Abstract:

As data-driven technologies increasingly shape everyday life, equipping students to critically evaluate the ethical dimensions of data collection, use, and monetization has become an essential learning outcome in higher education. This paper presents the design and pedagogical rationale of a data ethics module embedded within an introductory, programming-based data science course for undergraduates, many of whom are not data science or computer science majors. The module uses real-world datasets and industry case studies to move students beyond technical competence and toward ethical reasoning about how data affects individuals and communities.

The module is organized around four interconnected thematic units. First, students analyze survey data on social media platforms and their influence on consumer behavior, then reflect on the tension between free services and forced data-sharing policies. Second, students examine revenue data from major technology companies and consider how corporate profits are tied to monetizing personal data. Third, students work with a dataset of documented data breaches to quantify how frequently sensitive personal information is exposed, grounding concerns about data security in concrete evidence. Fourth, students investigate targeted advertising through case studies involving discriminatory ad delivery, including the exclusion of protected groups from housing advertisements and gender-biased distribution of job postings.

Throughout the module, students engage in hands-on data manipulation and visualization using Python, reinforcing programming skills while confronting the ethical implications of the data they analyze. Each unit pairs technical tasks with open-ended reflection questions on issues such as informed consent, corporate responsibility, and algorithmic discrimination. Designed without a single correct answer, these prompts foster classroom discourse and encourage students to weigh competing values — convenience versus privacy, personalization versus exclusion, profit versus public trust. The module also exposes students to published research demonstrating how microtargeted advertising systems can be exploited to infer private user information, illustrating that privacy risks are empirically documented.

Planned extensions address two rapidly evolving areas. The first introduces the ethics of generative artificial intelligence training data, asking students to examine how large-scale models are built on web-scraped content — often without creators' consent — and to analyze tensions among innovation, intellectual property, and fair compensation. The second adds a self-reflective unit in which students evaluate how artificial intelligence tools are used within their own educational experience, creating a pedagogical loop where they apply the critical thinking skills developed in the module to the technologies shaping their learning.

The approach in this module reflects a commitment to integrating ethics not as a standalone lecture but as a recurring, evidence-based thread woven through technical coursework. Early observations suggest that students are more engaged with ethical questions when those questions arise from data relevant to their daily lives that they have manipulated themselves. The module is adaptable to other introductory data science or data literacy courses and offers a replicable framework for cultivating both ethical awareness and technical proficiency.

Keywords:

Data science, ethics, critical thinking, education.