Wu Guixiang, Yang Yanfei, Xu Chenglin
School of Public Safety and Emergency Management, Kunming University of Science and Technology, Kunming, Yunnan, 650093, PR China.
Heliyon. 2023 Oct 11;9(10):e20616. doi: 10.1016/j.heliyon.2023.e20616. eCollection 2023 Oct.
University accidents in China are frequent, and to find out the relationship pattern of factors influencing accidents, 248 university accidents occurring within 2017-2021 were studied using difference analysis (Independent-samples T-test, Mann-Whitney test), logistic regression analysis, and diagnostic analysis of receiver operating characteristic curves. The results show: The variability in time, space, and qualifications was statistically significant (p < 0.05), and when the number of university safety policies ≥77 would significantly reduce the frequency of university accidents, with an influence strength value of 0.884 and a diagnostic accuracy of 79.8 %. In addition, the perpetrators, the time and the location of the accidents were usually undergraduate students, first semester of university, and economically developed and educationally rich provinces, respectively, with influence strength value and diagnostic accuracy of greater than 1 and 70%, respectively. Finally, specific suggestions are offered for the future prevention and reduction of accidents at the University based on the findings of the studies.
中国高校事故频发，为找出影响事故的因素之间的关系模式，采用差异分析（独立样本t检验、曼-惠特尼检验）、逻辑回归分析和受试者工作特征曲线诊断分析，对2017-2021年期间发生的248起高校事故进行了研究。结果表明：时间、空间和资质方面的变异性具有统计学意义（p<0.05），当高校安全政策数量≥77时，将显著降低高校事故发生率，影响强度值为0.884，诊断准确率为79.8%。此外，事故的肇事者、时间和地点通常分别是本科生、大学第一学期以及经济发达和教育资源丰富的省份，影响强度值和诊断准确率分别大于1和70%。最后，根据研究结果，为未来高校事故的预防和减少提供了具体建议。
