This study reports on the design, implementation, and impact of a remote professional development (PD) course for secondary school teachers who were transitioning to remote instruction during the COVID-19 pandemic. We designed technology innovations to strengthen the previously successful in-person course. The innovations support teachers to customize an instructional unit by setting and revising goals based on evidence from their students' prior work on the unit. A Curriculum Visualizer makes the pedagogy of the unit visible and guides planning for customization. Carefully curated small group activities using Zoom breakout rooms ensure that each teacher could share their thoughts, ideas, and impressions with other teachers. Participants were 23 science teachers from 12 different schools in a western U.S. state. We developed rubrics to code customization goals, plans, and moves using bottom-up methods and iterative refinement. Reflections on student work and use of the Curriculum Visualizer enabled teachers to set and refine customization goals and make evidence-based and pedagogy-aligned customization decisions that enhanced the interactive learning opportunities for their students. Our results reinforce the C-b model proposed by Sailer et al. (this issue) by illustrating the value of using technology to support collaborative, interactive PD activities.

With the widespread adoption of the Next Generation Science Standards (NGSS), science teachers and online learning environments face the challenge of evaluating students' integration of different dimensions of science learning. Recent advances in representation learning in natural language processing have proven effective across many natural language processing tasks, but a rigorous evaluation of the relative merits of these methods for scoring complex constructed response formative assessments has not previously been carried out. We present a detailed empirical investigation of feature-based, recurrent neural network, and pre-trained transformer models on scoring content in real-world formative assessment data. We demonstrate that recent neural methods can rival or exceed the performance of feature-based methods. We also provide evidence that different classes of neural models take advantage of different learning cues, and pre-trained transformer models may be more robust to spurious, dataset-specific learning cues, better reflecting scoring rubrics.

Prompted by the sudden shift to remote instruction in March 2020 brought on by the COVID-19 pandemic, teachers explored online resources to support their students learning from home. We report on how twelve teachers identified and creatively leveraged open educational resources (OERs) and practices to facilitate self-directed science learning. Based on interviews and logged data, we illustrate how teachers' use of OER starkly differed from the typical uses of technology for transmitting information or increasing productivity. These experiences provide insights into ways teachers and professional developers can take advantage of OER to promote self-directed learning when in-person instruction resumes.