Scholarlink

<p>The integration of interconnected devices with artificial intelligence enables Artificial Intelligence of Things (AIoT) that can assist teachers in monitoring student learning and providing timely interventions in classrooms. However, limited research has examined the factors influencing the quality of teachers’ AIoT integration, established theoretical foundations to predict quality, or applied comprehensive methodologies to address this issue, leaving open questions about how to promote high-quality AIoT integration. This study draws on the Capability, Opportunity, Motivation, and Behavior (COM-B) model and the Interactive-Constructive-Active–Passive (ICAP) framework through both linear (structural equation modeling) and nonlinear analyses (regression tree analysis). Survey data from 930 teachers in primary and middle schools with the adoption of an AIoT system supported by generative AI were analyzed to examine the impact of their capability (technological pedagogical content knowledge), opportunity (perceived support on first-order barriers), and motivation (self-efficacy, utility value, enjoyment, and habit) conceptualized in AIoT integration contexts on the integration quality (interactive, constructive, active, and passive learning). The linear analysis showed that opportunity directly and indirectly through motivation enhances AIoT integration quality, whereas capability affects AIoT integration quality only indirectly through motivation. The nonlinear analysis uncovered multiple effects of opportunity (the quality balance scale and brake) and motivation factors (the brake and amplifier). The findings advance the understanding of predictors of AIoT integration quality, informing interventions towards the effective application of novel technologies in education.</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

Examining teachers’ deep and shallow AIoT integration: SEM and regression tree analyses

  • 期刊论文

  • 发表日期：

    2026年3月24日

查看全文
  • Shuo Feng,
  • Xin Gao,
  • Qianqian Liu,
  • Zezhi Zhang,
  • Mingjie Li,
  • Qiyu Zhang,
  • Yingbin Zhang,
  • Shuai Wang

摘要

The integration of interconnected devices with artificial intelligence enables Artificial Intelligence of Things (AIoT) that can assist teachers in monitoring student learning and providing timely interventions in classrooms. However, limited research has examined the factors influencing the quality of teachers’ AIoT integration, established theoretical foundations to predict quality, or applied comprehensive methodologies to address this issue, leaving open questions about how to promote high-quality AIoT integration. This study draws on the Capability, Opportunity, Motivation, and Behavior (COM-B) model and the Interactive-Constructive-Active–Passive (ICAP) framework through both linear (structural equation modeling) and nonlinear analyses (regression tree analysis). Survey data from 930 teachers in primary and middle schools with the adoption of an AIoT system supported by generative AI were analyzed to examine the impact of their capability (technological pedagogical content knowledge), opportunity (perceived support on first-order barriers), and motivation (self-efficacy, utility value, enjoyment, and habit) conceptualized in AIoT integration contexts on the integration quality (interactive, constructive, active, and passive learning). The linear analysis showed that opportunity directly and indirectly through motivation enhances AIoT integration quality, whereas capability affects AIoT integration quality only indirectly through motivation. The nonlinear analysis uncovered multiple effects of opportunity (the quality balance scale and brake) and motivation factors (the brake and amplifier). The findings advance the understanding of predictors of AIoT integration quality, informing interventions towards the effective application of novel technologies in education.