question on repeated measures using xtmelogit

[Maria Gardener]

I have a question regarding a multilevel logit model. I want to study the effect of several independent variables on care sufficiency among older adults. For all the participants I have repeated measures of the dependent variable ‘care sufficiency’ and almost all of the independent variables (except for sex, and education, since these variables remain the same for the participants in my sample).
I have read the following article ( https://www.rips-irsp.com/articles/10.5334/irsp.90/ ) on how to perform a multilevel logistic regression, but I am not sure which variables are on which levels and how I should put that in my STATA-code. I think all of the repeatedly measured variables are on level 1, and only respondent-id is level 2, is that correct?

[Clyde Schechter]

I think all of the repeatedly measured variables are on level 1, and only respondent-id is level 2, is that correct?

Quite possibly. But without more information, a firm answer cannot be given. Are there perhaps level 3 variables such as care provider identifier, or city, or other things that constitute meaningful clusters of respondents? Or maybe even higher level variables, like metropolitan statistical area and census region?

On the assumption that there are no higher levels in your data, just respondents who were surveyed repeatedly, then, yes respond-id is the only level 2 variable. The generic command for a multi-level model of this data, with random intercepts at the respondent_id level will be:

multi_level_model_command outcome_variable variables_other_than_respondent_id || respondent_id:

multi_level_model_command refers to any of the -me- suite of Stata commands (including -mixed-).

[Maria Gardener]

Hi Clyde, thank you for your reply! I was wondering if there’s a way to see the effect of time (the reported effect in xtmelogit appears to be the pooled effect, and whether it’s different per wave?). Do I have to use a different function in STATA?

[Clyde Schechter]

I don't understand the question. Is there a time variable separate from wave? Or are you just looking to include the separate values of wave in the model? I think it is best for you to show example data, the regression command you are currently running, and then explain what you would like to see in the output that is different from what you have.

To show the example data, be sure to use the -dataex- command. If you are running version 18, 17, 16 or a fully updated version 15.1 or 14.2, -dataex- is already part of your official Stata installation. If not, run -ssc install dataex- to get it. Either way, run -help dataex- to read the simple instructions for using it. -dataex- will save you time; it is easier and quicker than typing out tables. It includes complete information about aspects of the data that are often critical to answering your question but cannot be seen from tabular displays or screenshots. It also makes it possible for those who want to help you to create a faithful representation of your example to try out their code, which in turn makes it more likely that their answer will actually work in your data.