manage "time" in longitudinal dyadic MCMC model

[JeanSebastien]

Hi,

I have a dyadic data set with repeated measures. The time variable refers to the census years (6 consecutive years in total). In a non-dyadic setting, I think time would be a level-1 variable. In a dyadic setting I am not so sure. The existence of the dyad is based on time, so time is equivalent for both individuals (in the absence of one individual in the census the dyad does not exist in the dataset). However, the other individual might have other "active" dyads for that specific year. Also, time is closely correlated to age since the data covers more than 60 years. How should I treat time in that context? Thanks,

[billb]

Hi JeanSebastien,
This seems more of a concept question rather than an MLwiN question. I think if you give more information about your data then it might be easier to answer and I am assuming the question is related to the earlier one you sent.
Best wishes,
Bill.

[DarrylHicks]

In a non-dyadic context, time is usually regarded as a level-1 variable, which represents individual observations over time. In a dyadic context, however, the notion of time becomes intertwined with the dyadic structure. Since the existence of a dyad is dependent on time, both individuals within it share the same time variable. Therefore, time cannot be considered a level-1 variable in the conventional sense.