powRICLPM
Perform Power Analysis for the RI-CLPM and STARTS Model
Perform user-friendly power analyses for the random
intercept cross-lagged panel model (RI-CLPM) and the bivariate stable trait
autoregressive trait state (STARTS) model. The strategy as proposed by
Mulder (2023)
powRICLPM
powRICLPM is an R package that aids researchers with performing a
power analysis for the random intercept cross-lagged panel model
(RI-CLPM) by Hamaker, Kuiper, and Grasman (2015), and the Stable Trait
Autoregressive Trait State Model (STARTS) by Kenny and Zautra (1995) and
Kenny and Zautra (2001). It implements the strategy as proposed by
Mulder (2023). Its main functionalities include:
- Basic power
analysis:
Use Monte Carlo simulations to compute the power to reject the
null-hypothesis (as well as other performance measures such as bias,
mean square error) for all parameters in the RI-CLPM and STARTS, for a
specific experimental condition. A condition is defined by its sample
size, number of repeated measures, proportion of between-unit
variance, and reliability of the indicators.
powRICLPMcan perform power analyses across multiple experimental conditions simultaneously, and report the results back in a user-friendly manner. - Extensions: The basic power analysis setup can be extended to include the use of bounded estimation, various (stationarity) constraints over time on parameters of the estimation model, the generation of nonnormal data, among other things.
- Mplus:
When Mplus is installed,
powRICLPMcan create Mplus syntax, and run the power analyses in Mplus.
Documentation
There are four sources of documentation for powRICLPM:
- The rationale for the power analysis strategy underlying this package can be found in Mulder (2023).
- Every user-facing function in the package is documented, and the
documentation can be accessed by running
?function_namein the R console (e.g.,?powRICLPM). Here, you can find explanations on how to use the functions, as well as technical details. - More elaborate descriptions of this package’s functionality and
analysis options are described in vignettes. These are accessible via
the ‘Vignettes’ tab in the menu, or via R using
vignette(package = "powRICLPM"). - The FAQ contains answers to frequently asked question that reach me via email.
Installation
To install the development version of powRICLPM, including the latest
bug fixes and new features, run:
install.packages("devtools")
devtools::install_github("jeroendmulder/powRICLPM")
To install the latest release of powRICLPM from CRAN, run:
install.packages("powRICLPM")
Citing powRICLPM
You can cite the R-package with the following citation:
Mulder, J.D., (2023). Power analysis for the random intercept cross-lagged panel model using the powRICLPM R-package. Structural Equation Modeling: A Multidisciplinary Journal, 30(4), 645-658. https://doi.org/10.1080/10705511.2022.2122467
Contact
If you have ideas, comments, or issues you would like to raise, please get in touch.
- Issues and ideas can be raised on GitHub via https://github.com/jeroendmulder/powRICLPM
- Pull request can be created on GitHub via https://github.com/jeroendmulder/powRICLPM/pulls
References
Hamaker, Ellen L., Rebecca M. Kuiper, and Raoul P. P. P. Grasman. 2015. “A critique of the cross-lagged panel model.” Psychological Methods 20 (1): 102–16. https://doi.org/10.1037/a0038889.
Kenny, David A., and Alex Zautra. 1995. “The trait-state-error model for multiwave data.” Journal of Consulting and Clinical Psychology1 63 (1): 52–59.
———. 2001. “Trait–state models for longitudinal data.” In New Methods for the Analysis of Change, 243–63. Washington: American Psychological Association. https://doi.org/10.1037/10409-008.
Mulder, Jeroen D. 2023. “Power Analysis for the Random Intercept Cross-Lagged Panel Model Using the powRICLPM r-Package.” Structural Equation Modeling: A Multidisciplinary Journal 30 (4): 645–58. https://doi.org/10.1080/10705511.2022.2122467.