MplusAutomation icon indicating copy to clipboard operation
MplusAutomation copied to clipboard

"Error extracting SUMMARY OF DATA in output file" when running 3-level SEM

Open pehkawn opened this issue 3 years ago • 3 comments

I've been attempting to extract data from the output of an empty three-level model in Mplus. However, running mplusModeler() or readModels() returns the error message:

Error extracting SUMMARY OF DATA in output file: model0.out

<simpleError in extractValue(pattern = "^\\s*Average cluster size\\s*", section,     filename, type = "dec"): More than one match found for parameter: ^\s*Average cluster size\s*
  model0.out>

I tried defining the same model with only one cluster, and the error message is gone. I can therefore only assume this has something to do with the model have more than one cluster.

pehkawn avatar Jan 23 '22 21:01 pehkawn

I'm having a simular issue with a cross-nested model. I get almost the same error message:

<simpleError in extractValue(pattern = "^\sNumber of clusters\s", section, filename, type = "int"): More than one match found for parameter: ^\sNumber of clusters\s sim.out>

andrew0302 avatar Feb 08 '22 11:02 andrew0302

Hi there,

I just pushed a commit that now parses the summary of data section for crossclassified models. I tested this against Mplus User's Guide example 9.24, which is the only cross-classified output I personally have. If you'd like me to test your specific case, please upload your .out file here (you'll have to label it as .txt for Github to accept it).

Re: the threelevel output above, @pehkawn, I couldn't replicate the issue. Could you upload your output file here?

I should also clarify that these 'simpleError' messages by the package do not reflect complete failure of parsing, but a localized error in parsing a certain section. So, your outputs may otherwise be intact and perfectly usable.

michaelhallquist avatar Feb 09 '22 14:02 michaelhallquist

Sorry for my late reply. Here it is. (.txt suffix added in order to be able to upload GitHub.)

pehkawn avatar Mar 18 '22 09:03 pehkawn