2026 ASEE Annual Conference & Exposition

This is a work-in-progress (WIP) paper that investigates the concept and understanding of statistical mindsets among engineering students. In an era defined by data-driven decisions, a robust statistical mindset, which goes beyond academic knowledge to include data interpretation and critical thinking, is crucial. Despite its importance in engineering, where data interpretation and critical thinking are used, there is limited clarity on what this mindset entails. This need motivates the present study to explore and define statistical mindsets in engineering students. This study presents the initial phase of an ongoing investigation with three main objectives: (1) to assess engineering students’ understanding of statistical mindset concepts; (2) to define what constitutes a statistical mindset within the engineering student group; and (3) to conduct a preliminary thematic analysis that will inform the development of a rigorous statistical metric for assessing this mindset.

This study employs a qualitative method approach, beginning with the qualitative data collection through an online survey platform called QuestionPro. A total number of 38 engineering students from a Western university participated in the study by answering five open-ended questions that were specifically designed and tailored to elicit insights into their perceptions, understanding, and insights toward statistical mindset. Following the data collection, the research progressed to the thematic analysis phase. The preliminary thematic analysis allowed us to identify five recurring patterns and themes in responses, including acknowledging variability and uncertainty, data-driven decision-making, managing complexity in tasks, risk assessment and probability analysis, and confidence in evaluation and application. The findings from this preliminary thematic analysis provide an initial exploration of statistical mindsets among engineering students. The next step of the research will further explore thematic analysis, considering alternative approaches to ensure the findings extend beyond the structured questions. Moreover, having two or more researchers code the data strengthens inter-rater reliability.