Many institutions use undergraduate teaching assistants (tutors) in their computing courses to help provide more resources to students. Because of the role tutors play in students' learning experiences, recent work in computing education has begun to explore student-tutor interactions through the tutor's perspective and through direct observation of the interactions. The results suggest that these interactions are cognitively challenging for tutors and may not be as beneficial for students' learning as one might hope. Given that many of these interactions may be unproductive, this work seeks to understand how student expectations of these sessions might be impacting the interactions' effectiveness. We interviewed 15 students in a CS2 course to learn about the expectations and desires that students have when they attend tutoring sessions. Our findings indicate that there is variation in what students consider a desired result from the interaction, that assignment deadlines affect students' expectations and desires for interactions, and that students do not always want what they believe is beneficial for their learning. We discuss implications for instructors and potential guidance for students and tutors to make tutoring sessions more effective.

The academia-industry gap in undergraduate CS education has been a well recognized problem over the past two decades. Many papers have identified this gap or proposed novel solutions to parts of it. Yet, recent studies found the gap persists. This dissertation explores various possible reasons that could lie behind the persistence of this gap. We uncover that the majority of CS faculty support including more industry-relevant content in their courses. Faculty do however experience a skills, resources and awareness gap which forms a barrier to making their courses more relevant to industry practice. Moreover, we find that these barriers are experienced by CS faculty at top universities world-wide. Furthermore, many of the efforts to close the gap do not adequately address known barriers to adoption and barriers encountered by faculty. Additionally, one concern raised by some faculty is that students may not be learning the skills that we are teaching. These faculty are partially correct as we find evidence indicating that the average student proficiency with prerequisite materials may not be fully aligned with instructor expectations. Moreover, a different study indicates there are performance disparities between several subpopulations of students. However, we also find evidence that indicates that students on average continue to improve their understanding of CS fundamentals as they progress through their program, indicating that students on average do appear to learn the materials that we teach.

Introductory computer programming (i.e. CS1) is the entry point into the computer science major at higher education institutions worldwide. It introduces foundational concepts to students that are then built upon in future courses. Computer science as a whole has struggled to attract and retain students in the major, particularly women and students from minority backgrounds. Researchers have explored reasons as to why students are struggling to complete their CS1 course and remain in the major. Current research has uncovered a wide range of different factors which have been studied mostly in isolation from each other, thus not being able to provide a full view of the CS1 student experience. This dissertation presents an (1) exploration of the current state of CS1 research and its organization into an existing theoretical framework, (2) a holistic view of student experiences in early computing courses, and (3) the benefits and role of pedagogical best practices in supporting student learning in CS1.Regarding exploration: we surveyed the current state of CS1 literature to gain an understanding of what factors related to student CS1 success have been studied and mapped the findings to an existing student learning framework: Biggs' 3P model. We find the factors explored included areas such as internal student factors, instructional practices, the student's learning process, and many more. Regarding holistic: we collected student experience surveys from students of the first four courses students at UCSD stake during their computer science program. We found lower performing students report higher levels of stress across multiple areas compared to their higher performing peers. Regarding pedagogy: we analyzed 12 years of longitudinal student data to understand how the implementation of a trio of best practices (peer instruction, pair programming, and media computation) in introductory programming courses at UCSD demonstrated improvement to student performance and retention within the program. Regression models and significance testing provide evidence that the presence of these best practices were significant in these positive outcomes. Despite these improvements, certain groups such as women and minorities were still failing and being retained at lower rates in the major than their peers.

Recent work has identified a mismatch between instructor expectations of students' mastery of prerequisite course content and their actual ability. This invites the question of why this mismatch exists. We first examined grades in prerequisite courses and found they meaningfully correlated with performance on an assessment testing their knowledge of prerequisite material. In addition, we found neither taking alternatives to the primary identified prerequisites nor the delay between taking prerequisite courses and the follow-on course meaningfully impacts performance. Second, we confirmed that prerequisite course grades are significantly correlated with the grade in the follow-on course---confirming that the grades in the previous courses convey some information about student understanding of those topics. Perhaps surprisingly, we found that grades in courses outside computing were similarly correlated as those courses inside computing, suggesting that underlying factors such as general study skills may be as important as the domain-specific knowledge itself.

Instructors and institutions tend to treat students who have fulfilled their prerequisites for a course as if they are all equally well-prepared. Recent work in computing has shown that a substantial proportion of students are entering later courses with low prerequisite course proficiency, possibly hurting student outcomes. This dissertation explores (1) the extent of low prerequisite course proficiency throughout our CS curriculum, (2) the impacts of low proficiency on student outcomes, with a focus on students typically underrepresented in computing, and (3) possible causes leading to low proficiency as well as possible reasons why students are able to succeed in courses despite entering underprepared.

Regarding extent: we find that a substantial proportion of students, across multiple courses, enter courses with low prerequisite proficiency.

Regarding impacts: encouragingly, we find that student grades in prior courses are predictive of prerequisite proficiency coming into a course and student success in the course. We also find a relationship between students' incoming proficiency and their performance in the course. This suggests that the low prerequisite proficiency may be negatively impacting student outcomes. In terms of subpopulations, there is a disproportionate impact on students typically underrepresented in computing.

Regarding causes: in contrast with previous research, we find no harmful effects on prerequisite proficiency due to the passage of time or taking alternative prerequisites. In terms of instructor expectations, we find that instructors overestimate students’ incoming prerequisite proficiency. Furthermore, we find that many instructors believe prerequisites are not fulfilling their purpose. They cited various reasons including the number of conflicting pressures instructors face with respect to ensuring students learn material expected for the next course (e.g., trying not to adversely impact students’ time-to-degree, being well-liked by students, and balancing the time spent on teaching with other responsibilities).

As enrollments and class sizes in postsecondary institutions have increased, instructors have sought automated means to identify students who are at risk of failing a course. This identification must be performed early enough in the term to allow instructors to assist those students before they fall irreparably behind. In this sense, this dissertation proposes early identification methods of at-risk students and characterizes behaviors of such students.

The first part of this work describes a modeling methodology that predicts student final exam scores in the third week of the term by using different sets of easy-to-collect data including student clicker responses, prerequisite course grades, online quiz scores, and assignment grades. The work uses different machine learning techniques, trained on one term of a course, to predict outcomes in subsequent terms. This allowed making predictions across terms in a natural setting (different final exams, minor changes to course content). We applied this modeling technique to five different courses across the computer science curriculum, taught by three different instructors at two different institutions. The results show consistent performance across multiple courses in a curriculum, and across multiple institutions. Also, prerequisite course grades and clicker responses are more predictive than online quiz and assignment grades.

The second part of the work is to understand what factors contribute to those students' difficulties. If we were able to better understand the characteristics of such students, we may be better able to help those students. This work examines the characteristics of lower- and higher-performing students through interviews with students from an introductory computing class. We identify a number of relevant areas of student behavior including how they approach their exam studies, how they approach completing programming assignments, whether they sought help after identifying misunderstandings, how and from whom they sought help, and how they reflected on assignments after submitting them. Particular behaviors within each area are coded and differences between groups of students are identified.

This dissertation ends with stating some future work of automating the identification process for general public and crafting effective student intervention frameworks.