Inquiry in warning message ""One or more analyses resulted in warnings!"

[Charlotte Xiong]

Dear LPA team,

Thanks for all of the amazing work you have done with "tidyLPA"! I am a new learner for R and LPA. I got one question when using the "MplusAutomation" function in tidyLPA package.

May I know the meaning of the warning message "One or more analyses resulted in warnings! Examine these analyses carefully: model_1_class_4, model_1_class_5, model_1_class_6" ? Does this mean that the model does not converge in the 4-cluster, 5-cluster, and 6-cluster solutions? But it seems that there is no issue with the corresponding model fit indices. May I ask what exactly is the issues mentioned in the warning message? BTW, I am using the no missing case data file of PISA.

Kindly check the attached code

LPA_Psy<-China%>%

• select(JOYSCIE,INTBRSCI,INSTSCIE,SCIEEFF,EPIST)%>%
• estimate_profiles(1:6, variances = "equal",covariances = "zero",package = "Mplusautomation")
  Warning message:

One or more analyses resulted in warnings! Examine these analyses carefully: model_1_class_4, model_1_class_5, model_1_class_6

LPA_Psy
tidyLPA analysis using mplus:

Model Classes AIC BIC Entropy prob_min prob_max n_min n_max BLRT_p
1 1 122927.31 122998.79 1.00 1.00 1.00 1.00 1.00
1 2 117545.48 117659.84 0.86 0.84 0.98 0.14 0.86 0.00
1 3 114762.64 114919.89 0.91 0.91 0.97 0.14 0.68 0.00
1 4 111129.19 111329.32 0.96 0.97 0.99 0.03 0.61 0.00
1 5 110191.32 110434.33 0.95 0.88 0.99 0.03 0.61 0.00
1 6 108947.28 109233.17 0.96 0.87 0.99 0.03 0.58 0.00