For example:
Lamb BW. Model (p-values in parenthesis) is: BW = juniper (0.53) day (0.02) juniper x day (0.16).
Considering that juniper is not significant in the model, I don't think I can look at any linear/quadratic trends, correct?
Basically, does the model variable have to be < 0.05 (or 0.1) to evaluate or discuss the linear/quad. trends?
Not necessarily. The omnibus F test is essentially whether at least one mean is different. Trend testing is a specific contrast that involves all of the means--I have seen several examples where a trend test is significant while the omnibus F is not. The single degree of freedom trend test is more powerful because of the assumption of linearity (or quadratic-ness).
This is a different situation than the comparison of two means following the omnibus test, where you already have information that no two means are likely to be different.
Steve Denham
Not necessarily. The omnibus F test is essentially whether at least one mean is different. Trend testing is a specific contrast that involves all of the means--I have seen several examples where a trend test is significant while the omnibus F is not. The single degree of freedom trend test is more powerful because of the assumption of linearity (or quadratic-ness).
This is a different situation than the comparison of two means following the omnibus test, where you already have information that no two means are likely to be different.
Steve Denham
Are you ready for the spotlight? We're accepting content ideas for SAS Innovate 2025 to be held May 6-9 in Orlando, FL. The call is open until September 25. Read more here about why you should contribute and what is in it for you!
ANOVA, or Analysis Of Variance, is used to compare the averages or means of two or more populations to better understand how they differ. Watch this tutorial for more.
Find more tutorials on the SAS Users YouTube channel.