Dear researchers, I am trying to assess specific indirect effects in my model with three moderators. However, AMOS always gives a syntax error and my estimand could not run. When I try it on R studio (with lavaan and psych packages), I could not assign parameters to calculate specific indirect effects. Could you please help me identify problems and solutions for this?
Below is the code in R studio:
library(psych)
library(lavaan)
# I already input my CSV data so now I just describe it
describe(my.data)
A =~ A2+ A3 + A4 + A5 + A7 + A8
MS =~ MS1 + MS2 + MS3 + MS4 + MS6 + MS7+ MS8
M =~ M1 + M2 + M4 + MA8
IM =~ IM1 + IM2 + IM3 + IM4
FLA =~ Listen + Speak + Read + Write
# Regression paths from IV to mediators
M ~ a1*IM
A ~ a2*IM
MS ~ a3*IM
# Regression paths from mediators to DV (FLA)
FLA ~ b1*M + b2*A + b3*MS + c1*IM
#From this moment, I tried to assign parameters to calculate specific indirect effects. However, none of the below functions works!
direct : c1
Error: object 'direct' not found
direct:= c1
Error in `:=`(direct, c1) : could not find function ":="
direct
I would try deleting the error term e26 from the model. This may solve the problem. Exogeneous variables (here: IMo) should not have an error term.
Dear Christian Geiser, Thank you for your suggestion.
Interestingly I made a basic mistake. However, the problem remains the same after I deleted that error term: The file contains syntax errors
As far as I recall, AMOS by default does not report indirect effects along individual paths when there is more than one indirect path between two factors/variables (e.g., in a parallel mediation model). Did you use user-defined estimands to get the estimates of the three indirect effects? If yes, maybe the syntax error is in the code defining these estimands, not in the model.
- Badges
- Science topic
- Similar topics
- Books