Completely Randomized Design (CRD)

The most basic experimental design is a completely randomized design (CRD) where experimental units are randomly assigned to treatments. If the design has multiple units for every treatment, the design is replicated. If the design has the same number of units for every treatment, the design is balanced. An example testing wood glue is provided.

Probability playlist:    • STAT 587 - Probability  
Inference playlist:    • STAT 587 - Inference  
Regression playlist:    • STAT 587 - Regression  

STAT 587: https://www.jarad.me/courses/stat587Eng/
STAT 587 Videos: https://www.jarad.me/courses/stat587E...
Slides: https://www.jarad.me/courses/stat587E...


Amazon affiliate links to props:
Tetris light: https://amzn.to/3g8BXde
Dragonwood: https://amzn.to/3k96IS7
Jumbo penny: https://amzn.to/3g8BXde
Jumbo quarter: ??
DCI Science Quiz Wal Clock: https://amzn.to/3g8BXde
Foam dice: https://amzn.to/3g8BXde
TIX LED Clock: https://www.geekalerts.com/tix-led-cl...
Bicycle Playing Cards: https://amzn.to/2E36Na8
Recording in progress door hang sign: https://amzn.to/3la32Av


Chapters:
00:00 Introduction
00:29 Random treatment assignment
02:07 Data collection
03:10 Strength of wood glue
04:10 Completely Randomized Design
05:57 Data visualization
06:16 Regression model
07:19 Model diagnostics
07:34 Regression statistics
08:00 Interpret results