Biostatistician

Job Summary

The Motivation Lab and Michigan Roybal Center in the School of Kinesiology are seeking a Biostatistician to serve as the quantitative backbone across the full lifecycle of the lab's behavioral, observational, and clinical intervention research. This position is embedded in an interdisciplinary research team led by Principal Investigator David E. Conroy, Ph.D., and works closely with co-investigators, project managers, and clinical research staff on NIA-, NHLBI-, and NIDDK-funded studies spanning precision behavioral intervention science, mHealth, just-in-time adaptive interventions (JITAIs), and rigorous clinical trial methodology.

Responsibilities

• Study Planning & Methodological Design (20%): Perform sample size and power calculations, including simulation-based power analyses in R, for complex study designs prior to data collection. Design data management protocols, randomization strategies, and data architecture for clinical trials and observational cohorts.
• Complex Statistical & Data Science Analysis (45%): Conduct advanced statistical analyses across cross-sectional, prospective longitudinal, randomized controlled trial (RCT), and intensive longitudinal data (ILD/ecological momentary assessment/wearable or passive sensor) formats. Implement frequentist mixed-effects models and generalized estimating equations in R (lme4, nlme, glmmTMB). Model latent variables, structural equations, and longitudinal growth patterns in Mplus or R. Diagnose missing data mechanisms and apply modern handling methods (e.g., FIML, multiple imputation) suited to longitudinal attrition. Apply advanced hierarchical/multilevel models using both frequentist (lme4, glmmTMB) and Bayesian (Stan, brms, rstan) approaches as appropriate to the research question. Build and evaluate time-series machine learning and deep learning pipelines in Python (scikit-learn, PyTorch).
• Data Governance, Safety Monitoring & Open Science (20%): Prepare interim reports, safety metrics, and blinded/unblinded summaries for Data Safety and Monitoring and funding sponsors (e.g., NIH). Establish data management frameworks that ensure data integrity, HIPAA/IRB compliance, and reproducible analytic pipelines. Package, annotate, and document clean codebooks, datasets, and analysis scripts for deposit in public/restricted repositories (e.g., ICPSR, GitHub).
• Collaborative Dissemination, Documentation & Mentorship (15%): Lead the development and team-wide adoption of best practices for study documentation, reproducible analytic workflows, and data management standards. Co-author peer-reviewed manuscripts, grant applications, and conference presentations, contributing statistical methods text, methodological descriptions, and publication-grade data visualizations. Review and provide feedback on analytic code and outputs produced by less experienced lab staff or trainees; serve as a technical resource on quantitative methods within the lab.

Required Qualifications

• Master's degree or Ph.D. in Quantitative Psychology, Biostatistics, Data Science, Applied Statistics, Information Science, or a related quantitative field, or an equivalent combination of education (with a minimum of a Bachelor's degree) and experience.
• 3+ years of post-degree experience supporting health, behavioral, or clinical research projects.
• Demonstrated hands-on proficiency in both R and Python for statistical computing and data science workflows.
• Experience with intensive longitudinal or ecological momentary assessment (EMA) data, including passive sensor/wearable data streams.
• Experience with reproducible research workflows, including version control (Git/GitHub) and dynamic reporting tools (Quarto, R Markdown, or Jupyter Notebooks).
• Demonstrated experience producing publication-grade data visualizations (e.g., ggplot2, matplotlib/seaborn, or similar) to communicate statistical results to scientific and non-specialist audiences.
• Experience with REDCap database management and automated API-based data ingestion from connected devices or third-party platforms.
• Excellent written and oral communication skills, including the ability to translate statistical methods for non-specialist audiences (e.g., manuscript co-authors, team members, sponsors).
• Proven ability to contribute to a positive workplace culture and demonstrate the school's core values.

Desired Qualifications

• Experience applying machine learning methods (e.g., scikit-learn, PyTorch) to time-series or behavioral data.
• Experience with latent growth curve modeling (e.g., Mplus, lavaan).
• Demonstrated experience using Bayesian estimation methods (e.g., Stan, brms/rstan).
• Experience preparing DSMB reports or interim safety summaries for NIH-funded clinical trials.
• Experience curating and depositing research data/codebooks in open-science repositories (OSF, ICPSR, GitHub) and reporting results on clinicaltrials.gov.
• Familiarity with CONSORT/GCP trial design standards.
• Experience mentoring junior analysts, research assistants, or trainees in statistical programming, data management, and study documentation.
• Experience building interactive dashboards or reporting tools (e.g., Shiny, Quarto dashboards, Plotly) for sharing results with investigators or sponsors.
• Experience with dynamical systems modeling, system identification, or control-theoretic approaches to adaptive intervention design.