Statistics 582

Statistical Consulting

Spring 2025

Course outline

The goal is to develop the skills needed by a statistical consultant. Topics to be covered include data analysis, problem solving, report writing, oral communication with clients, data gathering, consulting management, etc.

In-class activities

The diverse topics in the course require a variety of delivery modes not limited to class-room lecturing, and our in-class activities will cover a broad spectrum. Class participation is essential. Considerable time will be spent on data analysis discussing various examples. There will be time devoted to specific statistical topics, with lectures, discussions, and in-class exercises mixed together. Some time will also be spent discussing data screening, report writing, etc.

Course Work

There will be two major individual projects similar to MS exams, contributing 20% each to the course grade. A group project will add another 20%, working with real clients in the SCS setting. Class participation and a variety of other assignments will split the remaining 40%.

Prerequisite

Working knowledge on experimental design (STAT 514) and linear models (STAT 525) is assumed. Some knowledge on the analysis of non-normal data (STAT 526) is preferred but not required.

References

The course materials will be drawn from various sources, and there is no required textbook. Some of the references are listed here as pdf downloads. The two classics listed below should make enjoyable readings, and likely would benefit your future careers as writers and data analysts.

• The Elements of Style (4th ed), by W. Strunk and E.B. White.
• Applied Statistics: Principles and Examples, by D.R. Cox and E.J. Snell.

We will also watch and critique some videos from the book and the ASA site listed below.

• Statistical Consulting: A Guide to Effective Communication, by J. Derr.
• Statistical Collaboration Training for Applied Statisticians.

Software

R will be used as the primary platform for demonstrations and data/code sharing, but you are free to use your preferred platform in your analysis. In fact, it would be appreciated if you implement techniques in or translate R code into other platforms, say SAS, and share your success with the rest of class.

Basic Principles.

Notes, Examples, Guidelines, and References

Assignments