Duplicate warning in TidyLPA

[Smale-Jacobse, A.E.]

Dear all,

One more rookie-question... In my analyses with the TidyLPA package I keep getting warnings about duplicate rows (that are not used in the analyses). It is not clear to me why duplicates are found; the row names (column names in Excel) obviously vary. 

I tried to get around this problem by using random numbering as rownames (by using the command ->  rownames (data) = rep (1:2066) ) but I still get this warning. Any suggestions?

Kind regards, Annemieke Smale

[Joshua Rosenberg]

Hi Annemieke, thank you for this question. I wonder what function is returning this warning. Is there any way you could walk through line-by-line and tell us which function is associated with it? 

Thank you!

Josh

[J.D. Haltigan]

are sysmis (missing values) all specified correctly as N/A in data frame and are headers used in data frame?

Would check those potential trigger points as first step.

Best regards,

J.D.

--
You received this message because you are subscribed to the Google Groups "tidyLPA" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tidylpa+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/tidylpa/5531b094-17aa-40f7-931d-7033751acefa%40googlegroups.com.

[Hannah...@gmx.de]

Hi Annemieke,

Did you manage to fix the problem? I am encountering the same problem and I cannot find any information on how to deal with it anywhere.

If you did manage to fix it, do you have any tips for me?

Kind regards,

Hannah Peetz

[Caspar van Lissa]

What code is producing the error?

[Hannah...@gmx.de]

I was following the example from the Introduction to tidy LPA but none of the models worked.

I am basically just loading an excel file and then try to run the code with the relevant variables. It does not matter whether I omit all the missing values or not, I still get the same warnings.

Code:

data1 <- read.csv2('onlyr.csv')

data1%>%

  select(av_SeverityBullied, ZreceivedNom_Bullied)%>%

  single_imputation() %>%

  estimate_profiles(1:6)

Warning messages:

1: In data.row.names(row.names, rowsi, i) :

  some row.names duplicated: 4 --> row.names NOT used

2: In data.row.names(row.names, rowsi, i) :

  some row.names duplicated: 4,5,6,7,8 --> row.names NOT used

3: In data.row.names(row.names, rowsi, i) :

  some row.names duplicated: 4,5,6,7,8,9,10,11,12 --> row.names NOT used

4: In data.row.names(row.names, rowsi, i) :

  some row.names duplicated: 4,5,6,7,8,9,10,11,12,13,14,15,16 --> row.names NOT used

5: In data.row.names(row.names, rowsi, i) :

  some row.names duplicated: 4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20 --> row.names NOT used

6: In data.row.names(row.names, rowsi, i) :

  some row.names duplicated: 4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24 --> row.names NOT used

7: 

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

[Caspar van Lissa]

This is not an error originating in tidyLPA. When I google your error, I find: https://stackoverflow.com/questions/50962364/error-in-row-names-data-frametmp-value-value-duplicate-row-names

[Hannah...@gmx.de]

I already reduced my dataset to contain only three colums: participant number and my two variables. I still get the same error. And there are no hidden columns (I think) since I am using R studio and it clearly says I only have three columns. So I don't think that that is the problem. 

However, when trying again this morning I noticed a warning message when loading the tidyLPA package:

Warning message:

running command '"where" mplus.exe' had status 1 

Not sure if this has anything to do with it though.

[Caspar van Lissa]

That warning message shouldn't have anything to do with it. It's not an error code, and it arises when we're checking if you have Mplus installed.

Can you run your code one line at a time to check which line gives the error?

[Hannah...@gmx.de]

Yeah I also figured that the MPLus warning wasn't relevant.

How can I run the code one line at a time? Since I am using piping, R automatically runs the whole function.

[Caspar van Lissa]

I don't think I can help with this; like I said, it's not an error in tidyLPA. Please see this manual for how to run one line at a time https://eeecon.uibk.ac.at/~discdown/rprogramming/first-steps-in-rstudio.html

[Hanna...@gmx.de]

I managed to get rid of the warning by updating R to a newer version. Now that the code actually runs, I get new warning messages:

1: In mclustBootstrapLRT(full_data, modelName = mod_name,  ... :

  only 1-component model could be fitted. No LRT is performed!

2: In mclustBootstrapLRT(full_data, modelName = mod_name,  ... :

  only 1-component model could be fitted. No LRT is performed!

3: Mclust could not estimate model 2 with 2 classes.

4: Mclust could not estimate model 2 with 3 classes.

5: Mclust could not estimate model 2 with 4 classes.

6: Mclust could not estimate model 2 with 5 classes.

7: Mclust could not estimate model 2 with 6 classes.

8: Mclust could not estimate model 6 with 2 classes.

9: Mclust could not estimate model 6 with 3 classes.

10: Mclust could not estimate model 6 with 4 classes.

11: Mclust could not estimate model 6 with 5 classes.

12: Mclust could not estimate model 6 with 6 classes.

13: 

One or more analyses resulted in warnings! Examine these analyses carefully: model_1_class_4, model_1_class_5, model_1_class_6, model_2_class_2, model_2_class_3, model_2_class_4, model_2_class_5, model_2_class_6, model_3_class_4, model_3_class_5, model_3_class_6, model_6_class_2, model_6_class_3, model_6_class_4, model_6_class_5, model_6_class_6

It basically cannot run model 2 and 6, and with model 1 and 3 it has difficulties when I want to estimate 4 or more classes.

I do get results for model 1 and 3 with 1-3 classes. 

[Caspar van Lissa]

Something unusual about your data, may be likert-scale, or too few observations. I'd really recommend involving a statistician to help run and interpret the analysis!