Monte Carlo simulation confirms weight and score stability of the expert-derived priorities under judgment perturbation¶

Claim · All claims
Evidence · 1 study (2 entries) · design r2 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person · recheck by 2029: tool and model accuracy changes quickly

Subclaims¶

q2 i? In a 10,000-run Monte Carlo simulation, the top AI competency indicator retained 99.8% rank stability and the composite score's 95% CI [3.673, 3.867] fell entirely within the Agree evaluation range. → Zijing Wu and Qiang Li 2026 q2 i? Deterministic sensitivity analysis showed the top sub-indicators exhibited no ranking changes under ±30% weight adjustments. → Zijing Wu and Qiang Li 2026 (2)

Evidence¶

Zijing Wu and Qiang Li 2026¶

Zijing Wu and Qiang Li. (2026). Evaluation indicator system for AI certificate programs. International Journal of Educational Technology in Higher Education. https://doi.org/10.1186/s41239-026-00604-x

q2 · i? · design · r2

Monte Carlo simulation (N = 10,000) perturbing AHP pairwise comparisons with normally distributed noise (SD = 20%) and FAHP ratings with 20% adjacent-category shift probability. Results show tight concentration of the composite score and high rank stability for top indicators (C6: 99.8%).

"The A1 composite score distribution (Fig. 2, Panel B) was tightly concentrated (simulated mean = 3.7681 vs. baseline = 3.7822; CV = 1.31%). The 95% CI [3.673, 3.867] fell entirely within the “ Agree” evaluation range (3.5–4.5)."

Zijing Wu and Qiang Li 2026 (2)¶

Zijing Wu and Qiang Li. (2026). Evaluation indicator system for AI certificate programs. International Journal of Educational Technology in Higher Education. https://doi.org/10.1186/s41239-026-00604-x

q2 · i? · design · r2

Deterministic sensitivity analysis under ±10% to ±30% weight variation (Table 7). The article attributes the stability to structural gaps in original weights among sub-indicators, so relative importance relationships remain unchanged.

"the top 10 sub-indicators (AI Core Competency Achievement Rate among Students, Alignment of Curriculum Content with AI Frontier Technologies, Cross-Cultural Student Satisfaction Ratings, Industry/International Certification Recognition of Learning Outcomes, Faculty-Student Interaction Response Efficiency) exhibited no ranking changes under ± 30% weight adjustments, indicating extremely strong ranking stability"

Discussion¶

Related Claims¶

• Expert rating consistency passed all AHP thresholds, with the main dimension matrix CR = 0.0307 — related
• Student AI competency achievement and curriculum-AI frontier alignment dominate the indicator-level priority ranking — related
• Faculty Professional Competence shows the largest gap between perceived importance and current satisfaction among the five dimensions — related
• Experts rank Curriculum System and Content Design as the highest-priority dimension for AI certificate program evaluation, with Cross-Cultural Adaptability lowest — related
• Experts rate AI-related competences for learners and educators as the highest-impact and most important AI-in-education trend — related
• Expert perceptions reveal a developmental asymmetry in which content knowledge outweighs integrated pedagogical capacity in AI education — related