This full research paper investigates upper-secondary students programming skills. In 2021, an initial survey assessed the programming skills of first-year upper-secondary school students, focusing on their ability to solve problems using block and text-based programming languages. The survey revealed significant differences between students with mandatory programming education (Group TECH) and those without (Group SCI), with Group TECH outperforming Group SCI across all categories. The follow-up survey in 2024 evaluated the same cohort to measure longitudinal changes in their programming knowledge and skills. The results showed marked improvements in Group TECH's performance, particularly in more complex programming tasks involving loops and embedded conditional statements, while Group SCI also showed progress, albeit to a lesser extent. The study found that students who engaged more frequently in programming activities outside of school demonstrated higher average scores. Overall, students in the study entered upper-secondary mostly prepared for programming topics mostly related to math like variables and conditionals in block programming languages. Students that left upper-secondary, after introductory programming, were better prepared to understand loops and complex problems in text.

In the face of learning to program, students are often divided into two camps: those who excel and those who struggle. Through this challenge, some students manage to persevere from difficulty to ability. This study aims to identify the students who have made the most significant improvements in their programming skills, by leveraging various learning analytics and metrics. The field of learning analytics in programming has often sought to identify struggling students, utilizing metrics such as the error quotient (EQ) and time-on-task. This research distinguishes itself by taking a longitudinal approach to analyzing students' programming data collected over a nine-month period. By framing error quotient through operant conditioning and time-on-task through expectancy-value theory, we aim to root these learning analytics in established learning theory to validate their relevance. These preliminary results illustrate how we can identify students who have improved the most based on these metrics.

This dissertation research explores the potential of using learning analytics to improve programming education. The research goals include replicating previous research through studying heterogeneous groups of students at upper secondary schools over several months. The expected contribution of this dissertation is to provide insights into how learning analytics can identify struggling students.

This Innovate Practice work-in-progress paper presents findings around how distance learning, due to the the COVID-19 pandemic, affected students as measured by time-on-task in programming.In this qualitative study, we examine a group of 36 second year upper-secondary students in Programming 1 during a nine week period in Spring 2021. During this time, they alternated between one whole week of distance learning followed by two weeks of in school instruction. For the Programming 1 lessons, students used an online platform to write, edit and run code in. We analyzed the log data from the platform to estimate time-on-task for each student for every lesson both at home and at school.We observed that students were affected differently by distance learning as measured by time-on-task. 12 students had more average time-on-task at school. 15 students had more average time-on-task at home. Nine students had less than five minutes difference on average.In addition to the analysis of time-on-task, students were given a survey in Fall 2021 to follow up on their experiences with in-school teaching and distance learning. In the survey, students were asked questions about their study environment at home during distance learning. From the responses, 13 students described their study environment as “in bed” despite having access to a table and chair in a room for themselves and twenty-three students described their study environment as “playing video games during online lectures”. Not surprisingly, students that said they were playing video games during online lectures had a lower average time-on-task by about ten minutes than their peers. Interestingly, students that said they participated in class in bed had a higher average time-on-task by about ten minutes than their peers.Correlating responses from the survey and time-on-task data, we reason about how students’ study environments at home affected their time-on-task and how distance learning has affected students in the pandemic.

This research work-in-progress paper presents findings related to upper secondary students’ understanding of programming concepts such as: (1) variable assignment, (2) if-statements, and (3) loops in Python and Scratch. In 2017, the Swedish education board added computer science curriculum to compulsory schooling. Students now entering upper secondary school are expected to have experience with programming. To find out how familiar first year upper secondary students are with programming, we assessed 172 students’ programming knowledge with a multiple choice questionnaire with block programming questions in Scratch and text programming questions in Python. Each question required students to read a program between 4-8 lines of code and correctly identify the output of the program. The questions included basic programming concepts such as: variable assignment, if-statements and loops. Additionally, we asked students to self-report their prior experiences with programming in terms of how many hours they had spent programming inside and outside of school, as well as if they had more experience with block or text programming. As expected, the students’ scores correlated positively with the amount of hours they self-reported to have spent programming inside and outside of school. In conclusion, we correlate students’ self-reported programming experience with their scores on the block and text sections and reason about these results based on teachers’ experience and research literature.