news.md
In Pflegermeister/wizirt: A United API for IRT
wizirt 0.1.2
Major Changes
The wizirt package has now been released. It is pretty new. Let me know
if there are any issues!
As of right now, it is known that the degrees of freedom used in the
anova() function when mirt is the engine are 1 larger than when mirt
itself is used.
wizirt 0.0.0.0
Package Overview
The wizirt package (currently named wizirt2) is up and running, it seems
to work for different data sets. It has a limited functionality, but it
is currently working.
The wizirt package represents the work of my dissertation. As such I
have agreed to do the following for my dissertation:
- Provide general model information including:
- Model call
- Description of model run
- Software and version
- Estimator
- Convergence status, criteria, values
- Estimation issues
- Response data described (sample size, demographics)
- Descriptions of items and test
- Missing data summary
- Description of missing data handling
- Provide model fit information
- Unidimensionality (DETECT)
- Absolute fit
- Relative fit
- Anova method for comparing models
- Provide item information
- Parameter estimates
- Summary statistics
- Provide item-fit information
- Conditional dependence
- Functional form (isn’t this model-data fit?)
- Fit statistics
- predicted vs observed item response functions
- misfitting items flagged
- Provide person information
- Person location estimates (including SE)
- Provide person-fit information
- Global detection (Ht, lz*, infit and outfit, with cutoffs where
applicable)
- Local detection (ICI or other)
- Tabulation or presentation of response patterns (winsteps tables
as guides)
- PRF (nonparametric, parametric may follow later)
- MLM (Reise, Conijn)
- Two levels of accessibility
- Comprehensive report for beginners
- Felxible functions for advanced users
- Learning Resources
- Github website
- CRAN-style reference page
- tidyverse-style cheat sheet
- Karabatsos’s simulation replication article
- Quality Assurance
- Accuracy check
- Match IRTPRO, Winsteps
- Replicate Karabatsos
- Informative check (meet all the checks above two levels…)
- Usefulness check
- Usability testing
- Call it from python on separate laptop
- Aesthetics check
- Repetition
- Contrast
- Alignment
- Proximity
My goal is to complete all of this by October 24. That will give me a
week to write it all up by October 30 and give it to my chair. Then I
will have a week to improve it according to my chair’s recommendations
so I can send it to my committee by November 6.
Pflegermeister/wizirt documentation built on Nov. 22, 2020, 8:27 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
wizirt 0.1.2
Major Changes
The wizirt package has now been released. It is pretty new. Let me know if there are any issues!
As of right now, it is known that the degrees of freedom used in the anova() function when mirt is the engine are 1 larger than when mirt itself is used.
wizirt 0.0.0.0
Package Overview
The wizirt package (currently named wizirt2) is up and running, it seems to work for different data sets. It has a limited functionality, but it is currently working.
The wizirt package represents the work of my dissertation. As such I have agreed to do the following for my dissertation:
- Provide general model information including:
- Model call
- Description of model run
- Software and version
- Estimator
- Convergence status, criteria, values
- Estimation issues
- Response data described (sample size, demographics)
- Descriptions of items and test
- Missing data summary
- Description of missing data handling
- Provide model fit information
- Unidimensionality (DETECT)
- Absolute fit
- Relative fit
- Anova method for comparing models
- Provide item information
- Parameter estimates
- Summary statistics
- Provide item-fit information
- Conditional dependence
- Functional form (isn’t this model-data fit?)
- Fit statistics
- predicted vs observed item response functions
- misfitting items flagged
- Provide person information
- Person location estimates (including SE)
- Provide person-fit information
- Global detection (Ht, lz*, infit and outfit, with cutoffs where applicable)
- Local detection (ICI or other)
- Tabulation or presentation of response patterns (winsteps tables as guides)
- PRF (nonparametric, parametric may follow later)
- MLM (Reise, Conijn)
- Two levels of accessibility
- Comprehensive report for beginners
- Felxible functions for advanced users
- Learning Resources
- Github website
- CRAN-style reference page
- tidyverse-style cheat sheet
- Karabatsos’s simulation replication article
- Quality Assurance
- Accuracy check
- Match IRTPRO, Winsteps
- Replicate Karabatsos
- Informative check (meet all the checks above two levels…)
- Usefulness check
- Usability testing
- Call it from python on separate laptop
- Aesthetics check
- Repetition
- Contrast
- Alignment
- Proximity
- Accuracy check
My goal is to complete all of this by October 24. That will give me a week to write it all up by October 30 and give it to my chair. Then I will have a week to improve it according to my chair’s recommendations so I can send it to my committee by November 6.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.