Daniel Lee posted on Saturday, May 19, 2012 - 10:24 pm
Hello! I'm trying to conduct a CFA on a 84-item measure. In the syntax, each item represents a variable. Unfortunately, with so many items, I am easily surpassing the 90 character limit. Do you have advice for conducting a CFA on an 84-item measure on MPlus? Thank you! -Daniel Lee
I am trying to adapt the code for item and scale reliability with binary items from Raykov et al. (2010) to my test, which however has 83 items. The resulting code apparently is too long, as I get the message "Statement exceeds line limit. Break your statement into several parts." What is the line limit in Mplus 7 and is there a way to solve this problem?
I am sorry if I was being unclear - the line limit, not the character limit, is exceeded. My Input file has about 900 lines, because the calculation of the scale and item reliability requires about 10 calculation steps per item (and I have 83 items) according to the Mplus code provided in the Raykov article.
Is there maybe a way of saving the values resulting from some of the calculations (for example, one intermediate step is the calculation of item mean true scores) as variables in the dataset, and then continuing the next calculation step with a new Input file, using the saved variables? Like this I could split the code into shorter parts, if the whole code is too long.
We impose no line limit. Please send the input file, data, and your license number to firstname.lastname@example.org so we can see where the error is coming from.
Tammy Tolar posted on Thursday, August 28, 2014 - 9:13 am
I am doing the same analyses (as Katja above) with 55 items and getting the same error message about exceeding the line limit. Based on reviewing her file, is there a generic answer to the problem? Thanks.
Hi, I too am getting the same message "Statement exceeds line limit. Break your statement into several parts." as my model constraint command has lots of variables following a CFA model. I don't exceed the character limit and you mentioned that there is no line limit in Mplus (I am using the latest version). Any idea as to what is causing this problem?
Margarita posted on Monday, October 16, 2017 - 3:58 am
Dear Dr. Muthen,
I am also getting the above message when running the reliability of an 18-binary-item scale (as proposed by Raykov et al,2010). I tried breaking the true variance of sum score equation into smaller parts and even though I no longer get that message I now get
STARTING VALUES FOR THE DEPENDENT PARAMETERS COULD NOT BE COMPUTED FROM THE STARTING VALUES FOR THE INDEPENDENT PARAMETERS. CHANGE THE STARTING VALUES FOR THE INDEPENDENT PARAMETERS.
THE MODEL ESTIMATION DID NOT TERMINATE NORMALLY DUE TO AN ERROR IN THE COMPUTATION. CHANGE YOUR MODEL AND/OR STARTING VALUES.
I checked the syntax multiple times and appears to be correct. Do you have ideas as to why this is happening?