Student and instructor perceptions of data science integration into science and engineering courses

Data science literacy is vital for undergraduate engineering and science students, yet questions remain about effective integration in curricula. This study investigates the impact of integrating discipline-specific data science modules into existing undergraduate STEM courses at three US universities through a research-practice partnership. Using mixed methods to analyze survey responses from 877 students and instructor grades and interviews across six courses, we examined changes in student data science perception across various demographics, academic levels, and disciplines and compared student and instructor perspective. Results show significant increases in student self-reported perception after completing one or more modules irrespective of course and institution differences. Analysis revealed alignment between student self-assessments and instructor evaluations. Students highlighted benefits including real-world applications and career relevance, while identifying challenges with data analysis tools and varying experience levels. These findings provide insights for educators seeking to integrate data science into curricula.

数据科学素养（data science literacy）对于工科与理科本科学生而言至关重要，但目前学界仍存在关于如何在课程体系中有效融入数据科学内容的疑问。本研究依托研究-实践伙伴合作模式，针对美国三所高校的本科科学、技术、工程与数学（STEM）课程，探究将适配学科方向的数据科学模块融入现有课程体系所产生的影响。本研究采用混合研究法（mixed methods），对六门课程的877名学生的问卷反馈、教师评分以及访谈资料进行分析，考察不同人口统计学特征、学业阶段与学科背景的学生的数据科学认知变化情况，并对比学生与教师的认知视角。研究结果显示，无论课程与高校存在何种差异，学生在完成一门或多门数据科学模块后，其自我报告的数据科学认知水平均有显著提升。分析结果还表明，学生的自我评估与教师的评价之间存在显著一致性。学生们指出了该融入模式的多项优势，如贴合实际应用场景、具备职业适配价值，但同时也提出了数据分析工具使用困难、学生基础经验参差不齐等挑战。本研究结果可为致力于将数据科学融入课程体系的教育工作者提供参考借鉴。