Hello,
I am working on a project investigating how certain variables impact the effectiveness of mental health interventions. I ran a RCT with two interventions. DV's were measured pre and post intervention and IV's were measured once pre intervention.
DV's: Anxiety, Positive Affect, Negative Affect
IV's: Previous positive experience with intervention, Attitude towards intervention, Expectancy, Credibility, and Trait anxiety
I am using a linear mixed effect model to test this.
anxiety_model <- lmer(Anxiety2 ~ Intervention + Positive_Experience + Credibility
+ Expectancy + Attitude + Trait_Anxiety + Anxiety1 + (1|PPT), DATA)
positive_model <- lmer(Positive_Affect_2 ~ Intervention + Positive_Experience + Credibility
+ Expectancy + Attitude + Trait_Anxiety + Positive_Affect_1 + (1|PPT), DATA)
negative_model <- lmer(Negative_Affect_2 ~ Intervention + Positive_Experience + Credibility
+ Expectancy + Attitude + Trait_Anxiety + Negative_Affect_1 + (1|PPT), DATA)
Negative_Affect_2 is negative affect measured after the intervention (DV), Negative_Affect_1 is the negative affect measured before the intervention. The same applies for anxiety and positive affect.
I have 68 participants in intervention 1 and 53 in intervention 2. I have left NAs in as I have read that the lmer() can handle missing data.
Intervention is a factor variable with levels 1 and 2. PPT (participant number) is a character variable.
I have never ran this type of model before, I am just posting to seek some help with this model as I am unfamiliar with it and want to understand where I am going wrong.
Questions:
- are my models fitted correctly?
- I keep getting warning messages concerning eigenvalues, can any give me advice on these?
Warning message:
In as_lmerModLT(model, devfun) :
Model may not have converged with 1 eigenvalue close to zero: -6.9e-09
Warning message:
Model failed to converge with 1 negative eigenvalue: -4.1e-08
This is my first post and I am relatively new to R so I'm sorry if this is a confusing post. Please feel free to ask questions about my coding and I will answer them.
Thank you,
Karen