How to treat "Time" in a longitudinal study

General
#1

Hi everyone, I am doing research on same sex couple at the moment and I am having some questions longitudinal model in R that I hope someone can help me with.

  • I want to examine if Partner A (IV) at time 1 predicts Partner A(DV) and Partner B(DV) at time 1,2,3,4,5. So I test "Time" as one predictor with value of 0,1,2,3,4; and then test interaction effect of "Time" and IV (DV ~ Time*IV). But I am not sure if this way is appropriate or it only tells me about the growth trajectories of DV and the effect of IV on this trajectory?
    Thanks a lot for your help.
#2

is your data sufficiently anonymised so that you could share it, or a portion with the forum?
Also would you share the code snippet relating to what you tried?

#4

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

#3

I can share a section of my data, but I do not know how to get a small dataset from my original data. Can you show me?
The code chunk i use is as below:

model <-lme(selfCSI~ time*SelfIS + time*PartnerIS , 
               data = data2, random = ~time|CoupleID, na.action = "na.omit", method = "ML",control = lmeControl(opt = "optim"))

#>Linear mixed-effects model fit by maximum likelihood
 Data: data2 
        AIC       BIC   logLik
  -3211.466 -3160.716 1615.733

Random effects:
 Formula: ~time | CoupleID
 Structure: General positive-definite, Log-Cholesky parametrization
            StdDev     Corr  
(Intercept) 3.55456615 (Intr)
time        2.09213669 -0.769
Residual    0.00122212       

Fixed effects: selfCSI ~ time + SelfIS1 + PartnerIS1 + time * SelfIS1 + time *      PartnerIS1 
                    Value Std.Error  DF   t-value p-value
(Intercept)     16.539228 0.7995484 847 20.685714  0.0000
time             3.162756 0.5027901 847  6.290410  0.0000
SelfIS1         -0.091397 0.0333046 847 -2.744277  0.0062
PartnerIS1      -0.091397 0.0333046 847 -2.744277  0.0062
time:SelfIS1    -0.006323 0.0209046 847 -0.302467  0.7624
time:PartnerIS1 -0.006323 0.0209046 847 -0.302467  0.7624
 Correlation: 
                (Intr) time   SlfIS1 PrtIS1 t:SIS1
time            -0.785                            
SelfIS1         -0.967  0.758                     
PartnerIS1      -0.967  0.758  1.000              
time:SelfIS1     0.759 -0.967 -0.785 -0.785       
time:PartnerIS1  0.759 -0.967 -0.785 -0.785  1.000

Standardized Within-Group Residuals:
          Min            Q1           Med            Q3           Max 
-0.0018831382 -0.0002364230  0.0000944857  0.0002675336  0.0016769274 

Number of Observations: 1182
Number of Groups: 330

SelfIS and PartnerIS are my predictors.