Situated Algorithmic Thinking: Preparing the Future Computing Workforce for Ethical Decision-Making through Interactive Case Studies

George Mason University, Fairfax VA

Investigators

Abstract

This project aims to serve the national interest by studying how undergraduate students develop ethical decision-making skills. It will specifically focus on these skills in the context of algorithmic thinking. Algorithmic thinking solves problems by identifying step-by-step instructions that can solve the original problem as well as be used again and again to solve related problems. Algorithmic thinking is a foundation of computational thinking, information technology, and computing, all of which have had positive effects on many aspects of life. However, there is increasing concern about how algorithm-based technology may harm individuals and society. One concern is the potential for the algorithms that run software to have unintended outcomes, including production of biased results. For example, a hiring algorithm designed to evaluate candidates for an engineering job may be biased toward hiring male engineers because the data about successful engineers is mostly about successful male engineers. Therefore, it will be important to integrate ethical decision making into the computer science curriculum so that future programmers are aware of and can mitigate unintended algorithmic outcomes. Toward this goal, the project will develop, implement, and test six interactive case studies to engage undergraduate students in the ethical aspects of algorithmic thinking and algorithm design. The case studies will enable students to think through issues of algorithmic decision making from different perspectives. This experience is expected to promote student understanding of ethical considerations in the context of algorithmic thinking. The interactive case studies will be adaptable for use in other courses, in standalone workshops, and at other institutions. The ability to use algorithmic thinking to design and develop technology is a core concern for education of the future workforce. The challenge with algorithmic decision-making at a societal level is that algorithms can be (1) biased due to the characteristics of the underlying data used to train different models; (2) unaccountable in that there is no mechanism to audit them or for redress if an algorithm is misused or makes an error; and (3) misunderstood regarding how algorithm-driven decisions impact social justice or ethical issues. Given the importance of algorithms and their complexity, it is critical that students learn ethical decision-making as part of their computer science education. Building on the situated cognition paradigm of learning, this project will develop interactive case studies that provide students the opportunity to think through issues of algorithmic design. Engaging with the material from different perspectives and domains will allow students to develop a well-grounded understanding of the importance of ethics in the context of algorithm development. The research component of the project will measure how perspectival understanding develops using interactive case studies. The research study will use a mixed methods approach involving analysis of student-generated artifacts such as concept maps and analysis of student explanations, using coding strategies to document important trends and outcomes. This project is supported by the NSF Improving Undergraduate STEM Education Program: Education and Human Resources, which supports research and development projects to improve the effectiveness of STEM education for all students. Through the Engaged Student Learning track, the program supports the creation, exploration, and implementation of promising practices and tools. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.