Ultimately, we hope to provide in this SAS resource a guide for essentially every statistical situation that students and researchers are liable to encounter in agricultural and biological research.

PLAN AN EXPERIMENT:

Good statistical practice suggests that you choose your design before doing the experiment. This step will lead you through a series of questions by which you can identify an appropriate design. This will also give you an opportunity to learn about all of the various designs. At the end of this key, you will also learn about statistical tools for choosing how large the experiment needs to be (statistical power).

Click here for a key that helps you identify an appropriate design.
     Otherwise, skip to the next step, where you will analyze an experiment.

ANALYZE AN EXPERIMENT:

First, choose whether ANOVA or regression is the proper analysis approach for your experiment:

     Use a regression model for your research design if you are primarily interested in the relationship of
       continuous explanatory variables with your dependent variable . To enter the regression module,
       click the Regression tab.

     If your interest is in comparing means, then your research design uses an ANOVA model and you should
       continue on this page.

For ANOVA, you may choose among seven experimental designs and seven treatment designs, with the choice to further refine the analysis with any conbination of three specialized features.

Experimental Design Treatment Design Specialized Features     CRD Single Factor Sampling     RCBD Nested Replication     Latin Square Factorial Covariate     MS-Latin Square Split-Plot none     Switchback Strip-Plot       Crossover Split-Split       Incomplete Block Repeated Measures

This design matrix allows for about 400 unique experimental designs, each having its own specific SAS program for proper statistical analysis.

Select the ANOVA analysis module that is appropriate for you:

	Experimental Design
	CRD
	RCBD
	Latin Square
	MS-Latin Square
	Switchback
	Crossover
	Incomplete Block

Step 1. Choose (click) your Experimental Design from the list at left (in the blue table). To learn more about the designs, to help you determine which of the seven is appropriate for your experiment, click a choice in the yellow table above. Step 2. Clicking one of the seven choices at left takes you to a page that allows you to continue specifying your design, using one of the seven choices for Treatment Design defined in the center column above. Step 2 brings you into the analysis module for one of the 49 Experiment/Treatment Design combinations. Within each module, you will also be given the option to further customize your analysis by choosing one or more Specialized Features, if one happens to apply: Sampling, Replication and/or Covariate.

[Start with Step 1 above]

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