Scholarlink

Objective <p>To construct precise learner personas for medical students by integrating social sensing data with the Entrustable Professional Activities (EPAs) framework, thereby informing targeted pedagogical strategies. The study aims to address the gap in integrating behavioral data with competency assessment for multidimensional learner characterization.</p> Methods <p>Purposive sampling was employed to recruit 45 clinical interns (June–December 2024) based on data saturation principles. Data collection combined semi-structured interviews, social sensing technology (multimodal data from medical forums and clinical sensor arrays with AES-256 encryption for privacy protection), and EPA-based clinical observations. Thematic analysis via NVivo 12 Plus was used to extract features, with manual coding validating theoretical saturation after 45 interviews.</p> Results <p>A four-dimensional persona framework (learning attitude, clinical competence, social interaction, self-improvement awareness) was discovered, yielding five archetypes: Proactive Achievers (28.89%), Skilled Practitioners (20.00%), Social Collaborators (17.78%), Passive Followers (24.44%), and Ambiguous Improvers (8.89%). Intergroup differences in EPA-assessed clinical skills were significant (ANOVA, <i>p</i> &lt; 0.05 for all dimensions).</p> Conclusion <p>The multimodal framework enables data-driven learner characterization, providing evidence for personalized medical education. Future work will validate persona utility in educational interventions.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Constructing medical student personas via social sensing data and entrustable professional activities (EPAs) framework: a multimodal thematic analysis approach

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年10月10日

查看全文
  • Chen Mengyu,
  • Fan Dan

摘要

Objective

To construct precise learner personas for medical students by integrating social sensing data with the Entrustable Professional Activities (EPAs) framework, thereby informing targeted pedagogical strategies. The study aims to address the gap in integrating behavioral data with competency assessment for multidimensional learner characterization.

Methods

Purposive sampling was employed to recruit 45 clinical interns (June–December 2024) based on data saturation principles. Data collection combined semi-structured interviews, social sensing technology (multimodal data from medical forums and clinical sensor arrays with AES-256 encryption for privacy protection), and EPA-based clinical observations. Thematic analysis via NVivo 12 Plus was used to extract features, with manual coding validating theoretical saturation after 45 interviews.

Results

A four-dimensional persona framework (learning attitude, clinical competence, social interaction, self-improvement awareness) was discovered, yielding five archetypes: Proactive Achievers (28.89%), Skilled Practitioners (20.00%), Social Collaborators (17.78%), Passive Followers (24.44%), and Ambiguous Improvers (8.89%). Intergroup differences in EPA-assessed clinical skills were significant (ANOVA, p < 0.05 for all dimensions).

Conclusion

The multimodal framework enables data-driven learner characterization, providing evidence for personalized medical education. Future work will validate persona utility in educational interventions.