This page integrates my instructional practice, curriculum leadership, and the scholarship that connects them. My approach centers on a consistent modeling framework (specify -> assume -> estimate -> diagnose -> interpret) and prioritizes reproducible computation and clear communication.

Teaching & Course Design

  • Modeling-centered curriculum: I organize courses around a consistent modeling framework that connects linear regression, generalized linear models, multilevel models, and modern predictive tools. This structure fosters transfer across problems and levels.

  • Active computational labs: Courses use R and R Markdown so students implement models, evaluate diagnostics, and produce reproducible interpretive write-ups.

  • Assessment design: I use a mix of written, oral, and project-based assessments; where appropriate, I have implemented mastery-based grading and flipped-classroom elements.

Representative courses: STAT-S 631/632 (Applied Linear Models I & II), STAT-S 431/432 (Applied Linear Models for undergraduates), STAT-S 350/352 (Inference & Data Modeling), STAT-S 100 (ModernDive). Enrollment and course redesign history available on the Courses page.

Program Development & Curriculum Architecture

As Director of Undergraduate Studies, I led the development and approval of two new undergraduate programs (B.S. in Data Science — Statistics Specialization; B.A. in Mathematics with Major in Statistics) and aligned required courses, learning outcomes, and assessment mechanisms. I drafted program proposals, mapped outcomes to courses, and shepherded approvals through departmental, college, and university committees.

Key elements:

  • Clear, measurable learning outcomes distinguishing applied competencies (implementation, interpretation, communication) from theoretical competencies (derivation, conceptual understanding).
  • Course restructuring (S431/S432 and S631/S632) to split applied and theory emphases while preserving coherent progression.

Educational Scholarship

  • Textbook: Co-author of Statistical Inference via Data Science (ModernDive approach), 2nd ed., CRC Press (2025). The book uses computation to support conceptual understanding and has been adopted beyond IU.

  • Workshops & Dissemination: Implementation-focused workshops at national meetings (USCOTS, WNAR/IMS) to help instructors adapt materials.

  • Manuscripts & Presentations: Multiple manuscripts in preparation focused on simulation-based pedagogy, leveling differences in math backgrounds, and simulation for linear models.

Research Contributions (Selected)

My methodological interests include random forest methods for interaction detection, generalized linear mixed models, and item response theory. Selected research outputs and technical reports are listed on the Publications page.