1
Show Posts
This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.
2
ConQuest News / Understanding Rasch Measurement Theory
« on: June 17, 2020, 07:34:31 AM »
The next iteration of Understanding Rasch Measurement Theory starts on Monday 13 July 2020. This is a 100% online, 10 week, part-time, masters level course.
Led by world-renowned researcher and psychometrician Professor Geoff Masters, this masters level course will unlock the practical understanding you need to get the most from Rasch.
Find out more about enrolments, course outline and contents, and what to expect from an ACER online course below.
https://www.acer.org/au/professional-learning/postgraduate/rasch
Led by world-renowned researcher and psychometrician Professor Geoff Masters, this masters level course will unlock the practical understanding you need to get the most from Rasch.
Find out more about enrolments, course outline and contents, and what to expect from an ACER online course below.
https://www.acer.org/au/professional-learning/postgraduate/rasch
3
Questions and Answers / Re: Very different WLE Person separation reliability and EAP/PV reliability
« on: May 26, 2020, 04:11:45 AM »
Thanks for the files.
I'd strongly encourage you to upgrade to a more recent version of ConQuest (version 5 is new this year).
Your model seems to run fine and converges happily. The estimated reliabilities are the same in a new build of ConQuest (attached).
The WLE reliability is low primarily because there are very few items in dimensions 2 and 3. Reviewing your WLE file, there are many cases where the max possible score for dimension 2 and 3 is 4.00. This can be interpreted as meaning there is little information available to space cases out along the latent trait. Your PV reliability is getting a boost because of the multidimenisfal structure. Your latent traits are highly correlated (> 0.8 ) and provide a lot of information that reduces the uncertainty in dimension 2 and 3. I think it's fair to conclude that this data is more useful for estimation of population parameters than individual abilities.
I'd strongly encourage you to upgrade to a more recent version of ConQuest (version 5 is new this year).
Your model seems to run fine and converges happily. The estimated reliabilities are the same in a new build of ConQuest (attached).
The WLE reliability is low primarily because there are very few items in dimensions 2 and 3. Reviewing your WLE file, there are many cases where the max possible score for dimension 2 and 3 is 4.00. This can be interpreted as meaning there is little information available to space cases out along the latent trait. Your PV reliability is getting a boost because of the multidimenisfal structure. Your latent traits are highly correlated (> 0.8 ) and provide a lot of information that reduces the uncertainty in dimension 2 and 3. I think it's fair to conclude that this data is more useful for estimation of population parameters than individual abilities.
4
Questions and Answers / Re: Very different WLE Person separation reliability and EAP/PV reliability
« on: May 21, 2020, 07:31:51 AM »
we will need to see you data and outputs to check this. Can you also let us know what version of ConQuest you are running?
Alternatively, send through a reproducible example: https://conquest-forums.acer.edu.au/index.php?topic=1239.0
Note, you don't technically need your recode AND your score statements, see: https://conquestmanual.netlify.app/s4-00.html#score
This is will give your items 1, 5, and 30, properties like in the ordered partition model.
Alternatively, send through a reproducible example: https://conquest-forums.acer.edu.au/index.php?topic=1239.0
Note, you don't technically need your recode AND your score statements, see: https://conquestmanual.netlify.app/s4-00.html#score
Code: [Select]
codes 0,1,2,3;
score (0,1,2,3) (0,0,0.5,1) ( ) ( ) !items (1);
score (0,1) (0,1) ( ) ( ) !items (2,6,7,8,12,16,19,20,23,25,26,31,32,35,39,42,43,44,45,48);
score (0,1) (0,1) ( ) ( ) !items (4,17,18,34,38,46,49);
score (0,1,2) (0,0.5,1) ( ) ( ) !items (11,47);
score (0,1) (0,1) ( ) ( ) !items (10,21,24,33,37);
score (0,1,2) ( ) (0,0,1) ( ) !items (5);
score (0,1) ( ) (0,1) ( ) !items (3,5,9,14,27,36,40);
score (0,1) ( ) ( ) (0,1) !items (13,15,22,28,29,30,41);
score (0,1,2) ( ) ( ) (0,0,1) !items (30);
This is will give your items 1, 5, and 30, properties like in the ordered partition model.
5
Questions and Answers / Re: EAP/PV reliability estimates
« 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).
6
Questions and Answers / Re: EAP/PV reliability estimates
« 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
7
ConQuest News / ACER ConQuest Version 5: available now!
« on: April 30, 2020, 04:51:55 AM »
Version 5 of ACER ConQuest is now available.
Version 5 is a major upgrade on version 4 and includes new features including:
Version 5 is a major upgrade on version 4 and includes new features including:
- Bayesian estimation using Markov Chain Monte Carlo (MCMC)
- Integration with R through the conquestr library
8
Questions and Answers / Re: standard error of conditional covariances/correlations in multidimensional run?
« on: April 27, 2020, 01:17:50 AM »
you can extract the full error variance-covariance matrix and include in your Show file. See here:
https://conquestmanual.netlify.app/s4-00.html#show
You need to:
1. use empirical errors
2. add the option to your show command
https://conquestmanual.netlify.app/s4-00.html#show
You need to:
1. use empirical errors
Code: [Select]
estimate ! ... stderr = empirical ...;This is the default behaviour if you do not specify the stderr option.2. add the option to your show command
Code: [Select]
show ! ... tables = 10 ...; Table 10 is the error variance-covariance matrix
9
Questions and Answers / Re: EAP/PV reliability estimates
« 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:
See: https://conquestmanual.netlify.app/s4-00.html#score
Code: [Select]
score (1,2,3,4,5) (0,1,2,3,4);See: https://conquestmanual.netlify.app/s4-00.html#score
10
Questions and Answers / Re: set directory
« on: December 19, 2019, 07:35:38 AM »Do you just copy and paste the desired directory in there?
yes, you can copy and paste a path into the `set` command.
11
Questions and Answers / Re: set directory
« on: November 27, 2019, 10:43:44 AM »
ConQuest will set the working directory to the path where the executable was called from. This is normal and useful behaviour for command like tools. If you are using the GUI version, the working directory will be the location of the executable.
To change the working directory, you can use the `set` command:
To change the working directory, you can use the `set` command:
Code: [Select]
set directory = “c:\user\”; will set the working directory to be the user folder on your drive mapped to the letter “c”. You can confirm this by printing the working directory: Code: [Select]
dir;
12
Questions and Answers / Re: Release constrained parameter
« on: November 26, 2019, 12:55:53 AM »
No, the constraint of the last step parameter to be the negative sum of the rest is a necessary identification constraint for the model.
An easier way to interpret the rating scale or partial credit model you are estimating would be to look at the Thurstonian Thresholds (e.g., table 7 in the Show: `show ! tables = 1:2:3:4:7;`). In your example of it item scored 0, 1, 2, there will be 2 thresholds, labeled x.1 and x.2 (where x is your item numbers). Each threshold can be interpreted as the location on the latent trait at which the probability of achieving score >= 1 is > 0.5 (x.1) and score >=2 is 0.5 (x.2).
An easier way to interpret the rating scale or partial credit model you are estimating would be to look at the Thurstonian Thresholds (e.g., table 7 in the Show: `show ! tables = 1:2:3:4:7;`). In your example of it item scored 0, 1, 2, there will be 2 thresholds, labeled x.1 and x.2 (where x is your item numbers). Each threshold can be interpreted as the location on the latent trait at which the probability of achieving score >= 1 is > 0.5 (x.1) and score >=2 is 0.5 (x.2).
13
Questions and Answers / Re: Some mistakes. Help me.
« on: November 07, 2019, 11:21:09 PM »
can you provide some more information?
what sort of mistake/error are you encoutnering?
Can you provide a minimum working example? https://conquest-forums.acer.edu.au/index.php?topic=1239.0
Can you also not upload in proprietary formats (rar) - if you want to upload files, it is best to upload small, minimum working examples as text files.
what sort of mistake/error are you encoutnering?
Can you provide a minimum working example? https://conquest-forums.acer.edu.au/index.php?topic=1239.0
Can you also not upload in proprietary formats (rar) - if you want to upload files, it is best to upload small, minimum working examples as text files.
14
Questions and Answers / Re: Asking about facet model with SPSS datafile
« on: September 06, 2019, 12:03:52 AM »
Hi Lan Anh,
You can specify a facet mode using an SPSS like this (based on your cqc file attached):
This will give you a partial credit with a main effect for rater (ObsID) harshness. You will need to carefully think about model identification if you want to fit more complex models, like then in your attached cqc file - I don't think you will have an observation of every rater using every response category for every item in order to include the item*step*ObsID interaction.
You can specify a facet mode using an SPSS like this (based on your cqc file attached):
Code: [Select]
datafile Obs.sav !
filetype=spss, responses=A1 to C6, keeps = ObsID;
...
model item+step+item*step+ObsID;This will give you a partial credit with a main effect for rater (ObsID) harshness. You will need to carefully think about model identification if you want to fit more complex models, like then in your attached cqc file - I don't think you will have an observation of every rater using every response category for every item in order to include the item*step*ObsID interaction.
15
Questions and Answers / Re: Multidimensional Within Item Partial Credit Model
« on: May 23, 2019, 12:22:32 AM »
check out:
Adams, R. J., Wilson, M., & Wang, W. (1997). The Multidimensional Random Coefficients Multinomial Logit Model. Applied Psychological Measurement, 21(1), 1–23. https://doi.org/10.1177/0146621697211001
Adams, R. J., Wilson, M., & Wang, W. (1997). The Multidimensional Random Coefficients Multinomial Logit Model. Applied Psychological Measurement, 21(1), 1–23. https://doi.org/10.1177/0146621697211001