2025 ASEE Annual Conference & Exposition

This is a Work-in-Progress paper. The semiconductor industry is experiencing rapid transformations driven by revolutionary developments in Artificial Intelligence (AI), Internet of Things (IoT), cybersecurity, and a renewed emphasis on sustainable technologies. While hardware engineers create the essential physical elements and systems that power these advancements, there is a deficit of hardware engineers. Many first-year computer science and engineering students are inclined to follow software career paths, frequently due to their minimal exposure to hardware issues and trends.

With this context in mind, our research began with the question of what can assist first-year engineering students in broadening their perspectives and discovering opportunities in hardware-related professions. To address this question, we leverage the Social Cognitive Career Theory (SCCT) (Lent et al., 2019) based on Bandura’s social cognitive theory (2001). Building on the foundations of SCCT's choice-making model (Lent et al., 2019), our team focuses specifically on how first-year engineering students make career choices and the impact of hands-on AI hardware experiences. Expanding on previous cohorts’ experiences (Authors, 2023; 2024), which focused on factors influencing outcome expectations, this work-in-progress paper presents an updated version of curriculum implementations aimed at improving students' self-efficacy beliefs and outcome expectations. We anticipate these improvements will increase students’ interest as well as help them set career goals related to their involvement in hands-on AI hardware activities.

During this 8-week module focused on AIoT applications, each student receives a custom-made AIoT learning board, which includes an ESP32 microcontroller, a breadboard, a battery, power management components, and various sensors. The module activities encourage students to collect data using the board's sensors. With guidance from the instructors and detailed exercise outlines, students are able to experiment with provided code that features machine learning models. They then adapt these codes to address problems and scenarios relevant to both their personal interests and broader societal issues.

For the Fall 2023 and Fall 2024 semesters, we have implemented the AIoT module during the second half of an elective course offered by the Electrical and Computer Engineering (ECE) department to first-year engineering students from different majors at a large public southeastern R1 institution. The course has no prerequisites, assuming participants have no prior knowledge or skills. Both cohorts have been taught by the same instructors. In these two cohorts, 55 first-year engineering students participated, of whom 34 students (16 from Fall 2023 and 18 from Fall 2024) have completed pre- and post-surveys. Through these surveys, the researchers measured the effects of hands-on AIoT experience on the differences in students’ self-efficacy, outcome expectations, and changes in their interests and career choices.

While there exists diversity in racial and gender identity, in the descriptive analysis of the pre-survey with a 5-point Likert scale, our team observed a noticeable discrepancy in self-efficacy among students (M=3.32, SD=0.91) compared to outcome expectations (M=3.78, SD=0.53) and interest (M=3.83, SD=0.69). By the time of this work-in-progress presentation, we will be able to share the instructional design of the AIoT module from the past two years’ implementations and interim findings from our quantitative and qualitative research surveys.