Hello,
I am trying to carry out a multilevel model using lmer to look at how the effect of (birthweight) changes the outcome (cognitive function) time whilst controlling for variables.
I have created a null model, then compared this null model with an effect of birthweight (in order to try asses whether there is a change over time.
Then I have conducted a model comparison to isolate the effect of overall change over time.
when I did this it came up with an error (as shown bellow)
Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, :
unable to evaluate scaled gradient
Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, :
Model failed to converge: degenerate Hessian with 1 negative eigenvalues