cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
JMP is taking Discovery online, April 16 and 18. Register today and join us for interactive sessions featuring popular presentation topics, networking, and discussions with the experts.
Choose Language Hide Translation Bar

Pair wise comparison of slopes in ANCOVA

I am doing an ANCOVA analysis. I have a response variable water status which is continuous and I have soil moisture which is continuous as the predictor. I have a categorical variable which is variety and I desire to compare the slopes of each Y vs. X between different varieties to show if the linear regression has the same slope among varieties. The ANCOVA done on JMP and mentioned on this website many times yields to pair wise comparisons of the slopes between 2 varieties (one variety and a reference variety). However, taking all the comparisons together for 18 varieties (switching references every time) yields a high error rate. Is there anyway we can compare the slopes with a post-hoc doing multiple comparisons through SCheffe's or Tukey for example as a follow-up for ANCOVA and not only pair wise comparison through the indicator parameterizaton approach

thank you

1 REPLY 1

Re: Pair wise comparison of slopes in ANCOVA

First, you have the t ratio to test the value of the interaction term to the model. The interaction term represents a different effect of the predictor at different categories. (You know this because you selected ANCOVA.)

 

Second, the Fit Least Squares offers a selection of multiple comparison options. Click the red triangle at the top and select Estimates > Multiple Comparisons.

 

Screen Shot 2019-07-16 at 7.32.29 AM.png