学术生产者ChatGPT学术使用的影响因素及群体差异--基于SEM和fsQCA的实证分析  

Influencing Factors of Academic Producers to Use ChatGPT and Group Differences--Empirical AnalysisBased on SEM and fsQCA

作　　者：周云倩 张新彤 夏婧 ZHOU Yun-Qian;ZHANG Xin-Tong;XIA Jing(School ofJournalism and Communication,Nanchang University,Nanchang,Jiangxi,China 330031;School of Literature and Media,Wenzhou University of Technology,Wenzhou,Zhejiang,China 325000)

机构地区：[1]南昌大学新闻与传播学院,江西南昌330031 [2]温州理工学院文学与传媒学院,浙江温州325000

出　　处：《现代教育技术》2025年第2期117-127,共11页Modern Educational Technology

基　　金：江西省哲学社会科学重点研究基地重点项目“基于复杂性视角的社交网络舆情演化机理与善治研究”(项目编号:21SKJD01)的阶段性研究成果。

摘　　要：当前,在ChatGPT等生成式人工智能技术引发的学术生产变革中,学术生产者正面临机遇与风险交织的不确定性前景,其对ChatGPT的学术使用,实属基于潜在收益和损失的一种风险决策行为。因此,文章首先基于前景理论和技术接受模型构建学术生产者ChatGPT学术使用的影响因素假设模型。然后,文章分别采用结构方程模型(Structural Equation Model,SEM)和模糊集定性比较分析(Fuzzy-set Qualitative Comparative Analysis,fsQCA)对551份有效样本数据进行分析,发现“绩效驱动”和“意向驱动”是引发学术生产者ChatGPT学术使用的主要动因,且感知风险在其中发挥正向作用。最后,文章通过多群组分析,进一步揭示了不同性别、学科、学历的学术群体ChatGPT学术使用的影响机制差异,旨在为生成式人工智能学术应用相关监管政策的制定和优化提供参考,并为学术生态的良性发展寻求进路。At present,in the academic production revolution caused by generative artificial intelligence(GenAI)such as ChatGPT,academic producers are confronted with an uncertain prospect where opportunities and risks are intertwined.Their academic use of ChatGPT has also become a risky decision-making behavior based on potential gains and losses.Therefore,based on the prospect theory and the technology acceptance model,this paper first constructed the hypothesis model of influencing factors of the academic use of ChatGPT by academic producers.Then,the structural equation modeling(SEM)and fuzzy set qualitative comparative analysis(fsQCA)methods were adopted to analyze 551 valid sample data.The findings revealed that“performance drive”and“intention drive”were the main motivations for academic producers to use ChatGPT in academic production,with perceived risk playing a positive role in this process.Finally,through multi-group analysis,the differences in the influencing mechanisms of academic use of ChatGPT among academic groups of different genders,academic disciplines,and academic qualifications were further revealed,expecting to provide reference for the formulation and optimization of regulatory policies related to the GenAI academic application,and seek a way for the benign development of academic ecology.

关 键 词：学术生产 ChatGPT 生成式人工智能 结构方程模型 多群组分析 模糊集定性比较分析 

分 类 号：G40-057[文化科学—教育学原理]