Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

- JMP User Community
- :
- Discussions
- :
- Two way repeated measures ANOVA results

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

Highlighted

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

May 11, 2020 10:13 PM
(406 views)

Hello everyone, I hope you can help me.

I designed an experiment using two different genotypes (WT and TG) of mouse and they both were fed with either a control diet (Crtl) or a high-fat diet ( HFD). I measured the weight of each mice for several weeks and would like to statistically analyze the results.

For this approach since I have 2 factors, and time I decided to use the Two way repeated measures ANOVA . I did the procedure using the indications suggested here https://www.jmp.com/support/notes/30/584.html .I nested genotype inside diet and treated it as my random factor. Then did a full factorial cross of diet and time. However, I have problems with the interpretation of the results. My main objective is to test if both genotypes are affected by the HFD in the same way, meaning that they gained weight in the same way across time. But in the results section, it only lets me see the effect of each factor alone (diet, time, or genotype) or the effect of diet vs time. Where do I see the effect of diet on genotype along time?

Also is there a way to make a test on the difference in the mean of each group ( WT ctrl, WT HFD, Tg ctrl, and TG HFD) for each week? or do I need to run a two-way ANOVA separately to get this?

Thank you very much in advance for your response.

1 ACCEPTED SOLUTION

Accepted Solutions

Highlighted

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Yes, select **Analyze > Fit Model**. Use the **Add**, **Cross**, and **Macros** buttons to make the terms I suggested.

No, the N in N-way ANOVA indicates the number of categorical variables in the analysis. You have 2 categorical variables and 1 continuous variable, so we would have called it an Analysis of Covariance (ANCOVA) once be we do not (should not) bother with such distinctions any more.

Click the red triangle next to Fit Least Squares and select **Estimates** > **Custom Tests**.

Learn it once, use it forever!

4 REPLIES 4

Highlighted
##

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Re: Two way repeated measures ANOVA results

I think that your model is incorrect. The two factors are crossed, not nested. Both factors have only fixed effects. The subject should be the random effect, but there is no need to include the mouse explicitly in the model. Time should be entered alone and crossed with the other factors to determine if the time course is different under different conditions. The model might look like this (terms):

weight = intercept + time + genotype + diet + genotype*diet + genotype*time + diet*time

I would not think of time as a categorical factor (e.g., week 1 versus week 2). It likely exerts a continuous effect.

Learn it once, use it forever!

Highlighted
##

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Re: Two way repeated measures ANOVA results

Dear @markbailey

Thank you for your prompt reply. Regarding your suggested model...How would I do this in the JMP app? Simply crossing each factor in the "fit model" window? If I add one more row with the three factors crossed "genotype*diet*time" would it essentially be a three-way ANOVA?

Sorry if I wasn't clear on my original post on my second question, I do not want to compare each week, what I meant is to compare pairs (WT ctrl vs WT HFD, Tg ctrl, vs Tg HFD, and so on ...) for week 1, week 2, week 3, etc. Similar to what the "test slices" function does.

Once again thank you in advance for your time.

Kind regards.

Highlighted

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Yes, select **Analyze > Fit Model**. Use the **Add**, **Cross**, and **Macros** buttons to make the terms I suggested.

No, the N in N-way ANOVA indicates the number of categorical variables in the analysis. You have 2 categorical variables and 1 continuous variable, so we would have called it an Analysis of Covariance (ANCOVA) once be we do not (should not) bother with such distinctions any more.

Click the red triangle next to Fit Least Squares and select **Estimates** > **Custom Tests**.

Learn it once, use it forever!

Highlighted
##

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Re: Two way repeated measures ANOVA results

Thank you for the clarification.

I hope you have a nice day.

Article Labels

There are no labels assigned to this post.