This research paper describes the mathematics teacher knowledge transformation when computer science concepts are integrated with mathematics. During the last decade, schools and educators made a significant effort to introduce Computer Science (CS) and Computational Thinking (CT) in K-12. While this effort produced some positive results in schools where resources and specialized educators are available, there are school districts where creating new classes and/or hiring specialized teachers is not a sustainable option. This creates a need for teachers that are specialized in one subject area to expand their knowledge into CS and CT, and teach these concepts in their classes. This knowledge expansion and integration should result in creating integrated curricula using their subject matter knowledge and the newly acquired CS knowledge. This implies that the teacher should be able to identify the intersection of concepts between the two fields, and leverage their difference and similarities to promote deep learning for their students. The current study explores how the teacher candidates' technological, pedagogical content knowledge (TPACK) in mathematics is affected when mathematical concepts are integrated with CS and CT. This study is a three-day intervention in a graduate level secondary mathematics education methods course, where utilizing Math + CS integrated modules we infuse CS concepts into mathematics. The already known mathematics context reduced the cognitive load for familiarizing the teacher candidates with CS, and the integration of the CS concepts provided a different perspective for thinking about mathematics and teaching. After having the teacher candidates reflecting on the concepts of Variable and Equal Sign (=), our results showed that they started noticing the differences and nuances between the two fields, developing this way their interdisciplinary Math + CS TPACK (I-TPACK) knowledge. Furthermore, the study showed that representing multiplications using repetition structures illuminated the internal mechanic of multiplication and its properties. Finally, the experience with the strict grammar and syntax rules of programming illuminated the areas in the teacher candidates mathematics instruction that were ambiguous in the way they teach variables, equations, percentages, and multiplication.

We describe a new introductory CS curriculum initiative that uses analogies and active engagement to develop students' conceptual understanding before applying the concepts to programming. We believe that traditional coding approaches to introducing computer science concepts rely on students to build their own conceptual understanding, rather than grounding their understanding of concepts in what they know from everyday experiences. Using constructivism as a foundation for this curriculum initiative, our approach builds a framework for student understanding anchored in the physical world using simple games and stories to stimulate mental engagement through embodied learning.

While governments, teacher educators, and teachers recognize the importance of introducing computer science (CS) in K-12, there is limited understanding about when the teacher preparation should start and how it needs to be done. Furthermore, with the rapid expansion of digital technology in our society and the way it affects our lives, new educational frameworks for better teacher preparation have been developed, such as TPACK and CTIntegration, that situate new technologies and their role in education to meet the new societal demands. There is a larger body of research in Mathematics on how on how to prepare teachers to teach than in CS. This study explores how graduate teaching candidates and undergraduate pre-service teachers engage with Math+CS modules differently. We were interested in the similarities and differences in the way senior-level pre-service teachers in a mathematics and technology class and graduate-level teacher candidates in a methods class interact with Math+CS integrated modules. Working with the concepts of variable, input, and output, we investigated which knowledge domains were leveraged among groups in different classes, what connections they make between the concepts in math and CS, and their uncertainty. The results indicated that there was a difference in the way the teaching candidates in their methods class approached the computational modules than the pre-service teachers. The pre-service teachers focused more on developing their CS content knowledge, and the teacher candidates were interested in making connections between their knowledge domains, reflecting on how their students would think about the concepts in math, expressing more concerns, and asking more questions to develop their TPACK.

Despite many initiatives to increase participation in K-12 computer science (CS), only about half of the public high schools offer CS, and only about a third of the K-8 public schools offer CS. To make matters worse, even if schools offer CS, interests wane from late elementary through post-secondary education. The lack of participation has been attributed to feelings of not belonging, technology-rich programs creating divides among students, and negative belief systems that CS is socially isolating, lacks creativity or fun, and is for intelligent, white males, and we believe one contributing factor is the way CS is introduced, taught, and scaffolded. In this full paper, we present innovative pedagogical approaches to teach fundamental CS concepts, such as abstraction, representation, algorithms, and computation, to 6th grade students using manipulatives, which are physical objects that students interact with to teach or reinforce a concept. Teaching and learning using manipulatives has a long history in science and mathematics education, but the development of and research on manipulatives to teach CS concepts is less common. Through observational field notes from a 6th-grade classroom and interviews with the teacher, we discuss the affordances and drawbacks of the different approaches and manipulatives. We found that using manipulatives led to increased student engagement and participation with the material and made teaching the material more exciting and engaging for the teacher. In addition, we found that the manipulatives provided a way for student misunderstandings and errors to be more apparent through tinkering with the physical objects, and the teacher was able to easily understand how to extend the activities and find ways to connect multiple CS concepts. However, we observed several drawbacks with different manipulatives and approaches for using the manipulatives that could be helpful for future changes and the development of new manipulatives. For example, the puzzle-like games we gave students to construct algorithms for tossing a coin to see who goes first in a game and playing Rock, Paper, Scissors were challenging for students to recreate a given algorithm that was not necessarily an algorithm they would have designed themselves.

In the ever-evolving field of computer science (CS) education, the significance of K-12 teachers and their backgrounds have often been overshadowed by the predominant focus on students. Teachers in K-12 often lack the necessary expertise in CS and have limited support provided by existing CS-based curricula. While research on CS education effectiveness grows, limited attention has been given to the factors influencing teachers' intention to use or adoption of CS-based curricula. Prior research revealed that despite professional development and positive reactions to teaching innovations, curriculum adoption cannot be guaranteed. However, existing studies have concentrated on university-level instructors who possess extensive CS knowledge and a strong passion for teaching it. In this research study, we apply the Technology Acceptance Model (TAM), which factors perceived ease of use, perceived usefulness, and attitude into the intention to use technology, as a framework for understanding and predicting the intention to use a CS-based curriculum among secondary mathematics graduate teacher candidates in their last quarter of student teaching. Our findings highlight the TAM as a valuable tool for evaluating teachers' attitude toward and intention to use CS-based curricula, enabling informed decision-making for future curriculum developers. Teacher candidates who expressed a negative intention to use expressed concerns regarding Teacher Knowledge, Student Understanding, and Student Resources, while those who intended to use emphasized their appreciation for Lesson Plan Content and Lesson Plan Quality. It was interesting that prominent themes in perceived ease of use, which were Lesson Plan Content and Lesson Plan Quality, differed from the most frequent themes that emerged in perceived usefulness, namely Student Understanding, Student Engagement, and Pedagogical Alternative. This contrast highlights the importance of considering multiple factors and perspectives when designing and assessing CS-based curricula, which can help curriculum developers create more inclusive CS-based curricula that address and support the diverse backgrounds of K-12 teachers.

In this research to practice report, we present student perceptions of a new first-year engineering programming class that was designed by informed research practices. While the College of Engineering at the participating university saw a lot of major switching in the first year, there were not many students switching into computer science (CS). This could have been because other engineering major classes did not transfer well to CS or that students would be a year behind in the programming classes. In either case, the participating university felt that computational thinking and programming were an important part of first-year engineering exploration and believed that all 21st century engineering majors should learn to program in a general purpose language, such as Python or C++. Past research found that prior programming experience, knowledge organization, self-efficacy, and class size were factors impacting student performance in early programming courses, and learning about programming enhanced students' problem-solving skills and improved perceptions about CS. Many other research studies showed that the context in which programming material was presented to non-majors was important for their learning and interest, and themed, mixed-context courses with real-world engineering problems worked well. Therefore, in the spring 2022, we created 12, 100-person sections of a course titled “Engineering Computational and Algorithmic Thinking” for a new first-year engineering experience. Each section was taught by a different instructor representing a variety of disciplines and topics in the College of Engineering. This paper describes the research to practice and engineering student perceptions of the new, required programming class. We hypothesized that engineering students would say that the course should be required for all engineering majors, but we wanted to know why. We believed that student prior programming experience, engineering major, and other demographics would play a role in their responses. We thematically analyzed student justifications to identify emerging themes, and we used a mixed-method approach to correlate themes to different student demographics, engineering major interest, and prior programming experience. The majority of engineering students agreed that the first-year engineering course should be required, and prior programming experience and major played the largest role.

The understanding of Discipline-Specific Language is an important competency for students of any field to begin mastering early in their studies, since it serves as a prerequisite for both the analysis of expert text and precise communication. Therefore, an introductory curriculum should pay careful attention to how it incorporates, defines, and uses Discipline-Specific Language. While close examination of the language used in instruction has been studied in applied linguistics for more than a decade, this idea has not yet been extensively applied to the language used in Computer Science (CS) instruction. A handful of authors in CS literature have suggested that Discipline-Specific Language be examined in CS education or software engineering education, but little published work has been done to justify the value of further research in this area. In this exploratory research paper, we examine how a well-known Advanced Placement (AP) CS curriculum uses CS-specific language. We posed the following questions to guide our research: 1)Is discipline-specific language identified and defined in the example curricula? 2)Is there temporal contiguity between: a)Occurrences of the same CS-specific word? b)The first occurrence of a CS-specific word and it's definition? We look at what words are used, when they are introduced, how they are defined, and the consistency of their application. To establish an initial set of CS-specific words, we analyzed the free-response questions of past AP CS-A exams and the official AP CS-A course and exam description document. We then recorded every location of each of these words within the curriculum so that we could examine the context around their usage. While we found that technical CS-specific words were regularly defined and applied consistently, common (everyday) words with CS-specific context and connotations were either absent, ill-defined, or not applied consistently. Our results suggest that the example AP CS-A curriculum could improve its introduction of Discipline-Specific Language, and we believe that this preliminary review of a well-known AP CS-A curriculum supports further investigation of other CS curricula and the impacts language has on students' learning and academic success.

We describe a new introductory CS curriculum for middle schools that focuses on teaching CS concepts using the instructions and rules for playing simple, physical games. We deliberately avoid the use of technology and, in particular, programming, and we focus on games, such as tossing a coin to see who goes first and playing Tic-Tac-Toe. We report on middle-school students' understanding of basic CS concepts and their experiences with the curriculum.

This Research to Practice Full Paper presents the findings of our research about the comprehensiveness of continuous improvement plans and initiatives at academic computing programs. Continuous improvement (CI) plans consist of 8 components: (1) Administration, (2) Curriculum, (3) Course, (4) Faculty, (5) Research, (6) Academic Advising, (7) Facilities, and (8) Support Staff. To implement a truly comprehensive CI plan, an academic computing program must address all 8 CI components by identifying the types of data used for improving each one. For each component, this data can be produced internally (by the CI component itself) or externally (by other CI components). The degree to which two components exchange CI data determines the quality of their CI integration.In this paper, we detail a comprehensive list of data types used for each component’s continuous improvement. We also propose a quantifiable model (360-CI) to "measure" the comprehensiveness of an academic computing program’s continuous improvement plan or initiative. This model is represented by the Academic Computing Continuous Improvement Scoring Survey (ACCISS). ACCISS uses the list of continuous improvement data types to reveal the ones produced and utilized within an academic computing program. ACCISS also aims at identifying potential new data types that can be added to the comprehensive list. The results of the questionnaire can then be mapped to the 360-CI model to generate a comprehensiveness score out of 360.The questionnaire tool was administered at a large electrical engineering and computer science program. A high response rate provided us with confidence in the results of the generated score. The paper details the process of the questionnaire implementation, the results (including the comprehensiveness score), and the lessons to be learned when the questionnaire is administered in the future. The results also help this specific program identify areas for improvement in their continuous improvement process.

In this work (a full paper on an innovative practice), we report on middle school students’ experiences while learning a new text-based, functional domain-specific teaching language for programming well-known, simple physical games, such as tossing a coin to see who goes first or playing Tic-Tac-Toe. Based on students’ responses after taking an 18-week, 7th grade elective, we find that the majority of the students like learning the new language because it is not block-based, it is not complicated, and it is in the domain of games. However, we also find that there are some students who say programming is what they like the least about the class, and the majority of the students report that they struggle the most with writing the syntax. Overall, the majority of students like the curriculum, language, and using games as a way to explain CS concepts and teach programming. Even though learning a text-based, functional programming language may be difficult for middle-school students, these results show that the domain-specific teaching language is an effective teaching vehicle at the middle school level.

This full research paper explores two factors of increasing importance for first-year university engineering curricula: sustainability and diversity. Over the past fifteen years, many universities in the United States have adjusted their engineering programs in response to these two values expressed by industry, professional organizations, and the Accreditation Board for Engineering and Technology. This research study addresses these issues by integrating environmentally-responsible engineering (ERE) concepts into a series of courses in the pilot of a new first-year engineering experience.We surveyed students at the end of the experience to determine the impacts of integrating ERE into the course sequence on students’ enthusiasm for engineering, interest in ERE and ERE professions, concern for the environment, and sense of belonging. We found that students reported much greater satisfaction and positive impacts than we expected. Student responses suggest that incoming first-year engineering students are concerned about environmental issues and feel both responsible and inspired to use engineering skills to design solutions for problems impacting the environment and society. These results provide motivation to pursue further research on the effects of integrating sustainability concepts into engineering curricula on students’ motivation and consequential retention, especially among underrepresented students.

University students learning about computer science (CS) can be intimidated and frustrated by programming, and to make matters worse, the general-purpose programming languages chosen for introducing students to programming contain too many features that have the potential to overwhelm and distract students. We hypothesize that by using a delayed-coding approach with a language designed for teaching a smaller set of features focused on the fundamental CS concepts, such as types, values, conditions, control structures, and functions, student retention and success would improve, especially for those with no or little prior programming experience. To test this hypothesis, we split a college computer science orientation class into two sections. One section began programming with a general-purpose language, Python, during week 1. The second section used a new, functional domain-specific teaching language themed around programming simple, well-known physical games. A group of researchers designed the new language with the purpose of giving students a more focused approach to learning basic computer science concepts and emphasizing good programming practices early, such as working with user-defined types and decomposition. Based on student survey responses before and after the two sections and their grades through the two subsequent CS courses, we find that students in the delayed-coding section using the new language had lower engagement in their class. In addition, we find no evidence of a higher pass rate for students from this section in their subsequent computer science courses.

There is wide-spread agreement that K-12 students need opportunities to explore computer science (CS) concepts and computational thinking within a wide array of disciplines for advancing, broadening, and diversifying the participation in CS. Programs such as "Computer Science for All" were created by the US government to motivate and help students of all ages to engage with CS, which was described as the "new basic skill for economic opportunity and social mobility". However, what is less understood is how to prepare teachers to engage with CS concepts and computational thinking, expanding their specialized and pedagogical content knowledge on these concepts. This study explores this gap in the context of a graduate-level secondary math education methods course in a university environment. To reduce the cognitive load for the teacher candidates as well as their students, we utilize the secondary mathematics curriculum to explore CS concepts infused into mathematics, avoiding at the same time an increase to the teaching hours to the extent that it is detrimental to the existing K-12 curriculum. Our study uses hybrid block-text programming-based teaching modules specifically designed to expose the similarities and the differences between mathematics and CS. We utilize the concept of variable and operations around it to explore how the teacher candidates' conceptions and misconceptions about CS make the understanding of those concepts easy or challenging, and how that affects their ability to incorporate them into their teaching.

This Research Full Paper presents a study on the accuracy of computer science (CS) novices' mental models about linked lists in the C programming language. In CS, learning abstract fundamental concepts that require students to understand memory management can be very difficult and lead to misunderstandings that carry on into the advanced topics. This is especially true for linked lists data structures because they serve as a bridge to understanding more advanced data structures. Therefore, it is important to understand how students think about linked lists for improving teaching and learning. Exploratory research on mental models in CS is not as well-known as in other disciplines, such as Psychology, Education, Chemistry, Physics, and Mathematics. Since CS is based on abstract concepts like in Chemistry, Physics, and Mathematics, we believe the CS education community can benefit from more research on mental models and student reasoning, especially in fundamental areas such as data structures and algorithms. Hence, we conducted 2-hour semi-structured, think-aloud interviews with 11 undergraduate students to uncover the accuracy of their mental models, including their conceptual and procedural understanding, about singly linked lists. Our results suggest that none of the participants have an accurate mental model of a singly linked list, after learning about them and implementing them in their data structures course. Students struggle with expressing their conceptual understanding in their verbal responses to interview questions, while their scores for the coding questions are much better. Overall, the majority of students have a good procedural understanding of how to use the components of a linked list and implement the operations on a singly linked list, but most students cannot express their conceptual understanding of the components in a singly linked list. The results from this research suggest that educators need to spend more time covering linked lists in their data structures courses and make sure that students understand how their prerequisite knowledge of pointers and memory management integrate with the new knowledge about linked lists.

In today's competitive academic environment, academic computing programs must continuously improve, and for accredited programs, establishing and documenting a continuous improvement (CI) plan is a main requirement for accreditation. While many academic computing programs strive to implement a comprehensive CI plan that addresses all angles of the process, which we call 360-CI, they rarely do. One of the reasons for this deficiency is the ambiguity of what comprehensive CI (or 360-CI) is. From the literature, we identify 8 components of CI that should be addressed in every academic computing program's CI plan. These components include Administration, Curriculum, Course, Faculty, Research, Academic Advising, Facilities, and Support Staff. Each CI component is not addressed equally in the literature. The most emphasis is on Curriculum, Course, and Faculty, while the others receive much less attention. In this paper, we introduce an “ideal” 360-CI model utilizing all 8 CI components, and we use the 360-CI model to develop a method for scoring the comprehensiveness of an academic computing program's CI plan. To evaluate this method, we conducted a series of 21 semi-structured interviews and followup questionnaires with administrators and faculty in a large electrical engineering and computer science program. The results are consistent with the literature showing the most emphasis on Curriculum, Course, and Faculty CI and the least emphasis on Advising, Facilities, and Support Staff CI. Based on the results from this research, we propose potential approaches to help academic computing programs establish and maintain a CI plan that maximizes their CI score.

Continuous Improvement (CI) of academic computing programs is a main requirement of accreditation. Academic computing programs must have a CI plan that is documented in order to be granted accreditation. A comprehensive CI plan consists of 8 components: Course, Curriculum, Administration, Faculty, Research, Advising, Facilities, & Support Staff. In order for academic computing programs to achieve a truly comprehensive CI, these CI components must feed into themselves, as well as feed into each other. In this paper, we examine how two large ABET-accredited Electrical & Computer Engineering (ECE) and Computer Science (CS) programs handle these 8 components in their CI process by conducting 21 semi-structured interviews and following up with an online questionnaire. We use a coding scheme to analyze the results from the interview participants and find that all 8 CI components are recognized across both programs. However, the degree of integration between these components varies greatly with some components being highly coupled and others not being integrated at all. We present the types of data collected and used by the different ECE and CS programs, and we compare these results with those from the academic computing literature. From our analysis, we find that the CI focus is consistently on Curriculum, Course, Faculty, and, to a lesser extent, Administration, while Facilities, Advising, and Support Staff are barely recognized as part of the CI plan.

In computer science, learning abstract fundamental concepts requiring students to understand memory management can be very difficult and lead to misunderstandings that carryon into advanced topics. This is especially true in data structures with abstract data types. Understanding how novice students think and reason about data structures is important for improving teaching and learning in computer science. Most studies focus on student misunderstanding of advanced algorithms and data structures related to topics such as heaps, binary search trees, hash tables, dynamic programming, and recursion. Whereas, fewer studies focus on more elementary data structures, such as arrays and linked lists. Since linked lists serve as a bridge to understanding more advanced data structures, we believe that it is critical to identify students' conceptual and procedural misunderstandings earlier rather later. Therefore, directly after learning about linked lists using the C language, we conduct semi-structured, think-aloud interviews with 11 students to uncover their reasoning and misunderstandings about singly linked lists in C. Using rubrics to code responses to interview questions, we reveal students' confusion around the node containing a node pointer, failure to define a node structure, lack of knowledge regarding typecasting malloc, and lack of attention to the importance of NULL.

In this poster, we report our experiences with implementing two virtual computer science camps for middle school children. The camps use a two-part curriculum: One designed for 6th grade students using computer science concepts from familiar unplugged games, and the other targeted at 7th grade students using a domain-specific teaching language designed for programming board games. We use the camps to pilot the curricular material and provide teachers with practical training to deliver the curriculum virtually. Due to the teachers' commitment to finding successful strategies for delivering the curriculum online, the unplugged and programming activities worked surprisingly well in the remote environment. Overall, we found the use of well-known, unplugged games to be effective in preparing students to think algorithmically, and students were able to successfully code small programs for simple games in a new, functional, text-based language.

This is an Innovate Practice Full Paper. Creating modern systems with tightly integrated sub-systems relies heavily on multidisciplinary cooperation. Establishing an early value for other areas of expertise is crucial to encouraging later collaborations. This paper discusses a new first-term orientation course in Electrical and Computer Engineering at Oregon State University that is primarily taken by students pursuing a degree in Electrical and Computer Engineering. This course emphasizes collaboration and the importance of a multidisciplinary approach in engineering. The 10-week course contains engineering skills, tools, and concepts from three disciplines of engineering: Computer Science, Electrical & Computer Engineering, and Mechanical Engineering. Materials focus on how each discipline supports the other disciplines in modern engineering. Students attend one, one-hour lecture and two, two-hour labs per week. The lecture covers general engineering concepts including engineering process, design, and tools/skills. Student engineers work in teams of three, with each student focusing on a different engineering discipline. Teams design, implement, and improve on a competitive robotic project. Assessment using student surveys shows a 13% higher self-reported ability to recognize and avoid personal and engineering discipline bias from students enrolled in the new course when compared to students in the traditional course. This higher assessment value has a confidence interval of p=.005 using a two-tail T-test. Using student surveys and assignments, additional preliminary improvements were seen in student's ability to identify the need of another engineer/engineering profession to complete a portion of a project and understanding of other majors.

Academic co-curricular activities (e.g., programming contests, hackathons, student clubs, tutoring, internships, undergrad research) are popular with students, promote academic engagement and retention, and provide a competitive advantage to students applying for jobs and grad schools. This panel will continue a conversation started at a 2019 BOF session on this topic.

Computing Education Research (CER) implicitly assumes that CS undergraduate students have no barrier to access a learning platform or software package. The assumption that "everyone has access to and uses a device" endangers the validity of CER studies by overlooking a critical element in what students access and how students use computing resources. First, in this work, we explicitly investigate undergraduate student usage (how often) of computing resources (laptops and computer labs) on a university campus. Furthermore, we investigate whether CS student usage of laptops and computer labs are factors of success in computing education to close a crucial feedback loop for CER and CS educators. Second, previous studies studying student's technology equipped used qualitative surveys, and lacked a systematic and continuous view of the student population. In this work, we address this shortcoming by developing a method to use operational data sets from the wireless networks and computer labs on campus. Operational data sets provide a systematic and continuous coverage of all students, that is, a student's absence of usage becomes a data point instead of a missing point. Results indicate that the use of equipment levels may be lower than national average. Nevertheless, there exists a positive correlation between higher frequency of laptop usage and success in computing education.

In this research study, we investigate the impact of using the Story Programming approach to teach CS concepts on student performance in a subsequent C++ class. In particular, we compare how students receiving little or no coding to learn and apply these concepts perform in comparison to students who learn these concepts only in the context of coding. While past research has shown that exposure to programming is not a predictor of success in such courses, these studies are based on a 15-week versus 10-week course and do not control for the CS concepts and programming to which the students have been exposed. Consequently, we hypothesize that students from the Story Programming approach will perform worse in the following C++ class. Surprisingly, we find that this is not true: Students from the Story Programming approach with little to no coding do not significantly differ from their peers receiving a traditional code-focused approach.

Academic co-curricular activities (e.g., programming contests, hackathons, student ACM clubs, tutoring, internships, undergrad research) are popular with students, may promote academic engagement, and can give a leg up to students applying for jobs and grad schools. Yet information about co-curricular activities in departments and schools can be hard to come by. This BOF will provide participants with a forum for comparing notes: What co-curricular opportunities exist in your department? Does your department or school explicitly support or promote undergraduate co-curricular activities? If so, how (e.g., staff positions, faculty release time, student leadership - volunteer or for pay or credit) What have you learned from your experiences with co-curricular activities? Are co-curricular activities a good investment of department resources?

Story Programming is an approach for teaching complex computational and algorithmic thinking skills using simple stories anyone can relate to. One could learn these skills independent of a computer or with the use of a computer as a tool to interact with the computation in the tale. This research study examines the use of Story Programming before teaching coding in a computer science orientation course to determine if it is a viable alternative to the code-focused way of teaching the class in the past. We measure the viability of the Story Programming approach by evaluating student-success and learning outcomes, as well as student reactions to post-survey questions.

Code-first approaches for introducing students to CS exclude those without preparatory privilege in programming and those intimidated by coding. Delaying coding or not using coding in an introductory CS course provides an equitable learning opportunity and includes a broader group of students in computational education. We present a study that compares a traditional Python code-first approach with an approach to delay or remove coding by first using simple, well-known stories to explain computation without the need for a computer or coding. We find that many students, especially female students and those without prior programming, are initially not interested in coding but in using stories to explain computing. We conclude that a traditional Python code-first approach excludes these students and an option using stories is a viable alternative.

This study examines the results of establishing a rubric for design in freshmen computer science courses. The rubric consists of six categories (Understanding the Problem, Relationship Among Parts, Logic, Diagrams, Code Present, Testing) on a binary scale indicating if the category was included or not. The analysis examines four terms of data, one course per term, where students are required to submit a program design a week before they submit their assignment code. The terms differ in population (traditional and non-traditional), content (CS 1 and CS 2) and the level of design guidance (increased guidance/details in the syllabus). Each qualitative design was evaluated using the same rubric to facilitate comparisons between different courses, correlation to assignment grades and overall course grades, there were notable differences between what the populations included in their designs and how frequently the populations included the various categories of the rubric over time. In particular, non-traditional students tend to include more diagrams, logic and relationship among parts than traditional students, and in a CS 1 class without object oriented programming (OOP) versus a CS 2 class with OOP, CS 2 students tend to provide a more global picture of the program than specific details outlined in CS 1 designs, which lacked a big-picture perspective. These insights allow us to better understand how students from varying populations and different course content approach problem solving and design in computer science differently.