Search for: All records

This report examines the similarities and differences between K-12 CS standards across seven international locations. This analysis includes background information on each of the locations included in the study, trends in content, and comparisons by topic area. 

Problem. With many teachers in the United States just starting to learn how to teach computer science (CS), many do not have others nearby with CS teaching experience to provide support on CS practices and concepts. To address this gap, we piloted a one-year remote coaching program designed to provide that missing individualized support to teachers. Research Question. Our research question for this project was: How does teachers’ ability to apply CS practices and knowledge of CS concepts change after the coaching process? Methodology. Our mixed-methods study leveraged three primary forms of data from teachers who were coached (coachees) and teachers providing coaching (coaches): pre- and post-surveys, coaching logs, and self-reflection checklists. Findings. CS coachees’ reported CS knowledge and skills and their ability to apply CS practices related to Standard 1 were significantly higher after the coaching intervention. Implications. As more teachers continue to learn how to teach CS and hone their skills and practices, engaging these teachers in coaching can be powerful in improving their student’s learning. Professional development providers and regional education agencies (districts and schools) could leverage the processes formed through this intervention (which is based on professional development practices with solid evidence for positive impacts) to provide similar coaching to teachers just learning how to teach CS. 

Introduction: Because developing integrated computer science (CS) curriculum is a resource-intensive process, there is interest in leveraging the capabilities of AI tools, including large language models (LLMs), to streamline this task. However, given the novelty of LLMs, little is known about their ability to generate appropriate curriculum content. Research Question: How do current LLMs perform on the task of creating appropriate learning activities for integrated computer science education? Methods: We tested two LLMs (Claude 3.5 Sonnet and ChatGPT 4-o) by providing them with a subset of national learning standards for both CS and language arts and asking them to generate a high-level description of learning activities that met standards for both disciplines. Four humans rated the LLM output – using an aggregate rating approach – in terms of (1) whether it met the CS learning standard, (2) whether it met the language arts learning standard, (3) whether it was equitable, and (4) its overall quality. Results: For Claude AI, 52% of the activities met language arts standards, 64% met CS standards, and the average quality rating was middling. For ChatGPT, 75% of the activities met language arts standards, 63% met CS standards, and the average quality rating was low. Virtually all activities from both LLMs were rated as neither actively promoting nor inhibiting equitable instruction. Discussion: Our results suggest that LLMs are not (yet) able to create appropriate learning activities from learning standards. The activities were generally not usable by classroom teachers without further elaboration and/or modification. There were also grammatical errors in the output, something not common with LLM-produced text. Further, standards in one or both disciplines were often not addressed, and the quality of the activities was often low. We conclude with recommendations for the use of LLMs in curriculum development in light of these findings. 

Although there is extensive research on what makes teacher computer science (CS) professional development (PD) effective, little attention has been given to how PD providers prefer to collect and report data. A platform that met the needs of teachers while also simultaneously meeting the needs of funding bodies could be powerful in answering questions about participation and experiences in CS PD. Our research question for this study was: Which features and types of data do teachers find most important to include in a platform designed to record data related to their engagement with PD?We used an exploratory-sequential mixed methods approach that included focus groups and a survey created from an analysis of the data from the focus groups. The three most desired feature included adding information about the subject/topics targeted by the CS PD offerings for the CS PD they took, the grade levels targeted by the CS PD offering, and the number of training hours, points, and/or CEU credits earned or available to be earned by each CS PD offering taken. The three least desired features included have the capability for teachers to take notes about a CS PD, QR codes for signing up for CS PD, and capability to enter data about non-CS PDs you have completed. While we will use this data to inform the development of the platform, this study is significant as states can leverage this knowledge as they create their own systems for creating platforms for teachers within their own states. 

What does it mean to conduct computer science education research in a manner that ensures that the evidence produced is high quality and benefits a wide variety of students? One can pour over various guides from institutions like What Works Clearinghouse (WWC) and the American Psychology Association (APA). However, what many standards fail to include is a holistic perspective of conducting education research, including guidelines for ensuring that the aggregated data presented represents the student population that the research will ultimately serve. In this panel, we tackle both and explore approaches that have been used in other education research fields as well as those appropriate to CS education research that can be leveraged to ensure that all students' needs, experiences, cultures, identities, and voices are captured and presented in our research. 

Problem. Extant measures of students’ cybersecurity self-efficacy lack sufficient evidence of validity based on internal structure. Such evidence of validity is needed to enhance confidence in conclusions drawn from use of self-efficacy measures in the cybersecurity domain. Research Question. To address this identified problem, we sought to answer our research question: What is the underlying factor structure of a new self-efficacy for Information Security measure? Method. We leveraged exploratory factor analysis (EFA) to deter- mine the number of factors underlying a new measure of student self-efficacy to conduct information security. This measure was created to align with the five elements of the information security section of the K-12 Cybersecurity Education framework. Participants were 190 undergraduate students recruited from computer science courses across the U.S. Findings. Results from the EFA indicated that a four-factor solution best fit the data while maximizing interpretability of the factors. The internal reliability of the measure was quite strong (𝛼 = .99). Implications. The psychometric quality of this measure was demonstrated, and thus evidence of validity based on internal structure has been established. Future work will conduct a confirmatory factor analysis (CFA) and assess measurement invariance across sub- groups of interest (e.g., over- vs. under-represented race/ethnicity groups, gender). 

This report examines the similarities and differences between state-adopted K-12 CS standards and the 2017 CSTA K-12 Standards. This analysis includes basic information about the standards (such as counts by state and level) and their cognitive complexity, as well as more detailed information about their relationship to the CSTA standards. 

In recent years, eight states have adopted a graduation requirement in computer science (CS), and other states are considering similar requirements. Due to the recency of these requirements, little is known about student and teacher perceptions of course(s) that fulfill the requirement and their content. This project seeks to answer the question, What are the perceptions of students who are studying CS beyond high school and CS teachers of a high school CS requirement and its content? We used a mixed methods approach that included interview transcripts from students who took CS coursework in high school and are currently studying it in college (n = 9). We also used quantitative data from a survey of CS teachers (n = 2, 238) that asked for their perceptions of a CS graduation requirement. Most of the students felt that CS should be required in high school, and there was a wide variety of sentiment regarding what content should be included in such a course. For the high school teachers, about 85% felt that CS should be required. It is perhaps not surprising that most students who studied CS in college valued it at the high school level and thus supported a graduation requirement. What is more interesting is the diversity of content that they felt should belong in such a course. These findings serve as an important consideration for those implementing a CS graduation requirement.