HOME Module List Site Index About This Site Glossary Resources Innovative Technology Center UT Statistics Courses
Using SAS Choose Design ANOVA Compare Means Regression Examples
Latin Square Design
  
Split-Split-Plot
     

Check that your experimental setup conforms to a
Latin Square Design with Split-Split-Plot

You have a Latin Square experimental design if your experimental units differ due to row and column variation, and each row and column have as many experimental units as there are treatments. The row and column factors will be blocks, and each treatment will appear once in every row and every column.

You have a Split-Split-Plot treatment design if you have three or more treatments (factors), each having discrete values or "levels". Each factor is applied to different experimental units. The experimental units are created by successively subdividing the largest experimental unit (whole plot).

Plant example: You are applying irrigation treatments to plots (whole plot experimental units), and these plots differ in moisture (row) and soil type (column); this makes it a Latin Square. You have 4 levels of irrigation that are applied to plots. Each plot is split into 3 sub-plots, to which one of the 3 levels of nitrogen fertilizer is applied. Then each of the sub-plots is split into 2 sub-sub-plots, each of which receives one of two mulch treatment levels.

Animal example: You are feeding diet treatments to animals (columns) over time (rows); this makes it a Latin Square. You have 4 protein level diets (whole-plot factor) assigned to animals. In each time period, a blood sample is taken and split into three tubes. Each tube is mixed a different buffer (sub-plot treatment factor). Then, in the lab, each tube is measured for amino acid concentration by two methods (sub-sub-plot treatment factors).


Begin SAS Analysis: Latin Square with Split-Split-Plot

Or select another design

Home | Contact us | Module list & summary | Glossary/Terms | About this site | Stats courses | Links | Index