Generation Science Standards–Aligned Test Items Assessing Student Ideas About Evolution and Natural Selection

• In Event: 25.087 - MTCC Poster Session Four
  In Poster Session: 25.087-6 - Assessment in Schools: Poster Session 2

Abstract

The Next Generation Science Standards call for assessments that assess integrated, three-dimensional (3D) science learning in which disciplinary core ideas, science practices, and crosscutting concepts are tested together. The National Research Council has suggested that these assessments should use a combination of item formats including constructed-response (CR) and multiple-choice (MC). Previous research has shown that both formats provide psychometrically equivalent information if attention is paid to aligning the items to a common construct. Less attention has been paid to the specific information that can be obtained or how to do this when the goal is to test integrated 3D learning. This study uses both psychometric and construct analyses to compare the benefits and limitations of the MC and CR formats.

Authors

• George E. DeBoer, American Association for the Advancement of Science

• Joseph Hardcastle, American Association for the Advancement of Science