Planning, Construction, and Statistical Analysis of Comparative Experiments. Datasets and SAS Code.

---

· Testing Hypotheses and Determining Sample Size

· Methods of Reducing Unexplained Variation

· Latin Squares

· Split Plot and Related Designs

· Incomplete Block Designs

· Repeated Treatments Designs

· Analysis of Experiments without Designed Error Terms

· Confounding Effects with Blocks

· Response Surface Designs

· Sequences of Fractions of Factorials

· Multi-stage Experiments

---

Testing Hypotheses and Determining Sample Size

Example 4.1 randomization test data

Methods of Reducing Unexplained Variation

· Rice/CO2 example data

· Rice/CO2 analysis of variance with subsamples

Latin Squares

· Randomize a 5x5 Latin square

· Dairy cow example data

· Switchgrass problem data

Split Plot and Related Designs

· 1999 Rice yield data

· Example 7.1 Proc GLM code and output

· Example 7.1 Proc Mixed code and output

· Silvicultural experiment data. Problem 8.4

Incomplete Block Designs

· Simulated BIB data

· Analysis of Variance of Simulated BIB data

· Compute power for testing treatment effects in BIB

· Proc Optex code to find an incomplete block design

· Proc Plan code to generate an alpha design

Repeated Treatments Designs

· Carburetor example data. Latin square balanced for first order carryover effects.

· Carburetor example SAS code and output.

Analysis of Experiments without Designed Error Terms

· Simulated Data for a 2x2x2x2

· Recycling paper data

Confounding Effects with Blocks

· Simulated data from 3x3x3 factorial in blocks of size 9

Response Surface Designs

· SAS code to generate an information surface on an irregular region

Sequences of Fractions of Factorials

· SAS code to construct a 3/32 fraction of a $2**7$ in three blocks

Multi-Stage Experiments

· SAS code to generate a fractional factorial for a split plot

· Output

---

Go to Marcia Gumpertz' home page

Go to NCSU Statistics Department home page

Last Updated 2/28/02

© Marcia L. Gumpertz 2002