Scholarlink

<p>Artificial intelligence (AI) integration in education is advancing, but to use AI algorithmic systems, such as adaptive learning platforms and intelligent tutoring systems, productively and ethically in instruction, teachers require a degree of metacognitive malleability to ‘align’ their thinking with machine logic. Despite the need, existing instruments assess only proximal constructs (teacher metacognition; algorithmic/AI literacy). A validated, integrated, AI-education-context-specific instrument that operationalizes teachers’ metacognitive calibration of reliance on algorithmic decision support for instructional judgment in the Jordanian secondary-school context does not exist. This study aimed to design and validate an instrument, the Metacognitive–Algorithmic Alignment Scale (MAAS), to assess secondary school teachers’ capacity for reflective, ethical, and strategic thinking in AI-supported instructional settings in Jordan. A sequential exploratory mixed-methods study design was used. In Jordan, the study was conducted in 2025. For face and content validity, a panel of experts reviewed and pilot-tested the scale’s initial items. The final 43-item scale was administered to a sample of 832 secondary school teachers selected using stratified random sampling. The psychometric properties were examined using EFA, CFA (first- and second-order), and EGA. Reliability and measurement invariance were assessed by calculating Cronbach’s alpha, McDonald’s omega, Composite Reliability (CR), and Intraclass Correlation Coefficient (ICC), and by testing measurement invariance across gender. Five factors were generated by EFA: Ethical Alignment in Metacognitive–AI Processes; Reflective–Algorithmic Thinking; Cognitive Co-authoring with AI; Algorithmic Transparency for Teacher Metacognition; and Algorithmic–Cognitive Self-Regulation, which explained 64.30% of the total variance. CFA confirmed the five-factor structure (CFI = 0.994; RMSEA = 0.073), and EGA supported the stability of the network structure. The scale’s internal consistency was high (α = 0.92–0.96; ω = 0.90–0.96), the composite reliability was good (CR = 0.951–0.982), and the test–retest reliability was acceptable (ICC = 0.757–0.857). Structural invariance was upheld across genders. Overall, the MAAS is a valid and reliable instrument for measuring teachers’ metacognitive–algorithmic alignment, with implications for teacher training, policy-making, and future research on the educational use of AI. </p> Graphical Abstract <p></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

Assessing teachers’ AI-algorithmic competence: development and validation of the Metacognitive-Algorithmic Alignment Scale (MAAS)

  • 期刊论文

  • 发表日期：

    2026年3月30日

查看全文
  • Mohammad Nayef Ayasrah,
  • Ayoub Hamdan Al-Rousan,
  • Mazen Omar Almulla,
  • Abdullah Ahmed Almulla

摘要

Artificial intelligence (AI) integration in education is advancing, but to use AI algorithmic systems, such as adaptive learning platforms and intelligent tutoring systems, productively and ethically in instruction, teachers require a degree of metacognitive malleability to ‘align’ their thinking with machine logic. Despite the need, existing instruments assess only proximal constructs (teacher metacognition; algorithmic/AI literacy). A validated, integrated, AI-education-context-specific instrument that operationalizes teachers’ metacognitive calibration of reliance on algorithmic decision support for instructional judgment in the Jordanian secondary-school context does not exist. This study aimed to design and validate an instrument, the Metacognitive–Algorithmic Alignment Scale (MAAS), to assess secondary school teachers’ capacity for reflective, ethical, and strategic thinking in AI-supported instructional settings in Jordan. A sequential exploratory mixed-methods study design was used. In Jordan, the study was conducted in 2025. For face and content validity, a panel of experts reviewed and pilot-tested the scale’s initial items. The final 43-item scale was administered to a sample of 832 secondary school teachers selected using stratified random sampling. The psychometric properties were examined using EFA, CFA (first- and second-order), and EGA. Reliability and measurement invariance were assessed by calculating Cronbach’s alpha, McDonald’s omega, Composite Reliability (CR), and Intraclass Correlation Coefficient (ICC), and by testing measurement invariance across gender. Five factors were generated by EFA: Ethical Alignment in Metacognitive–AI Processes; Reflective–Algorithmic Thinking; Cognitive Co-authoring with AI; Algorithmic Transparency for Teacher Metacognition; and Algorithmic–Cognitive Self-Regulation, which explained 64.30% of the total variance. CFA confirmed the five-factor structure (CFI = 0.994; RMSEA = 0.073), and EGA supported the stability of the network structure. The scale’s internal consistency was high (α = 0.92–0.96; ω = 0.90–0.96), the composite reliability was good (CR = 0.951–0.982), and the test–retest reliability was acceptable (ICC = 0.757–0.857). Structural invariance was upheld across genders. Overall, the MAAS is a valid and reliable instrument for measuring teachers’ metacognitive–algorithmic alignment, with implications for teacher training, policy-making, and future research on the educational use of AI.

Graphical Abstract