Hi, Kynda!
Your question has many different exclusive interpretations, so I'm almost sure I have misunderstood your intention...
...but couldn't you structure the dataset for a repeated measures MANOVA like this?
More information might help.
Good luck.
Hi @Kevin_Anderson Thank you for the reply.
Here is an one of data sample, I've 3 groups (Control, 1mM treatment, 2mM treatment), Each group has two replications (R1 & R2) and I take measurements at 4 time points (0d,1d, 3d,5d). Each replication has six samples (subsampling) and same sample use for all the time points (repeated measures).
As you have illustrated I see, I can enter the data like that, Could you please let me know how can I perform the analysis and mean separation (posthoc-turkey) for eachtime point? I'm expecting to draw a graph like this and denote significance difference at each tim epoint?
Hi, Kynda!
How's this?
Your Excel data were transferred to a JMP data table, which is enclosed.
The graph is a simple bar chart from the Graph -> Chart pulldown that I customized a bit to look somewhat like yours.
Dr. John W. Tukey (please note: no "r". :-) may be spinning like a lathe right now, but I got the somewhat odd confidence interval number from a Mixed Model analysis of the repeated measures experiment (Fixed=Group, Random=Rep(Group), Repeated=Time Group=Group), then choosing the Multiple Comparison -> Tukey HSD, noting the Quantile and Adjusted DF from that analysis, and geting the t distribution value (98.8857...)from that. Yeah, that's kinda gnarley and someone may (should?!) have a better approach, but there you go. The Mixed Model analysis is somewhat more rigorous than your desired histograms, and basically say the same thing with different graphs.
Of course, this isn't the Multivariate part of the MANOVA you mentioned in the subject. I assume you have more response variables, but didn't include them in your example?
'Hope this helps.
Thank you very much Kevin!
Yes for this Repeated Measures Analysis (Mixed Model) is fine.
Could you please tell me how you perform turkey test here in JMP and got this graph?
This is how I perform the analysis:
Analyze > Fir Model
EG1 to Y
Group, Rep, Time added to Construct model
Then, Rep(Group) & Random (by changing attributes to Random effect) [is this correct?]
Method: REML
Run!
Is this the way I have to perform?
If yes, How can I proceed ahead with multiple comparision Turky?
Because my output looks like this and which button I have to click for Mean separation?
In Effect details
I want to get the mean separation for each time point separately, as shown in the graph?
Time"0": what is the mean separation of Control /1mm/2mM
Time "1": etc.
And please tell me how did you create this graph in JMP?
Thank you for your kind support