The American Institute for Research (AIR) is holding a conference on August 22-23, 2016 as an attempt to facilitate the transition from the Next Generation Science Standards (NGSS) content standards to NGSS assessments. Content and assessment leaders from about 10 states as well as nationally recognized NGSS experts (Peter McLaren, Brett Moulding and Aneesha Badrinarayan) and psychometricians (Laurie Wise and Tom Fisher) will discuss what needs to be done to translate NGSS content standards into reliable and valid NGSS summative assessment.
The NGSS were developed by a consortium of states and by the National Science Teachers Association, the American Association for the Advancement of Science, the National Research Council and Achieve. The standards represent the scientific core ideas and practices that students should learn by the end of high school. The standards go beyond the subject matter orientation that was used in previous science standards.
The NGSS conceptualizes science learning as a multidimensional process where students learn the fundamental core ideas in science as well as the scientific practices that are at the heart of the scientific method. Various states are in the process of adopting the NGSS as well as developing curriculum and instruction materials for the implementation of the standards. States are also planning to develop assessments that measure the multidimensional nature of the NGSS. This presents item writing challenges for test developers who must write items that measure multiple dimensions and statistical challenges for psychometricians that must report reliable and valid science results in multiple dimensions.
Test design begins with a clear statement of the purpose of the assessment, and the inferences that the test is designed to support. This panel will focus on the summative assessment, its intended uses, and the inferences that will be required to serve those purposes. The bulk of the panel will focus on defining the inferences—the claims that the test should support about the student, the school, and the district.
If successful, participants and panelists will leave the event with a clear statement of the inferences the summative assessment should support. These inferences, or claims, will be realistic in the sense that they will require a test design that is of reasonable length to support the validity of those inference with reasonable reliability. The psychometricians in attendance will help evaluate the implication of proposals for test design.
Proceedings of the conference will be available following the conference.
NGSS Panel - Sample of Sessions
State Policy Perspectives on NGSS
- Ajit Gopalakrishnan CT
- Vaughn Rhudy, WV
Inferences from NGSS Tests
NGSS experts will respond to questions from AIR about how to turn NGSS into a statewide reliable and valid summative test.
- Aneesha Badrinarayan, Achieve
- Rodger Bybee, Biological Sciences Curriculum Study
- Peter McLaren, Next Gen Education Consulting
- Brett Moulding, Partnership for Effective Science Teaching/Learning
Psychometric Commentary in Response to Experts
Psychometricians will respond to the Experts’ presentations
- Laurie Wise, HumRRO
- Tom Hirsch, AES
What Claims Will We Report On?
Implications of Claims for Test Design
Summary of Claims and Overview of Test Design
Next Steps, Moving Forward
- Gary Phillips, AIR