The problem is how best to analyse data which involves multiple dependent measures, taken on multiple/repeated occasions and determine whether there's a difference between two groups.
I'm trying to analyse data from a study involving cochlear implants. Now a cochlear implant has an array with multiple electrodes, and for each electrode it's possible to measure various different parameters. These parameters can be, and are, typically measured over time for the purpose of understanding how the parameter changes through time. Per good study design, in the problem I'm considering there's the control group and the experimental group. To note is that there's no cross-over element in the design.
To be specific:
- Two groups of N test subjects: control and experimental
- 12 different values for any "measurement set", from a single person at a single point in time
- Multiple measurement points (in time), the repeated measures part
- Does the control and experimental group differ, generally?
- Is there a significant difference at the different time points?
- Are there any differences between the 12 values through time?
Thus my questions:
- My thought is that this is a repeated measures MANOVA analysis type. Have I understood the situation correctly?
- How should the measurements at multiple time points be taken into account? Or maybe this is catered for with the (multiple) repeated nature of the analysis.
Thanks in advance!