Scholarlink

Background <p>Statistics education is essential for evidence-based medical practice, yet many medical students face challenges in achieving statistical proficiency. While attitudes, anxiety, and self-efficacy are known to influence learning, the sequential pathways through which they interact remain unclear. This study examines a chain mediation model to clarify how attitudes towards statistics forecast proficiency through statistics anxiety and self-efficacy among medical students.</p> Methods <p>A cross-sectional study was conducted with 576 medical students from Shandong University. Participants completed validated Chinese versions of the Statistics Anxiety Rating Scale (STARS) and the Current Statistics Self-Efficacy (CSSE) scale. Statistical proficiency was measured via final examination scores of the medical statistics course. Data were analyzed using structural equation modelling (SEM) with bootstrapping to test the direct and indirect effects within the proposed model.</p> Results <p>The SEM model demonstrated good fit. Negative attitudes towards statistics were associated with both lower proficiency directly (direct effect=-0.120, 95% CI: -0.224 to -0.027) and indirectly through two significant pathways: an independent mediation involving reduced self-efficacy (indirect effect = -0.035, 95% CI:-0.069 to -0.008), and a chain mediation where negative attitudes were positively correlated with higher anxiety, which in turn was associated with diminished self-efficacy, ultimately linked to lower proficiency (indirect effect = -0.007, 95% CI: -0.017 to -0.001). Crucially, the direct association between statistics anxiety and proficiency was non-significant (-0.018, 95% CI: -0.049 to 0.033), indicating that the relationship between anxiety and proficiency is fully mediated by self-efficacy.</p> Conclusion <p>The findings validate a chain mediation model, highlighting self-efficacy as the critical cognitive link connecting both emotional factors like anxiety and academic performance. This suggests a “confidence crisis” as a central feature in the learning process. Consequently, medical statistics teaching may benefit from moving beyond just managing anxiety to actively building students’ confidence. Using practical, relevant tasks helps students feel more capable, which can enhance learning resilience and improve their performance. Finally, it should be noted that causal interpretations require confirmation through longitudinal or experimental designs.</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

The influence of attitudes towards statistics on statistical proficiency among medical students: a chain mediating effect of statistics anxiety and statistics self-efficacy

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年1月2日

查看全文
  • Jing Liu,
  • Siqi Fan,
  • Yunxia Liu,
  • Shukang Wang,
  • Xiubin Sun,
  • Xiujun Li,
  • Zhongshang Yuan

摘要

Background

Statistics education is essential for evidence-based medical practice, yet many medical students face challenges in achieving statistical proficiency. While attitudes, anxiety, and self-efficacy are known to influence learning, the sequential pathways through which they interact remain unclear. This study examines a chain mediation model to clarify how attitudes towards statistics forecast proficiency through statistics anxiety and self-efficacy among medical students.

Methods

A cross-sectional study was conducted with 576 medical students from Shandong University. Participants completed validated Chinese versions of the Statistics Anxiety Rating Scale (STARS) and the Current Statistics Self-Efficacy (CSSE) scale. Statistical proficiency was measured via final examination scores of the medical statistics course. Data were analyzed using structural equation modelling (SEM) with bootstrapping to test the direct and indirect effects within the proposed model.

Results

The SEM model demonstrated good fit. Negative attitudes towards statistics were associated with both lower proficiency directly (direct effect=-0.120, 95% CI: -0.224 to -0.027) and indirectly through two significant pathways: an independent mediation involving reduced self-efficacy (indirect effect = -0.035, 95% CI:-0.069 to -0.008), and a chain mediation where negative attitudes were positively correlated with higher anxiety, which in turn was associated with diminished self-efficacy, ultimately linked to lower proficiency (indirect effect = -0.007, 95% CI: -0.017 to -0.001). Crucially, the direct association between statistics anxiety and proficiency was non-significant (-0.018, 95% CI: -0.049 to 0.033), indicating that the relationship between anxiety and proficiency is fully mediated by self-efficacy.

Conclusion

The findings validate a chain mediation model, highlighting self-efficacy as the critical cognitive link connecting both emotional factors like anxiety and academic performance. This suggests a “confidence crisis” as a central feature in the learning process. Consequently, medical statistics teaching may benefit from moving beyond just managing anxiety to actively building students’ confidence. Using practical, relevant tasks helps students feel more capable, which can enhance learning resilience and improve their performance. Finally, it should be noted that causal interpretations require confirmation through longitudinal or experimental designs.