README.md
In DJAnderson07/r2Winsteps: A package for interfacing between R and the Rasch modeling software Winsteps
Overview
The r2Winsteps package was developed to provide a convenient interface
between R and the Rasch modeling software Winsteps
(http://www.winsteps.com). The package is not intended to encompass the full
capabilities of Winsteps, but rather to provide a simple framework for
estimating many commonly applied models. There are also a few extensions to
Winsteps that are used within R.
The primary features of the package include the ability to:
- Write control and data files for Winsteps with the
r2Winsteps() function,
which includes automatic detection of polytomous item types. Either the
rating scale or partial credit models can be estimated.
- Run Winsteps directly from R with the
runWinsteps() function, which writes
and executes a .bat file to call Winsteps. Item, person, and structure
(if a polytmous model is estimated) parameters are returned in a list.
Intermediary files (control, data, output, etc.) can be stored or discarded
(default).
- Batch run a set of models with the
batchRunWinsteps() files. Essentially
calls runWinsteps(), but takes as its argument a list of
data frames, with a different model fit and parameter estimates returned for
each data frame in the list. At present, batchRunWinsteps() only works for
dichotomous items.
- Both
runWinsteps() and batchRunWinsteps() can include anchor files to
anchor specific items at specific values. The function write.anchor() can also be used to help produce anchor files, given a previously estimated ifile and a new dataset to analyze.
- Numerous default plotting options, as well as fully user-manipulable plot
functionality.
For the primary motivations behind the development of the package, see the
overview vignette. Note that the package is still in active development.
Use at your own risk. If you find any bugs, please email them to
daniela@uoregon.edu, or log them at
https://github.com/DJAnderson07/r2Winsteps/issues.
Installation
library(devtools)
install_github("DJAnderson07/r2Winsteps")
Citation info
citation(package = "r2Winsteps")
##
## To cite package 'r2Winsteps' in publications use:
##
## Daniel Anderson (2015). r2Winsteps: A package for interfacing
## between R and the Rasch modeling software Winsteps. R package
## version 0.0.0.9000. https://github.com/DJAnderson07/r2Winsteps
##
## A BibTeX entry for LaTeX users is
##
## @Manual{,
## title = {r2Winsteps: A package for interfacing between R and the Rasch modeling software
## Winsteps},
## author = {Daniel Anderson},
## year = {2015},
## note = {R package version 0.0.0.9000},
## url = {https://github.com/DJAnderson07/r2Winsteps},
## }
Basic usage
We will analyze the science data and look at a few plots.
library(r2Winsteps)
data(science)
head(science)
## item1 item2 item3 item4 item5 item6 item7 item8 item9 item10 item11
## 1 1 2 1 1 1 0 2 0 1 2 2
## 2 2 2 2 2 2 2 2 2 2 2 2
## 3 2 2 1 1 0 1 1 0 1 2 2
## 4 1 0 1 0 0 1 0 1 2 2 1
## 5 1 0 1 0 1 0 1 0 0 1 1
## 6 1 0 1 1 2 1 1 0 1 1 1
## item12 item13 item14 item15 item16 item17 item18 item19 item20 item21
## 1 2 2 0 2 1 1 2 2 0 2
## 2 2 2 2 2 2 2 2 2 2 2
## 3 2 2 1 2 2 1 2 2 1 2
## 4 2 2 1 1 1 1 2 2 0 2
## 5 1 1 0 0 1 1 2 2 1 1
## 6 2 1 0 1 0 1 2 2 1 0
## item22 item23 item24 item25 Sex LastName FirstName MiddleName
## 1 1 0 2 0 M Rossner Marc Daniel
## 2 2 2 2 2 M Rossner Lawrence F.
## 3 1 1 1 1 M Rossner Toby G.
## 4 1 1 1 1 M Rossner Michael T.
## 5 1 1 1 0 F Rossner Rebecca A.
## 6 1 2 1 0 M Rossner Tr Cat
Notice that these data are polytomous, which will be detected automatically.
Andrich's Rating Scale model will be fit by default, but we can also fit
Masters' Partial Credit Model, as shown below. To fit the model, we split the
dataset into person demographics, and item responses, then call runWinsteps()
# Split data
itemsSci <- science[ ,1:25]
demosSci <- science[ ,26:ncol(science)]
# Run model
parsSci <- runWinsteps(itms = itemsSci, dems = demosSci, partialCredit = TRUE)
All the parameters have been estimated with Winsteps, and have been read back into R, including the partial credit structure file.
str(parsSci)
## List of 3
## $ ItemParameters :'data.frame': 25 obs. of 22 variables:
## ..$ Entry : int [1:25] 1 2 3 4 5 6 7 8 9 10 ...
## ..$ Difficulty : num [1:25] -0.89 -0.61 0.28 1.8 2.19 0.16 0.99 1.64 0.6 -1.57 ...
## ..$ Status : int [1:25] 1 1 1 1 1 1 1 1 1 1 ...
## ..$ Count : num [1:25] 75 75 75 75 75 75 75 75 75 75 ...
## ..$ RawScore : num [1:25] 109 116 88 52 37 91 69 54 80 130 ...
## ..$ SE : num [1:25] 0.23 0.2 0.2 0.21 0.22 0.2 0.19 0.21 0.18 0.26 ...
## ..$ Infit : num [1:25] 0.73 0.75 0.66 0.99 2.28 0.88 0.97 1.17 1.11 0.78 ...
## ..$ Infit_Z : num [1:25] -1.93 -1.49 -2.63 -0.03 5.28 -0.79 -0.17 1.14 0.83 -1.09 ...
## ..$ Outfit : num [1:25] 0.67 0.56 0.62 0.98 3.98 0.82 1 1.23 1.13 0.56 ...
## ..$ Outfit_Z : num [1:25] -1.83 -1.44 -2.51 -0.05 6.98 -1.01 0.08 1.41 0.76 -1.14 ...
## ..$ Displacement : num [1:25] 0 0 0 0 0 0 0 0 0 0 ...
## ..$ PointMeasureCorr : num [1:25] 0.64 0.58 0.72 0.6 0.05 0.61 0.59 0.51 0.53 0.5 ...
## ..$ Weight : num [1:25] 1 1 1 1 1 1 1 1 1 1 ...
## ..$ ObservMatch : num [1:25] 74.3 73 74.3 71.6 52.7 68.9 52.7 45.9 56.8 78.4 ...
## ..$ ExpectMatch : num [1:25] 66.5 62.5 60.4 61.5 68.7 60.3 55.2 60.2 54.5 77 ...
## ..$ PointMeasureExpected: num [1:25] 0.48 0.46 0.55 0.59 0.63 0.54 0.59 0.6 0.57 0.38 ...
## ..$ RMSR : num [1:25] 0.43 0.5 0.47 0.54 0.81 0.55 0.6 0.61 0.67 0.4 ...
## ..$ WMLE : num [1:25] -0.89 -0.6 0.28 1.8 2.18 0.17 0.99 1.64 0.6 -1.55 ...
## ..$ Group : int [1:25] 0 0 0 0 0 0 0 0 0 0 ...
## ..$ Model : Factor w/ 1 level " R": 1 1 1 1 1 1 1 1 1 1 ...
## ..$ Recoding : Factor w/ 1 level " .": 1 1 1 1 1 1 1 1 1 1 ...
## ..$ ItemID : Factor w/ 25 levels " item1"," item10",..: 1 12 19 20 21 22 23 24 25 2 ...
## $ PersonParameters:'data.frame': 75 obs. of 22 variables:
## ..$ Entry : int [1:75] 1 2 3 4 5 6 7 8 9 10 ...
## ..$ Theta : num [1:75] 0.49 5.98 0.99 0.14 -0.77 -0.2 2.63 0.86 -0.2 0.26 ...
## ..$ Status : int [1:75] 1 0 1 1 1 1 1 1 1 1 ...
## ..$ Count : num [1:75] 25 25 25 25 25 25 25 25 25 25 ...
## ..$ RawScore : num [1:75] 30 50 34 27 19 24 44 33 24 28 ...
## ..$ SE : num [1:75] 0.35 1.84 0.36 0.34 0.34 0.34 0.48 0.36 0.34 0.34 ...
## ..$ Infit : num [1:75] 0.96 1 0.46 0.73 0.81 1.61 1.85 1.01 1.42 0.74 ...
## ..$ Infit_Z : num [1:75] -0.07 0 -2.29 -1.04 -0.71 2.08 1.93 0.13 1.54 -0.98 ...
## ..$ Outfit : num [1:75] 0.84 1 0.42 0.66 1.4 2.51 1.18 0.85 2.01 0.66 ...
## ..$ Outfit_Z : num [1:75] -0.31 0 -1.44 -1 1.2 3.44 0.5 -0.21 2.53 -0.95 ...
## ..$ Displacement : num [1:75] 0 0 0 0 0 0 0 0 0 0 ...
## ..$ PointMeasureCorr : num [1:75] 0 0 0 0 0 0 0 0 0 0 ...
## ..$ Weight : num [1:75] 1 1 1 1 1 1 1 1 1 1 ...
## ..$ ObservMatch : num [1:75] 68 100 88 76 64 56 76 64 52 64 ...
## ..$ ExpectMatch : num [1:75] 61.7 100 63.4 59.9 60 59.1 79.3 63.9 59.1 61 ...
## ..$ PointMeasureExpected: num [1:75] 0 0 0 0 0 0 0 0 0 0 ...
## ..$ RMSR : num [1:75] 0.56 0 0.38 0.5 0.53 0.75 0.57 0.56 0.71 0.5 ...
## ..$ WMLE : num [1:75] 0.48 5.98 0.98 0.14 -0.77 -0.2 2.57 0.85 -0.2 0.25 ...
## ..$ Sex : Factor w/ 2 levels " F "," M ": 2 2 2 2 1 2 2 2 2 2 ...
## ..$ LastName : Factor w/ 62 levels " Airehead ",..: 48 48 48 48 48 48 62 31 53 27 ...
## ..$ FirstName : Factor w/ 66 levels " Alan ",..: 36 33 61 39 49 63 7 53 38 17 ...
## ..$ MiddleName : Factor w/ 19 levels ""," A."," Baby",..: 5 7 9 18 2 4 1 19 1 16 ...
## $ StructureFiles :'data.frame': 74 obs. of 3 variables:
## ..$ Item : int [1:74] 1 1 1 2 2 2 3 3 3 4 ...
## ..$ Category: int [1:74] 0 1 2 0 1 2 0 1 2 0 ...
## ..$ delta : num [1:74] 0 -1.64 1.64 0 -0.3 0.3 0 -1.13 1.13 0 ...
## - attr(*, "class")= chr "r2Winsteps"
Test information function
We can view the test information function by just calling plot on the object
plot(parsSci)
Notice that the regions where reliability is 0.8 and 0.7 are shaded by default.
Item category probabilities
Next let's look at the item category probabilities for the first six items.
par(mfrow = c(3, 2))
plot(parsSci, type = "ICP", itemSelect = 1:6)
Let's view some more plots
Item information function
plot(parsSci, type = "IIFs")
Test characteristic curve
plot(parsSci, type = "TCC")
Item characteristic curves
plot(parsSci, type = "ICCs")
Item-Person density
plot(parsSci, type = "ipDens")
Thresholds: First 6 items
par(mfrow = c(3, 2))
plot(parsSci, type = "thresholds", itemSelect = 1:6)
There are also other plots and batch processing capabilities. The software is under active development and is released "as is". For more complete documentation see the vignettes or contact the author.
DJAnderson07/r2Winsteps documentation built on Sept. 12, 2019, 1:06 a.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.
Overview
The r2Winsteps package was developed to provide a convenient interface between R and the Rasch modeling software Winsteps (http://www.winsteps.com). The package is not intended to encompass the full capabilities of Winsteps, but rather to provide a simple framework for estimating many commonly applied models. There are also a few extensions to Winsteps that are used within R.
The primary features of the package include the ability to:
- Write control and data files for Winsteps with the
r2Winsteps()function, which includes automatic detection of polytomous item types. Either the rating scale or partial credit models can be estimated. - Run Winsteps directly from R with the
runWinsteps()function, which writes and executes a.batfile to call Winsteps. Item, person, and structure (if a polytmous model is estimated) parameters are returned in a list. Intermediary files (control, data, output, etc.) can be stored or discarded (default). - Batch run a set of models with the
batchRunWinsteps()files. Essentially callsrunWinsteps(), but takes as its argument a list of data frames, with a different model fit and parameter estimates returned for each data frame in the list. At present,batchRunWinsteps()only works for dichotomous items. - Both
runWinsteps()andbatchRunWinsteps()can include anchor files to anchor specific items at specific values. The functionwrite.anchor()can also be used to help produce anchor files, given a previously estimated ifile and a new dataset to analyze. - Numerous default plotting options, as well as fully user-manipulable plot functionality.
For the primary motivations behind the development of the package, see the overview vignette. Note that the package is still in active development. Use at your own risk. If you find any bugs, please email them to daniela@uoregon.edu, or log them at https://github.com/DJAnderson07/r2Winsteps/issues.
Installation
library(devtools)
install_github("DJAnderson07/r2Winsteps")
Citation info
citation(package = "r2Winsteps")
##
## To cite package 'r2Winsteps' in publications use:
##
## Daniel Anderson (2015). r2Winsteps: A package for interfacing
## between R and the Rasch modeling software Winsteps. R package
## version 0.0.0.9000. https://github.com/DJAnderson07/r2Winsteps
##
## A BibTeX entry for LaTeX users is
##
## @Manual{,
## title = {r2Winsteps: A package for interfacing between R and the Rasch modeling software
## Winsteps},
## author = {Daniel Anderson},
## year = {2015},
## note = {R package version 0.0.0.9000},
## url = {https://github.com/DJAnderson07/r2Winsteps},
## }
Basic usage
We will analyze the science data and look at a few plots.
library(r2Winsteps)
data(science)
head(science)
## item1 item2 item3 item4 item5 item6 item7 item8 item9 item10 item11
## 1 1 2 1 1 1 0 2 0 1 2 2
## 2 2 2 2 2 2 2 2 2 2 2 2
## 3 2 2 1 1 0 1 1 0 1 2 2
## 4 1 0 1 0 0 1 0 1 2 2 1
## 5 1 0 1 0 1 0 1 0 0 1 1
## 6 1 0 1 1 2 1 1 0 1 1 1
## item12 item13 item14 item15 item16 item17 item18 item19 item20 item21
## 1 2 2 0 2 1 1 2 2 0 2
## 2 2 2 2 2 2 2 2 2 2 2
## 3 2 2 1 2 2 1 2 2 1 2
## 4 2 2 1 1 1 1 2 2 0 2
## 5 1 1 0 0 1 1 2 2 1 1
## 6 2 1 0 1 0 1 2 2 1 0
## item22 item23 item24 item25 Sex LastName FirstName MiddleName
## 1 1 0 2 0 M Rossner Marc Daniel
## 2 2 2 2 2 M Rossner Lawrence F.
## 3 1 1 1 1 M Rossner Toby G.
## 4 1 1 1 1 M Rossner Michael T.
## 5 1 1 1 0 F Rossner Rebecca A.
## 6 1 2 1 0 M Rossner Tr Cat
Notice that these data are polytomous, which will be detected automatically.
Andrich's Rating Scale model will be fit by default, but we can also fit
Masters' Partial Credit Model, as shown below. To fit the model, we split the
dataset into person demographics, and item responses, then call runWinsteps()
# Split data
itemsSci <- science[ ,1:25]
demosSci <- science[ ,26:ncol(science)]
# Run model
parsSci <- runWinsteps(itms = itemsSci, dems = demosSci, partialCredit = TRUE)
All the parameters have been estimated with Winsteps, and have been read back into R, including the partial credit structure file.
str(parsSci)
## List of 3
## $ ItemParameters :'data.frame': 25 obs. of 22 variables:
## ..$ Entry : int [1:25] 1 2 3 4 5 6 7 8 9 10 ...
## ..$ Difficulty : num [1:25] -0.89 -0.61 0.28 1.8 2.19 0.16 0.99 1.64 0.6 -1.57 ...
## ..$ Status : int [1:25] 1 1 1 1 1 1 1 1 1 1 ...
## ..$ Count : num [1:25] 75 75 75 75 75 75 75 75 75 75 ...
## ..$ RawScore : num [1:25] 109 116 88 52 37 91 69 54 80 130 ...
## ..$ SE : num [1:25] 0.23 0.2 0.2 0.21 0.22 0.2 0.19 0.21 0.18 0.26 ...
## ..$ Infit : num [1:25] 0.73 0.75 0.66 0.99 2.28 0.88 0.97 1.17 1.11 0.78 ...
## ..$ Infit_Z : num [1:25] -1.93 -1.49 -2.63 -0.03 5.28 -0.79 -0.17 1.14 0.83 -1.09 ...
## ..$ Outfit : num [1:25] 0.67 0.56 0.62 0.98 3.98 0.82 1 1.23 1.13 0.56 ...
## ..$ Outfit_Z : num [1:25] -1.83 -1.44 -2.51 -0.05 6.98 -1.01 0.08 1.41 0.76 -1.14 ...
## ..$ Displacement : num [1:25] 0 0 0 0 0 0 0 0 0 0 ...
## ..$ PointMeasureCorr : num [1:25] 0.64 0.58 0.72 0.6 0.05 0.61 0.59 0.51 0.53 0.5 ...
## ..$ Weight : num [1:25] 1 1 1 1 1 1 1 1 1 1 ...
## ..$ ObservMatch : num [1:25] 74.3 73 74.3 71.6 52.7 68.9 52.7 45.9 56.8 78.4 ...
## ..$ ExpectMatch : num [1:25] 66.5 62.5 60.4 61.5 68.7 60.3 55.2 60.2 54.5 77 ...
## ..$ PointMeasureExpected: num [1:25] 0.48 0.46 0.55 0.59 0.63 0.54 0.59 0.6 0.57 0.38 ...
## ..$ RMSR : num [1:25] 0.43 0.5 0.47 0.54 0.81 0.55 0.6 0.61 0.67 0.4 ...
## ..$ WMLE : num [1:25] -0.89 -0.6 0.28 1.8 2.18 0.17 0.99 1.64 0.6 -1.55 ...
## ..$ Group : int [1:25] 0 0 0 0 0 0 0 0 0 0 ...
## ..$ Model : Factor w/ 1 level " R": 1 1 1 1 1 1 1 1 1 1 ...
## ..$ Recoding : Factor w/ 1 level " .": 1 1 1 1 1 1 1 1 1 1 ...
## ..$ ItemID : Factor w/ 25 levels " item1"," item10",..: 1 12 19 20 21 22 23 24 25 2 ...
## $ PersonParameters:'data.frame': 75 obs. of 22 variables:
## ..$ Entry : int [1:75] 1 2 3 4 5 6 7 8 9 10 ...
## ..$ Theta : num [1:75] 0.49 5.98 0.99 0.14 -0.77 -0.2 2.63 0.86 -0.2 0.26 ...
## ..$ Status : int [1:75] 1 0 1 1 1 1 1 1 1 1 ...
## ..$ Count : num [1:75] 25 25 25 25 25 25 25 25 25 25 ...
## ..$ RawScore : num [1:75] 30 50 34 27 19 24 44 33 24 28 ...
## ..$ SE : num [1:75] 0.35 1.84 0.36 0.34 0.34 0.34 0.48 0.36 0.34 0.34 ...
## ..$ Infit : num [1:75] 0.96 1 0.46 0.73 0.81 1.61 1.85 1.01 1.42 0.74 ...
## ..$ Infit_Z : num [1:75] -0.07 0 -2.29 -1.04 -0.71 2.08 1.93 0.13 1.54 -0.98 ...
## ..$ Outfit : num [1:75] 0.84 1 0.42 0.66 1.4 2.51 1.18 0.85 2.01 0.66 ...
## ..$ Outfit_Z : num [1:75] -0.31 0 -1.44 -1 1.2 3.44 0.5 -0.21 2.53 -0.95 ...
## ..$ Displacement : num [1:75] 0 0 0 0 0 0 0 0 0 0 ...
## ..$ PointMeasureCorr : num [1:75] 0 0 0 0 0 0 0 0 0 0 ...
## ..$ Weight : num [1:75] 1 1 1 1 1 1 1 1 1 1 ...
## ..$ ObservMatch : num [1:75] 68 100 88 76 64 56 76 64 52 64 ...
## ..$ ExpectMatch : num [1:75] 61.7 100 63.4 59.9 60 59.1 79.3 63.9 59.1 61 ...
## ..$ PointMeasureExpected: num [1:75] 0 0 0 0 0 0 0 0 0 0 ...
## ..$ RMSR : num [1:75] 0.56 0 0.38 0.5 0.53 0.75 0.57 0.56 0.71 0.5 ...
## ..$ WMLE : num [1:75] 0.48 5.98 0.98 0.14 -0.77 -0.2 2.57 0.85 -0.2 0.25 ...
## ..$ Sex : Factor w/ 2 levels " F "," M ": 2 2 2 2 1 2 2 2 2 2 ...
## ..$ LastName : Factor w/ 62 levels " Airehead ",..: 48 48 48 48 48 48 62 31 53 27 ...
## ..$ FirstName : Factor w/ 66 levels " Alan ",..: 36 33 61 39 49 63 7 53 38 17 ...
## ..$ MiddleName : Factor w/ 19 levels ""," A."," Baby",..: 5 7 9 18 2 4 1 19 1 16 ...
## $ StructureFiles :'data.frame': 74 obs. of 3 variables:
## ..$ Item : int [1:74] 1 1 1 2 2 2 3 3 3 4 ...
## ..$ Category: int [1:74] 0 1 2 0 1 2 0 1 2 0 ...
## ..$ delta : num [1:74] 0 -1.64 1.64 0 -0.3 0.3 0 -1.13 1.13 0 ...
## - attr(*, "class")= chr "r2Winsteps"
Test information function
We can view the test information function by just calling plot on the object
plot(parsSci)
Notice that the regions where reliability is 0.8 and 0.7 are shaded by default.
Item category probabilities
Next let's look at the item category probabilities for the first six items.
par(mfrow = c(3, 2))
plot(parsSci, type = "ICP", itemSelect = 1:6)
Let's view some more plots
Item information function
plot(parsSci, type = "IIFs")
Test characteristic curve
plot(parsSci, type = "TCC")
Item characteristic curves
plot(parsSci, type = "ICCs")
Item-Person density
plot(parsSci, type = "ipDens")
Thresholds: First 6 items
par(mfrow = c(3, 2))
plot(parsSci, type = "thresholds", itemSelect = 1:6)
There are also other plots and batch processing capabilities. The software is under active development and is released "as is". For more complete documentation see the vignettes or contact the author.
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.