sspLNIRT is a sample size planning tool for item calibration with the Joint Hierarchical Model (JHM) of response accuracy and response time. It estimates the minimum sample size required to achieve a target accuracy (RMSE) of item parameter estimates under a specified data-generating process.
The package provides:
- Precomputed results for various design conditions, accessible instantly via a Shiny app and/or R package.
-
Custom Sample Size Estimation via
optim_sample()for design conditions outside the precomputed data. - Visualization functions for inspecting parameter accuracy or bias, power curves, and implied response time and response accuracy distributions.
Usage
The tool is available as an R package and as an interactive Shiny app.
Web App
Use the app at sebastian-lortz.shinyapps.io/sspLNIRT.
Installation
You can install the latest version of the R package from GitHub:
# install devtools if needed
if (!requireNamespace("devtools")) {install.packages("devtools")}
# install from GitHub
devtools::install_github("sebastian-lortz/sspLNIRT")System Requirements
The sspLNIRT package was built under R version 4.5.2 using Apple clang version 16.0.0 (clang-1600.0.26.6) and GNU Fortran (GCC) 14.2.0. To compile R packages from source, install the appropriate toolchain:
- macOS: see https://mac.r-project.org/tools/
- Windows: see https://cran.r-project.org/bin/windows/Rtools/
Documentation
Vignettes and full function documentation are available at sebastian-lortz.github.io/sspLNIRT.
Citation
Please cite sspLNIRT if you use it. To cite the software, use:
Lortz S (2026). sspLNIRT: Sample Size Planning for Item Calibration using the Joint Hierarchical Model. R package version 0.0.0.9000, https://github.com/sebastian-lortz/sspLNIRT.
Or copy the reference information to your BibTeX file: