I am looking for some help with interpreting the coefficients in a linear mixed effects model for a time varying variable (c.lonescale_UCLA_##time). All the other covariates are time invariant.
There are only two time points. Time is coded 1 = survey1 and 2 = survey2
The dependent variable is depression symptoms coded from less to more symptoms.
Lonescale_UCLA is also coded from less to more loneliness.
Code:
mi estimate: xtmixed dep_cesd9_ i.AgeSES_fu1_recode##time clsa_cohort_fu1##time c.education_bl_recode##time i.regionprv_fu1_recode##time urban_rural_2_fu1##time immigrant_status_bl##time c.chronicconditions_10_fu1##time alone2_covid_recode##time alone3_covid_recode##time bhv_selfq_covid##time i.exp_pand_se_covid_recode2##time bhv_spbg_covid_recode##time c.lonescale_UCLA_##time|| entity_id:, pweight(wghts_analytic_bl) pwscale(size)
HTML Code:
dep_cesd9_ Coef. Std. Err. t P>|t| [95% Conf. Interval] lonescale_UCLA_ .9578595 .0439655 21.79 0.000 .8716886 1.04403 time#c.lonescale_UCLA_ 2 .3917519 .0524694 7.47 0.000 .2889137 .49459
For time#c.lonescale_UCLA_: would the correct interpretation for the coefficient be "increased loneliness from time 1 to time 2 is associated with a greater increase in depressive symptoms from time 1 to time 2."
Thanks,
Sean
0 Response to Interpreting a time variant covariate in a linear mixed effects model
Post a Comment