Generative artificial intelligence (GenAI) tools such as ChatGPT, Gemini, and Claude are transforming instructional design, yet most PK–12 educators lack effective strategies for engaging with them. To address this gap, we developed the CRAFT framework—a teacher-centered model for GenAI prompt design that emphasizes Context, Role, Audience, Format, and Tone. CRAFT translates principles from prompt engineering, instructional design, and Universal Design for Learning into a practical structure aligned with teachers' professional planning language. We implemented the framework in a professional development program with 94 PK–6 teachers who used GenAI to create lessons integrating computational thinking. Their 688 prompts and AI-generated responses were coded across 30 analytical features representing critique, launch behaviors, refinements, and supplemental material generation. Findings suggest that CRAFT helped teachers produce standards-aligned, differentiated, and instructionally coherent lessons while increasing confidence, creativity, and reflective engagement with GenAI tools.

As artificial intelligence (AI) becomes increasingly embedded in scientific and technical domains, the ability to engage in AI-integrated STEM problem-solving is emerging as a critical skill for the future STEM workforce. Supporting students in this type of problem-solving requires building a strong foundation in computational thinking, particularly through pedagogically effective and technically robust tools. In this paper, we propose augmenting i-Sail, a block-based programming environment designed for AI-integrated STEM problem-solving, with large language model-driven feedback capabilities to facilitate students' problem-solving while reinforcing key computational thinking skills for middle-grade students. We prompt a large language model with structured knowledge about breadth-first search to provide contextualized, adaptive feedback. The LLM helps students connect their problem-solving steps to the high-level structure of the breadth-first search algorithm and apply this understanding to pathfinding. We present a proof-of-concept evaluation that demonstrates the potential of the system to support the development of computational thinking through AI-integrated problem solving in diverse STEM contexts.

During the UnboxingCT project summer professional development, the Integrated CT Assessment was piloted with 72 elementary teachers. The assessment is based on computational thinking integration literature and asks teachers to identify different computational thinking concepts in content area scenarios. The assessment allowed us to identify which computational thinking concepts teachers were most familiar with prior to the professional development and assess changes in their understanding following the professional development. Our next step will be validation of the assessment with a larger group of teachers.

In this poster, we present our efforts to engage elementary teachers with learning trajectories as a tool for developing both their own and their students' comprehension of computational thinking (CT) and strategies for integrating CT learning in their classroom. Eleven teachers, who voluntarily joined a teacher professional development (PD) program to develop teacher leaders for CT integration in the elementary context, attended a one-day PD session aimed at reviewing their knowledge of CT, participating in CT-infused lessons, and engaging with CT learning trajectories. Over the next year, teachers will participate in monthly virtual PD to continue to grow both their CT content knowledge and pedagogical knowledge. Our goal is to develop these teachers as teacher leaders who will support others as they integrate CT. This poster will show our current progress on CT learning trajectories and teacher leaders' responses to the tool.

There is growing attention for developing professional learning experiences for content area teachers to infuse computational thinking (CT). However, there is little reporting on how teachers continue to implement the CT lessons once professional development (PD) is over. This study provides initial results on our efforts of building project sustainability through teacher autonomy in designing their own CT infusion projects or PDs for their schools. Our initial analysis indicates the need to continue to build teacher autonomy within the professional learning experiences for developing teacher confidence and sustainability of the project.

In this article, we leverage ideas from the theory of coevolutionary computation to analyze interactions of students with problems. We introduce the idea of informatively easy or hard concepts. Our approach is different from more traditional analyses of problem difficulty such as item analysis in the sense that we consider Pareto dominance relationships within the multidimensional structure of student–problem performance data rather than average performance measures. This method allows us to uncover not just the problems on which students are struggling but also the variety of difficulties different students face. Our approach is to apply methods from the Dimension Extraction Coevolutionary Algorithm to analyze problem-solving logs of students generated when they use an online software tutoring suite for introductory computer programming called problets . The results of our analysis not only have implications for how to scale up and improve adaptive tutoring software but also have the promise of contributing to the identification of common misconceptions held by students and thus, eventually, to the construction of a concept inventory for introductory programming.

In 2018 and 2019, Infusing Computing offered face-to-face summer PD workshops to support middle and high school teachers in integrating computational thinking into their classrooms through week-long summer PD workshops and academic-year support. Due to COVID-19, 151 teachers attended the Summer 2020 PD workshops in a week-long virtual conference format. In this paper, we describe Virtual Pivot: Infusing Computing, which employed emerging technology tools, pre-PD training, synchronous and asynchronous sessions, Snap! pair programming, live support, and live networking. Drawing on findings from participant interviews and post-PD surveys, we argue that three categories of changes (digital tools, formats, and supports for teacher engagement and collaboration) were effective in increasing participants' self-efficacy in teaching CT, supporting collaboration, and enabling participants to design CT-infused content-area lessons. We conclude by discussing how elements of this virtual PD can be replicated to increase teacher and student access to CT practices in middle and high school classrooms

Despite the increasing attention to infusing CT into middle and high school content area classrooms, there is a lack of information about the most effective practices and models to support teachers in their efforts to integrate disciplinary content and CT principles. To address this need, this paper proposes the Code, Connect and Create (3C) professional development (PD) model, which was designed to support middle and high school content area teachers in infusing computational thinking into their classrooms. To evaluate the model, we analyzed quantitative and qualitative data collected from Infusing Computing PD workshops designed for in-service science, math, English language arts, and social studies teachers located in two Southeastern states. Drawing on findings from our analysis of teacher-created learning segments, surveys, and interviews, we argue that the 3C professional development model supported shifts in teacher understandings of the role of computational thinking in content area classrooms, as well as their self-efficacy and beliefs regarding CT integration into disciplinary content. We conclude by offering implications for the use of this model to increase teacher and student access to computational thinking practices in middle and high school classrooms.

In summer 2018, we conducted two week-long professional development workshops for 116 middle and high school teachers interested in infusing computational thinking (CT) into their classrooms. Teachers learned to program in Snap!, connect CT to their disciplines, and create infused CT learning segments for their classes. This paper investigates the extent to which teachers were able to successfully infuse CT skills of pattern recognition, abstraction, decomposition, and algorithms into their learning products.

One way to increase access to education on computing is to integrate computational thinking (CT) into K12 disciplinary courses. However, this challenges teachers to both learn CT and decide how to best integrate CT into their classes. In this position paper, we present PRADA, an acronym for Pattern Recognition, Abstraction, Decomposition, and Algorithms, as a practical and understandable way of introducing the core ideas of CT to non-computing teachers. We piloted the PRADA model in two, separate, week-long professional development workshops designed for in-service middle and high school teachers and found that the PRADA model supported teachers in making connections between CT and their current course material. Initial findings, which emerged from the analysis of teacher-created learning materials, survey responses, and focus group interviews, indicate that the PRADA model supported core content teachers in successfully infusing CT into their existing curricula and increased their self-efficacy in CT integration.

This paper identifies design guidelines for the application of evolutionary techniques to the task of generating practice problems for learners in an Intelligent Tutoring System. To this end, we designed experiments that progressively incorporated an increasing number of the characteristics we expect to find in our target application. These features included noisy evaluations, overspecialization, and the need to mitigate user fatigue resulting from interactive evaluations of practice problems. As we did so, we evaluated the potential of recent breakthroughs in coevolutionary learning theory and identified the tradeoff specific to educational applications.

Computer Science Principles (CSP) will become an Advanced Placement course during the 2016-17 school year, and there is an immediate need to train new teachers to be leaders in computing classrooms. From 2012-2015, the Beauty and Joy of Computing team offered professional development (PD) to 133 teachers, resulting in 89 BJC CSP courses taught in high schools. Our data show that the PD improved teachers' confidence in our four core content categories and met its primary goal of training teachers in equitable, inquiry-based instruction. In this paper, we present the evolution of the BJC PD, its challenges and lessons that we learned while continually adapting to teachers' needs and contexts.

What are the concepts in introductory programming that are easy/hard for students? We propose to use Dimension Extraction algorithm (DECA) inspired by coevolution and co-optimization theory to answer this question. We propose and use the metrics of informatively easy/hard concepts to identify programming concepts that are solved correctly by the most "dominated student" versus solved incorrectly by the most "dominant student". As a proof of concept, we applied DECA to analyze the data collected by software tutors called problets used by introductory programming students in Spring 2014. We present the results, i.e., informatively easy/hard concepts on a dozen different topics covered in a typical introductory programming course. It is hoped that these results will inform programming instructors on the concepts they should (de)/emphasize in class. They will also contribute towards creating a concept inventory for introductory programming.

In this poster, we describe efforts to assess computational thinking activities that can be easily implemented in any science classroom. Studies have shown that a set of conditions must be met for computational thinking tools to be used in K-12 education and that when they are used, there is a wide spectrum in the level of computational thinking that the tool enables. This study extends this work by examining how middle school students translated their science fair projects into Scratch and what evidence of computational thinking is present. Scrape, a tool designed to analyze Scratch projects was used. Overall, it was found that most students simply created a presentation of their project without much complexity. Eight students created interactive projects that required user participation and used more advanced computational concepts. Finally, recommendations are given for next steps in the creation of a series of activities that would scaffold student learning as they apply to computational thinking concepts of a science concept.