This course covers classical Design of Experiments – DoE techniques to design efficient experiments as well as tools to analyse the collected data. The principles of sample size calculations, strategies to remove and control undesirable sources of variability like the use of blocks and controls, as well as the most commonly used experimental designs will be discussed. The statistical analysis of designed experiments will be progressively introduced, starting with the t-test method used to compare two groups. Then, the analysis of variance technique (ANOVA) will be extensively covered from simple one-factor experiments to more advanced multi-factor situations where the interaction between factors needs to be considered. Multiple comparisons techniques used to locate differences will also be presented.
Course Outline
The recommended duration for this course is 8 online session(s).
Session 1: Experiments for Comparing 2 Groups - Part 1- Experimental Designs : Design, Use, Construction & Statistical Analysis
- Completely Randomised Design (CRD) & Paired Design
- Design Structure: Experimental Units, Randomisation, Blocking
- Statistical Analysis with Student's t-Test & Conditions for Using the Test
- Sample Size Determination
- Application to Completely Randomised Design
- Application to the Paired Design
- Understanding, Controlling & Quantifying Variation
- Common Pitfalls in Experimentation to Avoid
- Role of DoE in R&D
- Sources of Variation & Impact on Data Quality
- Tools for Controlling Variation
- Structure of Experimental Designs
- Factorial Designs Overview
- Construction & Replication
- Detecting Factor Effects & Factor Synergy
- Construction of Factorial Designs
- Sample Size Determination
- Analysis of Variance - ANOVA Principle
- Conditions for Using ANOVA
- Detecting & Interpreting Factor Effects
- ANOVA for Designs with a Single Factor
- Multiple Comparisons Techniques to Locate Differences
- ANOVA for Designs with Two Factors or More
- Notion of Interactions: Detection & Interpretation
- Sample Size Calculations
- Application to Case Studies
- What to do when Underlying Assumptions are Not Met?
- Replication vs. Pseudo-replication
- Alternative Methods Overview
- Key Elements in DOE - Summary
- Key Elements in SSC & Significance Testing- Summary
- Statistical Analysis of Designed Experiments
- Impact of Missing Data / Design Imbalance
- Efficient Results Communication Tools
- What to Include in Reports
The course served to strengthen my understanding of concepts that I thought I knew. Furthermore, it provided me with the necessary tools to plan more powerful experiments from a statistical point of view. The course was well structured and quite enjoyable!
This was a very useful class, Taught in a no-nonsense manner so everyone could apply the concepts to their research and walk away with practical knowledge for their jobs. Thank you. I found the class to be very beneficial and concisely taught.
Nothing short of brilliant! There are a million things we could do, hundreds of things we should do, and only a handful of things we can do in a year. How do you decide which will have the greatest impact? Learn it here! Measuring the impact of your actions isn’t always easy. Measuring the combined impact of your actions and their interactions is even harder. This course makes it easy and can easily be applied no matter what industry you are in.
This was a great introductory course to DOE. It was just the right amount of material for a 2 day class and focused more on application than equations. A background in basic descriptive and inferential statistics is needed.
This was a good course in understanding the basics of DOE. I liked the focus on concepts in data analysis.
I have found that the “Introduction to the Design of Experiments” course is essential for anyone who wishes to apply a disciplined approach to practical applications in a product development or design applications. The program appropriately covered the required fundamentals by working through practical examples. The material was very well organized and due to having a small group all questions relating to concepts were very well explored. The instructor was very helpful and well acquainted with the subject matter.
DOE helped me to understand the different sources of error and variability as well as the tools available to minimize variability – repetition, randomization, blocking and controls. I now have a much better understanding of simple experimental designs and more complex factorial designs – along with the corresponding statistical tools – t-test, and Analysis of Variance (ANOVA) F-test. The breaking down of the components of variability when performing an ANOVA was of great benefit. Moreover, the notion of an interaction between two factors and how to test multiple comparisons were also very useful. The XLSTAT add-in to Microsoft Excel provides powerful and efficient statistical tools that allowed me to obtain faster, more detailed, reliable and accurate results than I have seen with other statistical packages.
Natalie, thank you so much for being such a great teacher. The Stats/DOE courses were absolutely brilliantly put together and taught. I learned so much in 3 days and feel so much more confident in my work environment now.