Date of Award
5-2024
Document Type
Dissertation
Degree Name
Doctor of Philosophy (PhD)
Department
Education and Human Development
Committee Chair/Advisor
Golnaz Arastoopour Irgens, Ph.D.
Committee Member
Danielle Herro, Ph.D.
Committee Member
Nathan McNeese, Ph.D.
Committee Member
Edmond Bowers, Ph.D.
Abstract
This study explores how middle school-aged youths at an afterschool center utilized, examined, and produced AI applications for social good with the support of adults and peers. The study employs a qualitative single case study design. It uses thematic analysis and quantitative ethnography methods to analyze data from multiple sources, including field notes, interviews, focus groups, story completions, video recordings, and artifacts. Findings indicated that engaging youths in critical exploration of AI tools enhanced their ability to design interest-based AI applications that provide solutions for healthcare problems, security, and accessibility. Moreover, the computational thinking practices that youth engaged in increased as they progressed from using AI tools to producing AI tools, with youth engaging in more comprehensive computational thinking skills during the production and sharing of AI media. Additionally, adult guidance played multiple roles, acting as facilitators, co-investigators, and guides as they supported youth in exploring, producing, and debugging AI applications that address social good. Finally, peer testing emerged as a significant avenue for adult-youth collaboration, fostering participants' ability to troubleshoot and consider the ethical dimensions of AI. The study recommends promoting youth's machine learning agency by engaging them in authentic hands-on activities and environments that enable youth to design, test, and refine AI projects in collaboration with peers, adults, and community members.
Recommended Citation
Adisa, Ibrahim Oluwajoba, "Developing Machine Learning Agency Among Youth: Characterizing Youth Critical Use, Examination, and Production of Machine Learning Applications" (2024). All Dissertations. 3581.
https://tigerprints.clemson.edu/all_dissertations/3581
Author ORCID Identifier
https://orcid.org/0000-0003-1657-2030