ConQuest Community

Help desk => Questions and Answers => Topic started by: cucho_veas on April 17, 2020, 12:06:05 PM

Title: EAP/PV reliability estimates
Post by: cucho_veas on April 17, 2020, 12:06:05 PM
Hi, I'm comparing two models, a unidimensional Rating scale Rasch model and a multidimensional RSRM. When running the URSRM, I'm trying to obtain the EAP/PV reliability but Conquest says they are unavailable. The basic syntax is:

Title URSRM;
datafile Scale.txt ;
format responses 1-32;
codes 1,2,3,4,5;
model item + step;
estimate !abilities =yes;
Title: Re: EAP/PV reliability estimates
Post by: dan_c on April 17, 2020, 12:27:03 PM
You are presumably using default scoring, such that your “lowest” response category is scored 1 (the lowest value of your code statement). You need a zero-scored response category, and you do that using a score statement:

Code: [Select]
score (1,2,3,4,5) (0,1,2,3,4);

See:  https://conquestmanual.netlify.app/s4-00.html#score (https://conquestmanual.netlify.app/s4-00.html#score)
Title: Re: EAP/PV reliability estimates
Post by: cucho_veas on April 21, 2020, 07:33:11 AM
Thanks, now it works. However, when using multidimensional syntax, the EAP/PV are available with the original 1-5 response categories. Why is that not possible with both models?
Title: Re: EAP/PV reliability estimates
Post by: dan_c on April 30, 2020, 04:54:29 AM
I'm not sure I follow - you would have to upload a minimum working example so I could recreate it.
https://conquest-forums.acer.edu.au/index.php?topic=1239.0 (https://conquest-forums.acer.edu.au/index.php?topic=1239.0)
Title: Re: EAP/PV reliability estimates
Post by: cucho_veas on May 01, 2020, 10:57:48 PM
This is the syntax I used:

Title model 1 factor;
datafile Scale.txt ;
format responses 1-32;
codes 1,2,3,4,5;
score (1,2,3,4,5) (0,1,2,3,4);
model item + step;
estimate !abilities =yes, method = Montecarlo;
reset;
Title model 4 factors;
Datafile Scale.txt;
Format responses 1-32;
codes 1,2,3,4,5;
score (1,2,3,4,5) (1,2,3,4,5) () () () !items (1,2,3,4,5,6,7,8,9);
score (1,2,3,4,5) () (1,2,3,4,5) () () !items (10,11,12,13,14,15);
score (1,2,3,4,5) () () (1,2,3,4,5) () !items (16,17,18,19,20,21,22,23);
score (1,2,3,4,5) () () () (1,2,3,4,5) !items (24,25,26,27,28,29,30,31,32);
Model item+step;
Estimate ! abilities =yes, method=Montecarlo, nodes =2000;
Title: Re: EAP/PV reliability estimates
Post by: dan_c on May 04, 2020, 06:21:37 AM
If you are comparing "model 1 factor" to "model 4 factors", I think it makes sense to impose the same scoring regime in both models (0,1,2,3,4).