lavaan ERROR: wrong number of arguments in modifier ()

[jessica...@socialneuro.psych.utoronto.ca]

Hello everyone,

I am trying to specify a model and I keep getting an error and I don't understand the errors. I am not sure how I need to fix my model to get it to run. Any help would be great. Here is the model I am trying to run:

model.2 <- '

fear  =~ fear_dur + fear_intensity + fear_freq 

anger =~ anger_dur + anger_intensity + anger_freq 

sad   =~ sad_dur + sad_intensity + sad_freq 

calm  =~ calm_dur + calm_intensity + calm_freq  

happy =~ happy_dur + happy_intensity + happy_freq 

anxiety  =~ anxiety_dur + anxiety_intensity + anxiety_freq

hope =~ hope_dur + hope_intensity + hope_freq

pos_intensity =~ v1*happy_intensity + v1*hope_intensity +  v1*calm_intensity

pos_dur =~ v2*happy_dur + v2*hope_dur +  v2*calm_dur

pos_freq =~ v3*happy_freq + v3*hope_freq +  v3*calm_freq

neg_intensity =~ v4*anger_intensity + v4*sad_intensity +  v4*anxiety_intensity + v4*fear_intensity

neg_dur =~ v5*anger_dur + v5*sad_dur + v5*anxiety_dur + v5*fear_dur

neg_freq =~ v6*anger_freq + v6*sad_freq + v6*anxiety_freq + v6*fear_freq

pos_freq ~~ neg_intensity

pos_freq ~~ neg_freq

pos_freq ~~ neg_dur

pos_intensity ~~ neg_intensity

pos_intensity ~~ neg_freq

pos_intensity ~~ neg_dur

pos_dur ~~ neg_intensity

pos_dur ~~ neg_dur

pos_dur ~~ neg_freq

pos_dur ~~ pos_intensity

pos_dur ~~ pos_freq

pos_freq ~~ pos_intensity

neg_dur ~~ neg_intensity

neg_dur ~~ neg_freq

neg_freq ~~ neg_intensity

fear  ~~ 0*pos_dur

fear  ~~ 0*pos_intensity

fear  ~~ 0*pos_freq

fear  ~~ 0*neg_dur

fear  ~~ 0*neg_intensity

fear  ~~ 0*neg_freq

anger  ~~ 0*pos_dur

anger  ~~ 0*pos_intensity

anger  ~~ 0*pos_freq

anger  ~~ 0*neg_dur

anger  ~~ 0*neg_intensity

anger  ~~ 0*neg_freq

sad  ~~ 0*pos_dur

sad  ~~ 0*pos_intensity

sad  ~~ 0*pos_freq

sad  ~~ 0*neg_dur

sad  ~~ 0*neg_intensity

sad  ~~ 0*neg_freq

calm  ~~ 0*pos_dur

calm  ~~ 0*pos_intensity

calm  ~~ 0*pos_freq

calm  ~~ 0*neg_dur

calm  ~~ 0*neg_intensity

calm  ~~ 0*neg_freq

happy  ~~ 0*pos_dur

happy  ~~ 0*pos_intensity

happy  ~~ 0*pos_freq

happy  ~~ 0*neg_dur

happy  ~~ 0*neg_intensity

happy  ~~ 0*neg_freq

anxiety  ~~ 0*pos_dur

anxiety  ~~ 0*pos_intensity

anxiety  ~~ 0*pos_freq

anxiety  ~~ 0*neg_dur

anxiety  ~~ 0*neg_intensity

anxiety  ~~ 0*neg_freq

hope  ~~ 0*pos_dur

hope  ~~ 0*pos_intensity

hope  ~~ 0*pos_freq

hope  ~~ 0*neg_dur

hope  ~~ 0*neg_intensity

hope  ~~ 0*neg_freq

neg_dur ~~ 0*neg_intensity

neg_dur ~~ 0*neg_freq

neg_freq ~~ 0*neg_intensity

pos_dur ~~ 0*pos_intensity

pos_dur ~~ 0*pos_freq

pos_freq ~~ 0*pos_intensity

'

And here is the error.

Error in lavaanify(model = FLAT, meanstructure = lavoptions$meanstructure,  : 

  lavaan ERROR: wrong number of arguments in modifier () of element NANANA

In addition: Warning message:

In lav_partable_flat(FLAT, meanstructure = meanstructure, int.ov.free = int.ov.free,  :

  duplicated elements in model syntax have been ignored: c("neg_intensity", "neg_dur", "neg_intensity", "pos_intensity", "pos_dur", "pos_intensity")c("~~", "~~", "~~", "~~", "~~", "~~")c("neg_dur", "neg_freq", "neg_freq", "pos_dur", "pos_freq", "pos_freq")

Thank you in advance!

[Terrence Jorgensen]

Not sure without seeing the rest of your script, but if you have a multigroup model, then you need to repeat the parameter label/value for each group if you want them to be the same value in each group.  For example, in a 3-group model:

fear  ~~ c(0, 0, 0)*pos_dur

I usually see the "wrong number of arguments in modifier" message if I don't use the right number of groups.

Terrence D. Jorgensen

Postdoctoral Researcher, Methods and Statistics

Research Institute for Child Development and Education, the University of Amsterdam

UvA web page: http://www.uva.nl/profile/t.d.jorgensen

[jahug...@gmail.com]

Hi Terrence, 

Thank you very much for getting back to me. I don't gave more than one group in my experiment. I simplified some of the syntax and am getting different errors based on the order of the syntax, which shouldn't  be the case.

For example using the following syntax in my model:

neg_intensity ~~ 0*neg_freq        gives me the same error as previously stated. But if I write it like this:

neg_freq ~~ 0*neg_intensity

It gives me an error that there is a duplicate model element, which there isn't. Below is my simplified model that I ran:

model.2 <- '

fear  =~ fear_dur + fear_intensity + fear_freq 

anger =~ anger_dur + anger_intensity + anger_freq 

sad   =~ sad_dur + sad_intensity + sad_freq 

calm  =~ calm_dur + calm_intensity + calm_freq  

happy =~ happy_dur + happy_intensity + happy_freq 

anxiety  =~ anxiety_dur + anxiety_intensity + anxiety_freq

hope =~ hope_dur + hope_intensity + hope_freq

pos_intensity =~ v1*happy_intensity + v1*hope_intensity +  v1*calm_intensity

pos_dur =~ v2*happy_dur + v2*hope_dur +  v2*calm_dur

pos_freq =~ v3*happy_freq + v3*hope_freq +  v3*calm_freq

neg_intensity =~ v4*anger_intensity + v4*sad_intensity +  v4*anxiety_intensity + v4*fear_intensity

neg_dur =~ v5*anger_dur + v5*sad_dur + v5*anxiety_dur + v5*fear_dur

neg_freq =~ v6*anger_freq + v6*sad_freq + v6*anxiety_freq + v6*fear_freq

pos_freq ~~ neg_intensity + neg_freq + neg_dur

pos_intensity ~~ neg_intensity + neg_freq + neg_dur

pos_dur ~~ neg_intensity + neg_dur + neg_freq

pos_dur ~~ pos_intensity + pos_freq

pos_freq ~~ pos_intensity

neg_dur ~~ neg_intensity + neg_freq

neg_freq ~~ neg_intensity

fear  ~~ 0*pos_dur +  0*pos_intensity + 0*pos_freq + 0*neg_dur + 0*neg_intensity + 0*neg_freq

anger  ~~ 0*pos_dur + 0*pos_intensity + 0*pos_freq + 0*neg_dur + 0*neg_intensity + 0*neg_freq

sad  ~~ 0*pos_dur + 0*pos_intensity + 0*pos_freq + 0*neg_dur + 0*neg_intensity + 0*neg_freq

calm  ~~ 0*pos_dur + 0*pos_intensity + 0*pos_freq + 0*neg_dur + 0*neg_intensity + 0*neg_freq

happy  ~~ 0*pos_dur + 0*pos_intensity + 0*pos_freq + 0*neg_dur + 0*neg_intensity + 0*neg_freq

anxiety  ~~ 0*pos_dur + 0*pos_intensity + 0*pos_freq + 0*neg_dur + 0*neg_intensity + 0*neg_freq

hope  ~~ 0*pos_dur + 0*pos_intensity + 0*pos_freq + 0*neg_dur + 0*neg_intensity + 0*neg_freq

neg_dur ~~ 0*neg_intensity + 0*neg_freq 

neg_intensity ~~ 0*neg_freq 

pos_dur ~~ 0*pos_intensity + 0*pos_freq

pos_intensity ~~ 0*pos_freq 

'

[Terrence Jorgensen]

You still didn't provide your full R script, but I highlighted the duplicated elements I noticed in your model syntax.

model.2 <- '

           ## factor loadings ...  

pos_freq ~~ neg_intensity + neg_freq + neg_dur

pos_intensity ~~ neg_intensity + neg_freq + neg_dur

pos_dur ~~ neg_intensity + neg_dur + neg_freq

pos_dur ~~ pos_intensity + pos_freq

pos_freq ~~ pos_intensity

neg_dur ~~ neg_intensity + neg_freq

neg_freq ~~ neg_intensity

fear  ~~ 0*pos_dur +  0*pos_intensity + 0*pos_freq + 0*neg_dur + 0*neg_intensity + 0*neg_freq

anger  ~~ 0*pos_dur + 0*pos_intensity + 0*pos_freq + 0*neg_dur + 0*neg_intensity + 0*neg_freq

sad  ~~ 0*pos_dur + 0*pos_intensity + 0*pos_freq + 0*neg_dur + 0*neg_intensity + 0*neg_freq

calm  ~~ 0*pos_dur + 0*pos_intensity + 0*pos_freq + 0*neg_dur + 0*neg_intensity + 0*neg_freq

happy  ~~ 0*pos_dur + 0*pos_intensity + 0*pos_freq + 0*neg_dur + 0*neg_intensity + 0*neg_freq

anxiety  ~~ 0*pos_dur + 0*pos_intensity + 0*pos_freq + 0*neg_dur + 0*neg_intensity + 0*neg_freq

hope  ~~ 0*pos_dur + 0*pos_intensity + 0*pos_freq + 0*neg_dur + 0*neg_intensity + 0*neg_freq

neg_dur ~~ 0*neg_intensity + 0*neg_freq 

neg_intensity ~~ 0*neg_freq 

pos_dur ~~ 0*pos_intensity + 0*pos_freq

pos_intensity ~~ 0*pos_freq 

'

The first time you specify the highlighted parameters, you freely estimate them.  The second time you specify them, you fix them to zero.

[jahug...@gmail.com]

Hi Terrence,

Yes, thank you for catching that. I thought that would cause a problem in the model, so I left it out and then was told to put it back in. Thank you for highlighting that issue again. My script is pretty limited, just doing a basic cfa:

fit <- cfa(model.2, data = MTurk)

summary(fit, fit.measures = TRUE)

[Yves Rosseel]

> And here is the error.
>
> Error in lavaanify(model = FLAT, meanstructure =
> lavoptions$meanstructure, :
> lavaan ERROR: wrong number of arguments in modifier () of element NANANA

The error message is a result of the duplicated elements, as listed in
the warning message:

> In addition: Warning message:
> In lav_partable_flat(FLAT, meanstructure = meanstructure, int.ov.free =
> int.ov.free, :
> duplicated elements in model syntax have been ignored:
> c("neg_intensity", "neg_dur", "neg_intensity", "pos_intensity",
> "pos_dur", "pos_intensity")c("~~", "~~", "~~", "~~", "~~",
> "~~")c("neg_dur", "neg_freq", "neg_freq", "pos_dur", "pos_freq", "pos_freq")

Removing the duplicated elements from the model syntax will fix it. In
dev 0.5-21, the list of duplicated elements is much better readable, and
the error message is avoided.

Yves.

[Jessica Hughes]

Great, thank you for the suggestion! It worked.

Jessica A. Hughes, M.A.

Ph.D. Candidate | University of Toronto

Department of Psychology

Computational Affective Neuroscience Lab

--
You received this message because you are subscribed to a topic in the Google Groups "lavaan" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/lavaan/EVxo-24ozz8/unsubscribe.
To unsubscribe from this group and all its topics, send an email to lavaan+unsubscribe@googlegroups.com.
To post to this group, send email to lav...@googlegroups.com.
Visit this group at https://groups.google.com/group/lavaan.
For more options, visit https://groups.google.com/d/optout.

[franz...@gmail.com]

Hello,

I receive the same message but only when I write as many arguments in c() as I have groups (in my case 50). If I write c(a,a) which would be two groups, there is no error message. So I was wondering what is wrong?

Thanks in advance!

Franzi

[Terrence Jorgensen]

I receive the same message but only when I write as many arguments in c() as I have groups (in my case 50). 

You only get an error when the number of modifiers correctly matches the number of groups?  That seems strange. Is missing data preventing any groups from being represented in the analysis?  Does this happen in the latest development version?

install.packages("lavaan", repos="http://www.da.ugent.be", type="source")

With 50 groups, perhaps you should consider a multilevel rather than multigroup model.

http://users.ugent.be/~yrosseel/lavaan/zurich2017/MULTILEVEL/lavaan_multilevel_zurich2017.pdf

[franz...@gmail.com]

Thank you Terrence! Yes, indeed it works much better with a multilevel approach.