Development and validation of an observation-based protocol to measure the eight scientific practices of the next generation science standards in K-12 science classrooms

The eight scientific practices in the Next Generation Science Standards (NGSS) are significant theoretical constructs that reflect the nature of science; they are intended to guide science teaching and learning. Yet, operationalization of these practices in terms of student learning remains limited, and a measurement tool for the eight practices is needed. Using an Interactive-Constructive-Active-Passive (ICAP) theoretical framework, this study develops such a protocol, entitled ICAP to Measure by Observation NGSS Science Practice Implementation in the Classroom (IONIC). This protocol can be applied in K-12 science classrooms to determine the level of student engagement with the eight practices. Data were collected from 152 video-recorded lessons obtained after a 3-year professional development (PD) program designed to engage teachers to better understand and use the NGSS practices. We examined validity (content, face, translation, and construct) and reliability (internal and interrater) of IONIC. Confirmatory factor analysis was conducted to verify three competing theoretical models: Parsimonious (three phases of knowledge development: Investigation, Sensemaking, and Evaluating and Communicating, merged with the eight NGSS practices), NGSS (the eight practices), and Comprehensive. The latter consists of an upper-level (Parsimonious) and a lower-level (NGSS). The results suggest the Comprehensive Model is the preferred model among the three. In the upper-level, the score for Phase 3, Evaluating and communicating, is significantly lower than the other two. In the lower-level, classroom scores for Practice 5, Using mathematics and computational thinking, Practice 7, Engaging in argument from evidence, and Practice 8, Obtaining, evaluating, and communicating information, are significantly lower than most of the others. Results of regression analyses suggest that the number of years the teacher participated in PD can predict student IONIC scores. The findings suggest that IONIC is a valid and reliable observation-based protocol, facilitating an understanding of how students can effectively engage in scientific practices in the classroom.