michaelhallquist
Tech10 extraction
Hi - I recently updated to MPLUS 8.8 and since then get an error on running mplusautomation if I request TECH10: <simpleError in bivarFitData[mPos, ] <- c(vars, values): number of items to replace is not a multiple of replacement length>
I'm wondering if the 8.8 version has some changes to the bivariate model fit output causing this error?
It's very likely that the format in 8.8 has changed slightly, which always throws a wrench in the works of parsing outputs. Could you upload a .out file that has this problem (as a .txt file for Github to see it properly)? I can give it a look -- these version-to-version changes are often quick to fix.
Thanks for replying. I compared output files from 8.8 and version 8.6 and found several additions. It seems they have added 'Number of significant standardized residuals' into the Most frequent response patterns and chi-square contributions, Univariate and Bivariate Model Fit Information sections. Also they have added Univariate and Bivariate Distributions fit for each class I attach a text file, flagging these new lines or sections with *** Technical 10 output changes.txt
Not really the same issue, but while we are at it, the package does not really capture this tech10 message, as it only returns an empty list instead: TECH10 OUTPUT FOR THE CATEGORICAL VARIABLES IS NOT AVAILABLE BECAUSE THE FREQUENCY TABLE FOR THE LATENT CLASS INDICATOR MODEL PART IS TOO LARGE. Is this a design choice? Would it be better to be more explicit about capturing it, e.g. returning NA? An empty list makes me check the output file manually, in fear of it being missed. example.txt
Hi, just wondering if there is any update on this? I'm still having issues extracting tech10 results with the package. Thanks very much!
