Instruments to Measure Teachers’ Artificial Intelligence Competence: Comparison and Integration Based on the UNESCO Framework

Abstract

Introduction: AI training opportunities within higher education require a prior diagnostic assessment to determine faculty competence levels and suggest tailored learning paths. Although the UNESCO Framework stands out as a key reference due to its human-centered focus, it lacks a measurement instrument and quantitative ranges for proficiency levels. Available scales remain fragmented, in English, validated in other contexts, or poorly aligned with the framework. Given the heterogeneous training demands identified at the University of El Salvador (UES) between 2025 and 2026, integrating a suitable instrument became essential. Objective: To integrate a questionnaire to measure AI competencies among UES faculty using validated subscales aligned with the UNESCO Framework. Methodology: Between May and July 2026, a four-phase procedure evaluated six instruments (TAICS, AI-ED-SAT, AICTS, T-GAIC, AILST, and AITPACK). Key subscales were selected, five custom institutional contextualization blocks were added, and a pilot test was conducted with faculty members. Results and Discussion: A nine-block questionnaire was structured (AI-ED-SAT as the core base, three TAICS subscales, and custom contextual blocks). The pilot test confirmed comprehensibility, achieving an average clarity score of 96.09 % (Excellent). Findings demonstrate that the articulated integration of subscales overcomes local relevance limitations of international scales. Conclusions: The applied methodological framework provides a replicable procedure to assemble tools without compromising the original psychometric rigor. The resulting instrument yields a self-perception diagnosis essential for guiding faculty training at UES.