General Question

bongo's avatar

How to incorporate varying time intervals between repeated measures in a General Linear Model repeated measure covariate analysis in SPSS, HELP!

Asked by bongo (4302points) May 2nd, 2012

Ok so I am sure there are some SPSS whizzes on here so please help me even if its you trying to remember and sending me in sort of the right direction…

So I have some data of yolk sac absorption of trout and I want to look at the rate of absorption taking into account varying egg sizes at the start of the trial. I have been told by my tutor to perform a General Linear Model analysis with egg size as a controlled variable however when I plot the graph and run the stats the model assumes that I have equal time intervals between my repeated measures, this is not the case. This obviously affects the curvature of my graph and therefore whether it follows a log/linear etc scale (as is my understanding). How can I incorporate the varying time intervals into my repeated measure analysis so that my graph shows this variation?
I hope this makes sense, I am rather confused myself!

Please note: I know how to do a general linear model covariate analysis, it is incorporating the varying time between the repeated measures I am having a problem with!

Observing members: 0 Composing members: 0

5 Answers

Kayak8's avatar

This may not be particularly helpful, but these guys did a somewhat similar analysis and you may be able to work backwards from how they laid out their graphs to figure out if your variables are in the right place. They used day/degree against length and wet weight. Later they compared dry yolk sac weight against dry larvae weight.

bongo's avatar

Thanks for that, I have read that paper already. It does mean that it is possible to do somehow but finding the right buttons to click and getting the data in the right places for the stats to work is currently really starting to annoy me!

Kayak8's avatar

Could you refer to them as Encounter 1, Encounter 2, Encounter 3 etc and get away from the variable interval lengths and see how they graph?

bongo's avatar

I can do that but it is the rate of yolk sac absorption I want to look at therefore the varying time intervals are very important. I have plotted a graph with this data using the egg size as a covariate and with the time increments all assumed to be equal but this affects the curvature of the graph and so does not have a true representation of what is happening (therefore I wont be able to use it in my thesis).

Response moderated (Spam)

Answer this question

Login

or

Join

to answer.

This question is in the General Section. Responses must be helpful and on-topic.

Your answer will be saved while you login or join.

Have a question? Ask Fluther!

What do you know more about?
or
Knowledge Networking @ Fluther