mediation Error in if (xhat == 0) out <- 1 else { :

Error in if (xhat == 0) out <- 1 else { :

Open mg2485 opened this issue 4 years ago • 7 comments

Using the mediate function (please see my code below), I receive the following error message. I would like to estimate a multilevel mediation for the following 2x5 mixed design. While the IV is manipulated between-subjects, the DV is a repeated measure (binary). In particular, each participant made 5 binary choices for 5 different choice sets (within-subject). The mediator (i.e., perceived expertise) is a repeatedly measured continuous variable, which was measured after each choice.

Do you have any idea what could be wrong with my code?

Error in if (xhat == 0) out <- 1 else { : missing value where TRUE/FALSE needed

med.fit <- lmer(Expertise ~ Condition + (1|SubjectID), data = data) out.fit <- glmer(Choice ~ Expertise + Condition + (1 + Expertise|SubjectID), family = binomial(link = "logit"), data = data)

med.out <- mediate(med.fit, out.fit, treat = "Condition", mediator = "Expertise", sims = 5000)

summary(med.out)

The data looks like the following: SubjectID I ChoiceSet I Condition I Expertise I Choice 1 I 1 I 0 I 5 I 1 1 I 2 I 0 I 4 I 1 1 I 3 I 0 I 4 I 0 1 I 4 I 0 I 4 I 0 1 I 5 I 0 I 3 I 0 2 I 1 I 1 I 3 I 1 2 I 2 I 1 I 3 I 1 2 I 3 I 1 I 4 I 1 2 I 4 I 1 I 4 I 1 2 I 5 I 1 I 5 I 0