In Pflegermeister/wizirt: A United API for IRT
title: r params$title
subtitle: r params$notes
author: r params$author
data: r params$data
date: r format(Sys.time(), '%B %d, %Y, %H:%M %p')
```{css, echo = FALSE, message = FALSE, warning = FALSE}
main .nav-pills > li.active > a,
main .nav-pills > li.active > a:hover,
main .nav-pills > li.active > a:focus {
background-color: #566D81;
}
main .nav-pills > li > a:hover {
background-color: #566D81;
}
h1, h2, h3, h4, h5, h6, legend {
color: #566D81;
}
nav-top span.glyphicon {
color: #566D81;
}
table-of-contents header {
color: #566D81;
}
table-of-contents h2 {
background-color: #566D81;
}
main a {
background-image: linear-gradient(180deg,#7896AE,#7896AE);
color: #7C7D7D;
}
a:hover {
color: #231E1F;
}
postamble {
color: #F4F5F5;
background: #020a23;
border-top: solid 10px #020a23;
font-family: "Lato","proxima-nova","Helvetica Neue",Arial,sans-serif;
font-size: 90%;
z-index: 400;
padding: 12px;
}
postamble p {
color: #F4F5F5;
font-size: 100%;
margin-bottom: 0;
}
body {
background-color: #fcfcf7;
}
content {
background: #fcfcf7;
max-width:1200px;
}
sidebar h2 {
z-index: 200;
background-color: #566D81;
text-align: center;
padding: 0.809em;
display: block;
color: #F4F5F5;
font-size: 100%;
margin-top: 0;
margin-bottom: 0.809em;
}
sidebar a {
color: #F4F5F5;
}
sidebar {
position: fixed;
top: 0;
overflow: hidden;
display: flex;
flex-direction: column;
height: 100%;
background: #020a23;
z-index: 200;
font-size: 16px;
}
p {
color:#201520;
}
img {
border-radius: 5px;
box-shadow: 3px 3px 10px #ccc;
}
.table-wrapper, .ReactTable {
padding: 7px;
border-radius: 5px;
background-color: white;
box-shadow: 3px 3px 10px #ccc;
}
pre {
box-shadow: 3px 3px 10px #ccc;
}
.logo {
position: absolute;
right:0;
width: 100px;
box-shadow: 0 0 0 #ccc;
}
```r
library(knitr)
library(rmdformats)
library(flextable)
library(reactable)
library(wizirt)
## Global options
options(max.print="75")
opts_chunk$set(echo=FALSE,
prompt=FALSE,
tidy=TRUE,
comment=NA,
message=FALSE,
warning=FALSE)
opts_knit$set(width=75)
if(is.null(params$data)){
cat('This report was generated using the practice data from the wizirt package')
data('responses')
data <- responses[,-1]
} else {
data <- readr::read_rds(params$data)
}
my_model <- wizirt(data = data, engine = params$engine, item_type = params$item_type, tol = params$tol, rownames = params$rownames)
cat(params$note)
Model Overview
Technical Information
A r my_model$fit$model$n_factors factor r my_model$fit$model$item_type model was estimated using the function r my_model$fit$model$engine$func from the package r my_model$fit$model$engine$pkg (v r my_model$fit$model$engine$ver). Estimation has r ifelse(my_model$fit$estimation$convergence, "", "not") converged using the r my_model$fit$estimation$method method after r my_model$fit$estimation$iterations iterations with a convergence criteria $r my_model$fit$estimation$criteria$.
Data Summary
Model built on a data set with r nrow(my_model$fit$data) examinees ($\bar{\theta}$ = r round(mean(my_model$fit$parameters$persons$ability, na.rm = T), 2), $\sigma_{\theta}$ = r round(sd(my_model$fit$parameters$persons$ability, na.rm = T), 2)) and r ncol(my_model$fit$data) items ($\bar{\delta}$ = r round(mean(my_model$fit$parameters$coefficients$difficulty, na.rm = T), 2), $\sigma_{\delta}$ = r round(sd(my_model$fit$parameters$coefficients$difficulty, na.rm = T), 2)).
Missing Data Summary
if(mean(is.na(my_model$fit$data)) == 0){
cat("No data were missing from this data set.")
} else {
print(my_model, type = 'na_item') %>% dplyr::filter(count != 0) %>% flextable() %>% autofit()
print(my_model, type = 'na_person') %>% dplyr::filter(count != 0) %>% flextable() %>% autofit()
}
if(mean(is.na(my_model$fit$data)) == 0){
cat("")
} else {
cat("## Description of Missing Data Handling
Generally, missing data are handled using full information during estimation. Additional functions for item and person fit tend to use list wise deletion or pairwise deletion.")
}
Assumptions {.tabset .tabset-fade .tabset-pills}
assumptions <- irt_assume(my_model)
Unidimensionality
print(assumptions, type = 'unid') %>% flextable() %>% autofit() %>% add_header_lines(c("Statistics Detecting Essential Unidimensionality", "Table 1")) %>% add_footer_lines(glue::glue("Statistics generated using sirt {packageVersion('sirt')}"))
Relative Fit
anova(my_model) %>% flextable() %>% autofit()
Local/Conditional Dependence
library(reactable)
assumptions$ld %>%
dplyr::mutate(dplyr::across(LD_std:ccov, round, 3)) %>%
dplyr::rename(`Item 1` = "item_1", `Item 2` = "item_2") %>%
reactable::reactable(filterable = T,
paginationType = "simple",
searchable = TRUE, columns = list(
pvals = colDef(style = function(value) {
if (value < .05) {
color <- "#9e0317"
} else {
color <- "#020a23"
}
list(color = color)
})
))
plot(my_model, type = 'ld_pairs')
Summary Plots {.tabset .tabset-fade .tabset-pills}
Test Information
plot(my_model, type = 'tinfo')
Theta Estimates and Item Difficulties
plot(my_model, type = 'theta_diff') + ggplot2::aes(color = type) + ggplot2::scale_color_manual(values = c("#020a23", "#566D81"))
Items
Item-level Statistics
ifa <- irt_item_fit(my_model, stats = params$ifa_stats)
print(ifa) %>% dplyr::mutate(dplyr::across(-item, round, 2)) %>%
reactable::reactable(paginationType = "simple",
searchable = TRUE)
Item-level Plots {.tabset .tabset-fade .tabset-pills}
Item Characteristic Curves and Residuals
plot(my_model, type = 'resid_trace') + ggplot2::coord_cartesian(xlim = c(-3,3))
Item Characteristic Curves and Observed Responses
plot(my_model, type = 'obs_trace') + ggplot2::coord_cartesian(xlim = c(-3,3))
Standardized Residuals
plot(my_model, type = 'stand') + ggplot2::coord_cartesian(xlim = c(-3,3))
Item Information
plot(my_model, type = 'info') + ggplot2::coord_cartesian(xlim = c(-3,3))
Persons
Person-level Plots {.tabset .tabset-fade .tabset-pills}
Districution of Person Abilities
plot(my_model, type = 'theta_SE')
Person Response Functions {.tabset .tabset-fade .tabset-pills}
persons_per = 30
out <- plot_wrap(object = my_model, type = 'np_prf', persons_per = persons_per)
for (i in seq_len(length(out))){
numbers <- seq_len(persons_per)+(i-1)*persons_per
numbers <- numbers[numbers %in% seq_len(nrow(my_model$fit$data))]
cat(glue::glue("#### Persons {min(numbers)}:{max(numbers)}
"))
print(out[[i]])
cat("
")
}
Person-level Statistics
Person-level Statistics
pfa <- irt_person_fit(my_model, stats = params$pfa_stats)
print(pfa, patterns = T, item_order = 'by_diff') %>%
dplyr::relocate(ids) %>%
dplyr::mutate(dplyr::across(c(-ids, -pattern), round, 2)) %>%
reactable::reactable(paginationType = "simple",
searchable = TRUE,
columns = list(
pattern = colDef(minWidth = 300)))
Quantitative Explanation of Person Misfit
pfa_mlm <- irt_model_pfa(my_model, pfa = pfa, predictors = params$pfa_predictors)
eval(parse(text = paste0("pfa_mlm$icc %>% as.data.frame() %>% tibble::rownames_to_column('ICC') %>% dplyr::mutate(dplyr::across(tidyselect::all_of(params$pfa_stats), function(x) round(as.numeric(x),3) )) %>%
reactable::reactable(columns = list(", paste( glue::glue('{params$pfa_stats} = colDef(maxWidth = 100)'), collapse = ',
'), "))")
)
)
mods <- lapply(1:length(pfa_mlm$models), function(x) {
pfa_mlm$models[[x]] %>% summary() %>% coef() %>% round(2) %>% as.data.frame() %>% tibble::rownames_to_column() %>% dplyr::mutate(Model = names(pfa_mlm$models)[x])
})
dplyr::bind_rows(mods) %>% reactable::reactable()
References
Estimation performed by the following packages:
-
Chung Y, Rabe-Hesketh S, Dorie V, Gelman A, Liu J (2013). A nondegenerate penalized likelihood estimator for variance parameters in multilevel models. Psychometrika, 78(4), 685-709.
-
Dimitris Rizopoulos (2006). ltm: An R package for Latent Variable Modelling and Item Response Theory Analyses, Journal of Statistical Software, 17 (5), 1-25. URL http://www.jstatsoft.org/v17/i05/
-
Jorge N. Tendeiro, Rob R. Meijer, A. Susan M. Niessen (2016). PerFit: An R Package for Person-Fit Analysis in IRT. Journal of Statistical Software, 74(5), 1-27. doi:10.18637/jss.v074.i05
-
Mair, P., Hatzinger, R., & Maier M. J. (2020). eRm: Extended Rasch Modeling. 1.0-1. https://cran.r-project.org/package=eRm
-
R. Philip Chalmers (2012). mirt: A Multidimensional Item Response Theory Package for the R Environment. Journal of Statistical Software, 48(6), 1-29. doi:10.18637/jss.v048.i06
-
Robitzsch, A. (2020). sirt: Supplementary Item Response Theory Models. R package version 3.9-4. https://CRAN.R-project.org/package=sirt
Powered by wizirt 
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.
title: r params$title
subtitle: r params$notes
author: r params$author
data: r params$data
date: r format(Sys.time(), '%B %d, %Y, %H:%M %p')
```{css, echo = FALSE, message = FALSE, warning = FALSE}
main .nav-pills > li.active > a,
main .nav-pills > li.active > a:hover,
main .nav-pills > li.active > a:focus {
background-color: #566D81;
}
main .nav-pills > li > a:hover {
background-color: #566D81;
}
h1, h2, h3, h4, h5, h6, legend { color: #566D81; }
nav-top span.glyphicon {
color: #566D81;
}
table-of-contents header {
color: #566D81;
}
table-of-contents h2 {
background-color: #566D81;
}
main a {
background-image: linear-gradient(180deg,#7896AE,#7896AE); color: #7C7D7D;
}
a:hover { color: #231E1F; }
postamble {
color: #F4F5F5; background: #020a23; border-top: solid 10px #020a23; font-family: "Lato","proxima-nova","Helvetica Neue",Arial,sans-serif; font-size: 90%; z-index: 400; padding: 12px;
}
postamble p {
color: #F4F5F5; font-size: 100%; margin-bottom: 0; }
body { background-color: #fcfcf7; }
content {
background: #fcfcf7; max-width:1200px;
}
sidebar h2 {
z-index: 200; background-color: #566D81; text-align: center; padding: 0.809em; display: block; color: #F4F5F5; font-size: 100%; margin-top: 0; margin-bottom: 0.809em;
}
sidebar a {
color: #F4F5F5;
}
sidebar {
position: fixed; top: 0; overflow: hidden; display: flex; flex-direction: column; height: 100%; background: #020a23; z-index: 200; font-size: 16px;
}
p { color:#201520; }
img { border-radius: 5px; box-shadow: 3px 3px 10px #ccc; }
.table-wrapper, .ReactTable { padding: 7px; border-radius: 5px; background-color: white; box-shadow: 3px 3px 10px #ccc; }
pre {
box-shadow: 3px 3px 10px #ccc;
}
.logo { position: absolute; right:0; width: 100px; box-shadow: 0 0 0 #ccc; }
```r
library(knitr)
library(rmdformats)
library(flextable)
library(reactable)
library(wizirt)
## Global options
options(max.print="75")
opts_chunk$set(echo=FALSE,
prompt=FALSE,
tidy=TRUE,
comment=NA,
message=FALSE,
warning=FALSE)
opts_knit$set(width=75)
if(is.null(params$data)){ cat('This report was generated using the practice data from the wizirt package') data('responses') data <- responses[,-1] } else { data <- readr::read_rds(params$data) }
my_model <- wizirt(data = data, engine = params$engine, item_type = params$item_type, tol = params$tol, rownames = params$rownames)
cat(params$note)
Model Overview
Technical Information
A r my_model$fit$model$n_factors factor r my_model$fit$model$item_type model was estimated using the function r my_model$fit$model$engine$func from the package r my_model$fit$model$engine$pkg (v r my_model$fit$model$engine$ver). Estimation has r ifelse(my_model$fit$estimation$convergence, "", "not") converged using the r my_model$fit$estimation$method method after r my_model$fit$estimation$iterations iterations with a convergence criteria $r my_model$fit$estimation$criteria$.
Data Summary
Model built on a data set with r nrow(my_model$fit$data) examinees ($\bar{\theta}$ = r round(mean(my_model$fit$parameters$persons$ability, na.rm = T), 2), $\sigma_{\theta}$ = r round(sd(my_model$fit$parameters$persons$ability, na.rm = T), 2)) and r ncol(my_model$fit$data) items ($\bar{\delta}$ = r round(mean(my_model$fit$parameters$coefficients$difficulty, na.rm = T), 2), $\sigma_{\delta}$ = r round(sd(my_model$fit$parameters$coefficients$difficulty, na.rm = T), 2)).
Missing Data Summary
if(mean(is.na(my_model$fit$data)) == 0){ cat("No data were missing from this data set.") } else { print(my_model, type = 'na_item') %>% dplyr::filter(count != 0) %>% flextable() %>% autofit() print(my_model, type = 'na_person') %>% dplyr::filter(count != 0) %>% flextable() %>% autofit() }
if(mean(is.na(my_model$fit$data)) == 0){ cat("") } else { cat("## Description of Missing Data Handling Generally, missing data are handled using full information during estimation. Additional functions for item and person fit tend to use list wise deletion or pairwise deletion.") }
Assumptions {.tabset .tabset-fade .tabset-pills}
assumptions <- irt_assume(my_model)
Unidimensionality
print(assumptions, type = 'unid') %>% flextable() %>% autofit() %>% add_header_lines(c("Statistics Detecting Essential Unidimensionality", "Table 1")) %>% add_footer_lines(glue::glue("Statistics generated using sirt {packageVersion('sirt')}"))
Relative Fit
anova(my_model) %>% flextable() %>% autofit()
Local/Conditional Dependence
library(reactable) assumptions$ld %>% dplyr::mutate(dplyr::across(LD_std:ccov, round, 3)) %>% dplyr::rename(`Item 1` = "item_1", `Item 2` = "item_2") %>% reactable::reactable(filterable = T, paginationType = "simple", searchable = TRUE, columns = list( pvals = colDef(style = function(value) { if (value < .05) { color <- "#9e0317" } else { color <- "#020a23" } list(color = color) }) ))
plot(my_model, type = 'ld_pairs')
Summary Plots {.tabset .tabset-fade .tabset-pills}
Test Information
plot(my_model, type = 'tinfo')
Theta Estimates and Item Difficulties
plot(my_model, type = 'theta_diff') + ggplot2::aes(color = type) + ggplot2::scale_color_manual(values = c("#020a23", "#566D81"))
Items
Item-level Statistics
ifa <- irt_item_fit(my_model, stats = params$ifa_stats) print(ifa) %>% dplyr::mutate(dplyr::across(-item, round, 2)) %>% reactable::reactable(paginationType = "simple", searchable = TRUE)
Item-level Plots {.tabset .tabset-fade .tabset-pills}
Item Characteristic Curves and Residuals
plot(my_model, type = 'resid_trace') + ggplot2::coord_cartesian(xlim = c(-3,3))
Item Characteristic Curves and Observed Responses
plot(my_model, type = 'obs_trace') + ggplot2::coord_cartesian(xlim = c(-3,3))
Standardized Residuals
plot(my_model, type = 'stand') + ggplot2::coord_cartesian(xlim = c(-3,3))
Item Information
plot(my_model, type = 'info') + ggplot2::coord_cartesian(xlim = c(-3,3))
Persons
Person-level Plots {.tabset .tabset-fade .tabset-pills}
Districution of Person Abilities
plot(my_model, type = 'theta_SE')
Person Response Functions {.tabset .tabset-fade .tabset-pills}
persons_per = 30 out <- plot_wrap(object = my_model, type = 'np_prf', persons_per = persons_per) for (i in seq_len(length(out))){ numbers <- seq_len(persons_per)+(i-1)*persons_per numbers <- numbers[numbers %in% seq_len(nrow(my_model$fit$data))] cat(glue::glue("#### Persons {min(numbers)}:{max(numbers)} ")) print(out[[i]]) cat(" ") }
Person-level Statistics
Person-level Statistics
pfa <- irt_person_fit(my_model, stats = params$pfa_stats) print(pfa, patterns = T, item_order = 'by_diff') %>% dplyr::relocate(ids) %>% dplyr::mutate(dplyr::across(c(-ids, -pattern), round, 2)) %>% reactable::reactable(paginationType = "simple", searchable = TRUE, columns = list( pattern = colDef(minWidth = 300)))
Quantitative Explanation of Person Misfit
pfa_mlm <- irt_model_pfa(my_model, pfa = pfa, predictors = params$pfa_predictors) eval(parse(text = paste0("pfa_mlm$icc %>% as.data.frame() %>% tibble::rownames_to_column('ICC') %>% dplyr::mutate(dplyr::across(tidyselect::all_of(params$pfa_stats), function(x) round(as.numeric(x),3) )) %>% reactable::reactable(columns = list(", paste( glue::glue('{params$pfa_stats} = colDef(maxWidth = 100)'), collapse = ', '), "))") ) ) mods <- lapply(1:length(pfa_mlm$models), function(x) { pfa_mlm$models[[x]] %>% summary() %>% coef() %>% round(2) %>% as.data.frame() %>% tibble::rownames_to_column() %>% dplyr::mutate(Model = names(pfa_mlm$models)[x]) }) dplyr::bind_rows(mods) %>% reactable::reactable()
References
Estimation performed by the following packages:
-
Chung Y, Rabe-Hesketh S, Dorie V, Gelman A, Liu J (2013). A nondegenerate penalized likelihood estimator for variance parameters in multilevel models. Psychometrika, 78(4), 685-709.
-
Dimitris Rizopoulos (2006). ltm: An R package for Latent Variable Modelling and Item Response Theory Analyses, Journal of Statistical Software, 17 (5), 1-25. URL http://www.jstatsoft.org/v17/i05/
-
Jorge N. Tendeiro, Rob R. Meijer, A. Susan M. Niessen (2016). PerFit: An R Package for Person-Fit Analysis in IRT. Journal of Statistical Software, 74(5), 1-27. doi:10.18637/jss.v074.i05
-
Mair, P., Hatzinger, R., & Maier M. J. (2020). eRm: Extended Rasch Modeling. 1.0-1. https://cran.r-project.org/package=eRm
-
R. Philip Chalmers (2012). mirt: A Multidimensional Item Response Theory Package for the R Environment. Journal of Statistical Software, 48(6), 1-29. doi:10.18637/jss.v048.i06
-
Robitzsch, A. (2020). sirt: Supplementary Item Response Theory Models. R package version 3.9-4. https://CRAN.R-project.org/package=sirt
Powered by wizirt
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.