Hello everyone,

I am currently dealing with psychometric evaluation of the croatian translation of a scale to assess chronic stress. The scale has nine dimensions and I am estimating PCM for every subscale separately using eRm-package in R.

I now want to conduct an Anderson LR-test with the function LRtest(). With some of the scales it has worked perfectly fine. However, for most of the dimensions I get an error message:

"No items with appropriate response patterns left to perform LR-test!".

I have encountered this message before, but I have always been able to still make it work, except for this time. Using other splitcriteria such as the mean or sex also does not seem to work and I keep getting the same error message. My sample is N= 257 and when I for example use the median there is approximately the same number of people in each group.

What could the reason for the error message be? I hope that someone of you can help me since I am really desperate.

Thank you and all the best,