Traditional item analysis / covariance matrix not positive definite

[jonmerl]

Hello,

I just started working with Conquest (4.5.2). I ran a model with one dimension and a model with two dimensions.

Regarding the descriptive statistics, the frequencies of right and wrong answers (0,1) are slightly different than the frequencies in the data file. Often a difference of one case per item.
I am not sure if I did a mistake with the import or in the syntax.

Moreover, the twodimensional model breaks up. Error message: "Covariance matrix is not positive definite". Is this a problem of the syntax or is it that this model can not be estimated with this data.

I would be very grateful for your help.
 
Thank you,
Jonas

[dan_c]

Hi Jonas,

your issues is with the data file you have crated. It is encoded UTF8 with a byte order marker (BOM). Early versions of ConQuest 4 did not support this encoding. I would recommend creating files with a BOM - it is not part of the Unicode standard, and only seems to appear with MS Excel when exporting files.

ConQuest 5 will attempt to strip the BOM when reading the file. I would encourage you to upgrade: https://shop.acer.edu.au/acer-conquest-5.html

Note that in ConQuest 5, you can read and write CSV files directly and do not need to create a text (.dat) or SPSS file.

Attached is a converted data file (I did this with VS Code - which is free to use and an excellent text editor) and output files. Note the issue with the covariance matrix is still present in your 2D model, I would guess it is to do with the very high correlation between your dimensions (>0.95), the very narrow variance (~0.1-0.3) and small sample size. Note the model deviance implies the 1D model is a better fit anyway.

[jonmerl]

Thank you very much for your helpful and quick response.
This forum is really great.

Just one question:
Is it possible to remove the data files from the posts, so the problem and solution would still be comprehensible for the community, only the data is not available. This would be a bit calming for me.

Thank you for your response.
Best wishes.

[dan_c]

of course - I have removed the attachments.