Overview:   
Large, multi-factor datasets are frequently encountered by today’s technical and business professionals.  Business forecasting, market research studies, product development experiments, and Lean Six Sigma projects represent examples where the course methods may be required.  This course picks up where the introductory Practical Data Analysis course leaves off, focusing on graphical and statistical methods particularly aimed at the analysis of multi-variable data.  Participants practice these methodologies using a leading PC-based statistical software package.  The course finishes with a workshop requiring participants to analyze a realistic, but challenging multi-variable dataset.

How You will Benefit:
By the end of the course, participants will have gained:

• A firm grasp of the basic concepts and tools of ANOVA and regression analysis.
• Ability to choose the best statistical methods to use to effectively model a set of multi-variable industrial / business data and correctly interpret the resulting analysis.
• Knowledge of the essential methods of model-building using multiple regression.
• Experience applying the course methods using the PC-based software package.

Course Outline:
Day 1

1. Course Introduction                  
   1. Review of Practical Data Analysis Methods: Five-Step Model for Data Analysis
   2. The Challenge of Analyzing Complex, Multi-factor datasets
   3. Review of Statistical Inference (e.g., Coefficient, Standard Error, p-value, etc.)
   4. Typical Erroneous Assumptions in Univariate Data Analyses
2. Analysis of Variance (ANOVA)            
   1. ANOVA terminology:  Sum of Squares, Mean Squares, F-ratio, etc.
   2. How ANOVA Works:  Partitioning the Sum of Squares
   3. Fixed vs. Random Factors, Crossed vs. Nested Designs
   4. One-factor & Multi-factor ANOVA:  Methods and Examples
   5. Graphical Methods in ANOVA:  Main Effects, Interaction, & Residuals Plots
3. Data Transformations
   1. Benefits to the analysis
   2. How to choose the transformation:  the Box-Cox Method
4. Components of Variance: ANOVA with Random Factors
   1. Useful COV Applications:  Measurement & Process Studies
   2. The ANOVA for COV Studies
   3. Calculation of Variance Components
   4. Graphical Methods in COV Studies

Day 2

1. Regression Analysis    
   1. Regression Terminology, Simple Linear Regression
   2. Least Squares, Interpretation of Regression Coefficients
   3. Decision-making on the Regression Line
   4. Goodness-of-fit Criteria:  r2, adjusted r2, residuals
   5. Calculating Predicted Responses and Input Values
2. Multiple Linear Regression
   1. Interpretation of MLR output
   2. Indicator Variables, Analysis of Covariance
   3. Polynomial Regression, 2nd Order model-building
   4. Data Transformations in MLR
   5. Variable Selection Procedures: Stepwise Regression
3. Other Modeling Procedures
   1. The General Linear Model:  Mixed Discrete & Continuous Factors
   2. Logistic Regression:  Regression with Discrete Responses

Who Should Attend:
Engineers, scientists, and other personnel in R&D, process/product engineering, manufacturing, quality assurance, business planning, and related disciplines who are responsible for data-based decision-making in their workplaces.  Completion of the Practical Data Analysis course or the equivalent is necessary for course participation.  Basic proficiency in use of a PC-based statistical software package is assumed.

Cost:
Two-day course cost: $600 per student for OBA members and $700 per student for nono-members.  Tuition includes: class, bound presentation, and lunch for both days.

***10% Discount for all student who register for both the Analysing Multi-factor class and the Practical Data Analysis class on 3/16/10 & 3/30/10***
PLEASE RESISTER FOR THIS CLASS THROUGH DAY 1 OF 2 Posting

Instructor Bio:  Don Lewis Ph.D.
Don Lewis is Principal, Lewis Consulting LLC, whose mission is to enable clients to improve their competitive performance through effective application of proven quantitative decision-making methodologies.  Since establishing his consulting practice in 1986, Don has trained and mentored over five thousand technical professionals to apply quantitative methods, such as Statistical Process Control and Design of Experiments, in their project work.  His consulting experience accrues from 50+ organizations across a diverse group of industries, including biosciences.  Clients have achieved significant performance improvement, including proprietary breakthroughs, as a result of implementing his services. 

Recently, as a Lead Instructor in Motorola University’s Digital Six Sigma Black Belt training program, Don has trained over two hundred and fifty Motorola Black Belts throughout the U.S., Europe, and Asia.  Since 2003 his Northwest Lean Six Sigma clients have saved over $16 million in project work completed in conjunction with his training programs.  He is an Adjunct Professor in both the Department of Management of Science & Technology at the OGI School of Science & Engineering in Portland, Oregon and the Atkinson Graduate School of Management at Willamette University.  Don is also a chapter author of the recently published "Encyclopedia of Statistics in Quality and Reliability."  He received his B.A. in mathematics from Claremont McKenna College and Ph.D. in biostatistics from the University of North Carolina at Chapel Hill.  Don is an ASQ Certified Six Sigma Black Belt.