STAT 720
1
Welcome to STAT 720!
1.1
About this course:
1.1.1
Logistics
1.2
Learning goals
1.3
On notation
1.4
Why do designed experiments exist?
1.4.1
Example
1.5
The golden rules of designed experiments
1.6
Other concepts in designed experiments
1.7
For tomorrow
2
Basic types of designed experiments
2.1
Announcements
2.2
Review
2.3
Types of designs - the basics
2.3.1
Completely randomized design (CRD)
2.3.2
Randomized complete block design (RCBD)
2.3.3
Incomplete block design (IBD)
2.4
Skeleton ANOVA tables
2.4.1
Practice examples
2.5
Homeworks
3
Basic types of designed experiments
3.1
Review
3.2
Types of designs
3.2.1
Completely randomized design (CRD)
3.2.2
Randomized complete block design (RCBD)
3.3
Building an ANOVA skeleton using design (aka topographical) and treatment elements
3.4
Linear model good old friend
3.4.1
The most common assumptions behind most software
3.5
Some Practice
3.6
To do
4
Linear models, ANOVA shells applied to the more basic experiment designs
4.1
Review
4.2
Linear models
4.2.1
The most common model - Assumptions
4.2.2
Connection between this and your classical ANOVA table
4.2.3
Sorghum example
4.3
Takehomes
4.3.1
If you don’t include the design elements (blocks), their portion of the variance goes to the error
4.4
Tomorrow
5
Review, organizing data and other helpful tips
5.1
Kahoot! as review
5.2
Organizing data
5.3
Homework
6
The treatment structure
6.1
Announcements
6.2
Review
6.3
Treatment structure
6.3.1
Types of treatment structures
6.4
Where is the treatment structure connected to the statistical model?
6.5
Tomorrow:
7
What you ask of a designed experiment
7.1
Announcements
7.2
Review
7.3
ANOVA
7.3.1
In case you were wondering: ANOVA and types of sums of squares
7.4
Setting the stage: Estimated marginal means aka least squares means
7.4.1
Example:
7.4.2
Discussion
7.5
Tomorrow
8
Applied examples
8.1
Announcements
8.2
Applied example for today
8.3
Tomorrow
9
Review
9.1
Announcements
9.2
Next week
10
Applied Linear Mixed Models I
10.1
Announcements
10.2
Applied Linear Mixed Models
10.2.1
Review: random variables
10.3
Review: Designs
10.4
Tomorrow
11
Applied Linear Mixed Models II
11.1
Review: Designs
11.1.1
Let’s look at the anova table
11.2
Tomorrow
12
Practice: Hierarchical (Multilevel) Designs
12.1
Review: Hierarchical Designs
12.1.1
More pictures:
12.2
R demo
12.3
Tomorrow
13
More practice: Hierarchical (Multilevel) Designs
13.1
Review: Hierarchical Designs
13.1.1
Remember our example:
13.2
Building the ANOVA skeleton using design (aka topographical) and treatment elements
13.3
R demo
13.4
Tomorrow
14
Review
15
Review: RCBD and split-plot
15.1
Background
15.1.1
Baseline recipe for banana muffins:
15.2
Study / Research question
15.2.1
Proposed design experiment: RCBD
15.2.2
Treatment means and confidence intervals
15.3
Tomorrow:
16
Review: Applied split-plot
16.1
Announcements
16.2
Background
16.2.1
Research question
16.3
What is the best design?
16.3.1
Option A: another RCBD
16.3.2
Option B: split-plot design
16.4
Treatment means and confidence intervals for the split-plot design
16.5
Tomorrow:
17
Analyzing data from a split-plot design
17.1
Announcements
17.2
Background
17.2.1
Research question
17.3
Analyzing the data
17.4
Treatment means and confidence intervals for the split-plot design
17.5
Tomorrow:
18
Analysis and inference for a split-plot design
18.1
Announcements
18.2
Review of our experiment
18.3
ANOVA tables
18.3.1
Split-plot in a CRD
18.3.2
Other possible designs
18.4
Applied analysis in R
19
Analysis and inference for a split-plot design - Part II
19.1
Announcements
19.2
Review of our experiment
19.3
ANOVA tables
19.3.1
Split-plot in a CRD
19.3.2
Same design as above, with subsampling
19.4
Applied analysis in R
20
Power analysis
20.1
Announcements
20.2
Review of everything so far
20.3
Power of an experiment
20.3.1
Power calculations
20.4
Reminders
21
Power analysis II
21.1
Announcements
21.2
What are simulations?
21.2.1
Simulation demo
21.3
Power analysis demonstration
21.3.1
CRD - 2 reps
21.3.2
CRD - 3 reps
21.3.3
Split-plot - 2 reps
21.3.4
Split-plot - 3 reps
21.3.5
Split-plot - 2 reps + subsampling
21.3.6
Split-plot - 3 reps + subsampling
22
Review Friday
22.1
Announcements
23
Randomized complete block designs
23.1
Announcements
23.2
Randomized complete block designs
23.3
Blocks - fixed or random?
23.3.1
Applied design and analysis of an RCBD
23.4
Incomplete block designs
23.4.1
Applied design and analysis of a Balanced IBD
24
Split plot designs
24.1
Announcements
24.2
Split plot designs
24.2.1
Applied case: split-plot design
24.3
Summary of split-plot designs
25
More multilevel designs
25.1
Announcements
25.2
Review: statistical models
25.3
Multi-location trials
25.3.1
Example
25.4
Subsampling
25.4.1
Example
26
Planning a multi-location design
26.1
Announcements
26.2
Planning a multi-environment trial
26.2.1
On the importance of sample size
26.2.2
What do the results for different multi-environment trials look like?
27
Crossover Designs I
27.1
Announcements
27.2
Crossover designs
27.3
Variations to this example
28
Crossover Designs II
28.1
Announcements
28.2
Crossover designs
28.3
Applied example
28.4
Tomorrow
29
That’s all folks!
29.1
Announcements
29.2
Principles for conducting valid and efficient experiments
29.3
Modeling data generated by designed experiments
29.3.1
From ANOVA table to statistical model
29.3.2
What happens if we drop blocks, split-plots, etc?
29.3.3
Differences modeling multi-location designed experiments
29.4
Learn more about designing and analyzing experiments
30
Semester Project
30.1
Learning objectives
30.2
Partial deadlines
30.2.1
Project proposal - Due Friday June 20 at 2pm CT
30.2.2
Written report - Due Wednesday July 23 at 2pm CT for peer review
30.2.3
Oral presentation - Somewhere between July 21 - August 1
30.2.4
Practical reads
Published with bookdown
STAT 720 - Design of Experiments
Day 22
Review Friday
July 11th, 2025
22.1
Announcements
HW 3 due today