Learning Analytics Portfolio

This paper analyzes a supervision meeting between a music therapy intern and supervisor through four learning theories: distributed cognition, situated cognition, narrative intelligence, and expertise, to interpret and improve instructional practice. Designed for program faculty and clinical educators, it draws on direct observation, note-taking, and theoretical application to evaluate learning interactions and propose interventions. The analysis reveals how tools, storytelling, and reflective practice shape professional learning. Through this process, I strengthened my skills in theoretical analysis, observation, and translating learning science into actionable improvements.

This final paper reports a qualitative study of in-the-ear supervision in a music therapy company, exploring how supervisors’ real-time feedback balances clinician learning with client care. Intended for clinical supervisors, training directors, and practice leaders, the report draws on virtual observations, field notes, and semi-structured interviews analyzed thematically. I identified categories of feedback (directions with/without explanation, reinforcement) and their functions, generating implications for training and supervision design. Through this project, I strengthened skills in qualitative coding, thematic analysis, and combining theory with data into concrete recommendations: with more time, I would add longitudinal and demographic data. 

This final paper presents an independent quantitative study examining how socioeconomic status predicts 5th-grade science achievement using regression techniques. Written for K-12 education stakeholders and policy makers who rely on data to guide equitable decision-making, the study synthesizes nine research memos into a full empirical analysis. It incorporates descriptive statistics, visualizations, diagnostics, and a theoretically informed final model. The revision process strengthened clarity, interpretation, and policy relevance. Through this project, I deepened my skills in model specification, data interpretation, and academic writing, producing findings that can meaningfully inform resource allocation and intervention strategies in schools. 

This three-part assignment uses regression, decision trees, association rule mining, and k-means clustering to examine student frustration, confidence patterns, nd online engagement. Designed for educational decision-makers, it applies statistical and data-mining methods to address how learners interact and succeed. Using tools such as RapidMiner and LMS datasets, I collaboratively demonstrated skills in model building, clustering, and ethical data interpretation. The findings reveal meaningful patterns in hint use, confidence change, and engagement behaviors. Through this process, I strengthened my analytical reasoning, data-preparation skills, and understanding of ethical considerations in learning analytics. 

This study uses Epistemic Network Analysis (ENA) to examine how museum employees’ workplace experiences shared on a public social media platform changed over time. In 2020, posts most often connected inappropriate comments with leadership inaction, while by 2021, they more strongly reflected management monitoring and cautious sharing. A two-sample t-test confirmed these shifts were statistically significant. Qualitative examples further illustrate how initially open accounts of misconduct and inaction became more guarded as employees grew aware of organizational surveillance. These findings demonstrate how monitoring can constrain employee expression in anonymous forums, reshaping how individuals participate in digital communities of practice and ultimately limiting the potential for open dialogue and institutional accountability.