This longitudinal study examines the implementation and evaluation of ALEKS (Assessment and Learning in Knowledge Spaces), an AI-driven adaptive learning platform, in the chemistry curriculum at Amarillo College during the 2024-2025 academic year. It contributes to the literature on technology-enhanced learning and its impact on STEM student performance.

Employing a mixed-methods approach, the research combined quantitative performance metrics with qualitative student feedback to evaluate the effectiveness of integrating adaptive learning technology. Data were collected from diverse cohorts across various course formats, including traditional semester-length and intensive summer courses, for robust comparative analysis.

The introduction of ALEKS resulted in significant performance improvements. Course completion rates increased from 78.3% to 95.2%, reflecting a 16.9 percentage point rise. This suggests that ALEKS effectively addressed learning challenges and reduced dropout rates typical in STEM courses.

Student evaluations, based on 190 responses across 12 sections, yielded a 96.45% response rate, indicating significant improvements in perceived learning outcomes and satisfaction. Students reported improvements in understanding, engagement, and overall course effectiveness, with the Summer 2024 cohort achieving an average rating of 4.55 out of 5.

A comparative analysis revealed that the intensive 8-week summer format outperformed traditional 16-week courses in terms of learning outcomes, with an effect size of Cohen's d = 0.61, indicating significant practical relevance. This suggests the intensive format may enhance the benefits of adaptive learning technologies.

Despite positive results, the study highlighted pedagogical challenges, particularly in aligning digital content with traditional formats, emphasizing the need for comprehensive faculty training and curriculum redesign.

Quantitative findings also indicated an 18.65 percentage point increase in overall student success rates and a 22% improvement in learning outcomes as measured by standardized assessments. These underscore the potential of AI-driven adaptive learning in chemistry education and other STEM fields.

The research's implications extend beyond Amarillo College, offering insights for educational leaders, curriculum developers, and technology specialists. The findings enhance understanding of adaptive learning's effectiveness and provide practical guidance for implementation in higher education STEM programs.

In conclusion, this study demonstrates that the planned implementation of AI-based adaptive learning systems can significantly enhance student success and learning outcomes in chemistry, laying the groundwork for future research on optimal strategies and supporting the broader integration of adaptive learning across STEM curricula.

Keywords: Adaptive learning, high-success, ALEKS.