Subject: STA 260
Title: Statistical Practice and Data Analysis
School: College of Letters and Science LS
Department: Statistics STA
Effective Term: 2014 Fall
- Lecture/Discussion - 3.0 hours
Principles and practice of interdisciplinary collaboration in statistics, statistical consulting, ethical aspects, and basics of data analysis and study design. Emphasis on practical consulting and collaboration of statisticians with clients and scientists under instructor supervision. May be repeated for credit up to one time.
STA 207 or STA 232B; Working knowledge of advanced statistical software and the equivalent of STA 207 or STA 232B.
May be repeated for credit up to one time .
Open to students enrolled in the graduate program in Statistics or Biostatistics, as the class also serves to provide professional service to clients and collaborators who work with the students.
Expanded Course Description
Summary of Course Content:
Topics will be selected from the following: Communicating with Collaborators: Structuring Consulting Sessions, Structuring Collaborations, Effective Oral Communication, Report Writing, Grant Proposal Writing, Acquiring Background Knowledge and Familiarity with Relevant Literature. Principles and Practice of Consulting and Collaboration: Background of Consulting Problems, Ethical Aspects, Eliciting Relevant Variables, Analysis Plan Study Design: Sample Size Calculations, Experimental Studies, Observational Studies Data Analysis: Data Pre-processing, Handling Missing Data, Exploratory Data Analysis, Graphical Methods, Data Visualization Strategies for Statistical Modeling: Translating Applied Problems into Statistical Models and Testable Hypotheses, Practice of Linear, Nonlinear and Random Effects Models Communicating Results: Interpretation of Outputs, Reporting Statistical Analyses, Structuring and Writing Reports
Potential Course Overlap:
There is minor overlap in terms of underlying concepts with STA 206, 207, 232A, 232B and STA 223 where generalized linear models and applied linear models are discussed in-depth and with mathematical details, while the emphasis in this course is exclusively on practical and applied aspects and data analysis, such as the handling of missing data, so that the actual overlap is minimal.