Skip to main content

Timeline for GLMM: Many models?

Current License: CC BY-SA 3.0

7 events
when toggle format what by license comment
May 23, 2017 at 12:39 history edited CommunityBot
replaced http://stackoverflow.com/ with https://stackoverflow.com/
Mar 19, 2014 at 17:51 vote accept user3288202
Mar 19, 2014 at 17:47 vote accept user3288202
Mar 19, 2014 at 17:47
Mar 19, 2014 at 17:47 comment added user3288202 of course. Your reply is definitely helpful. I think it is easier to run two models with different baseline and report only comparison where relevant. Thank you very much.
Mar 19, 2014 at 17:18 comment added JupiterM104 Sorry, but I'm not entirely sure what you mean. However, if you're concerned about having multiple models, and just want one, be aware that changing which factor level is taken as the baseline does not change the underlying linear model. i.e. you need to make two models, one to test A vs B and A vs C, and another to test B vs C, but they are the same model. You'd only need to report 'using' one model, and present one set of results. e.g. take all results from the summary() of the first model, and just add relevant details of the B vs C test from summary() of the second model. Helpful?
Mar 19, 2014 at 16:52 comment added user3288202 Hi JupiterM104, I am interested in doing the simple way, i.e. recoding factor to compare B with C. However, I am still not sure if I can have to two model: 1) model with A as baseline and 2) model with B as baseline, and report the results from both models without selecting them. Is it possible?
Mar 19, 2014 at 13:01 history answered JupiterM104 CC BY-SA 3.0