Power Analysis for Parameter Estimation in Structural Equation Modeling: A Discussion and Tutorial

This repo contains code files referenced in the following paper:

Wang, Y. A., & Rhemtulla, M. (in press). Power analysis for parameter estimation in structural equation modeling: A discussion and tutorial. Advances in Methods and Practices in Psychological Science.

pwrSEM

pwrSEM is a Shiny web app developed by Y. Andre Wang for power analysis for parameter estimation in structural equation modeling.

Code

Source code of the app is available at: pwrSEM.R. Users can run pwrSEM locally in R Studio by downloading the source code file, opening it in R Studio, and then pressing “Run App” (procedure current as of R Studio version 1.2.5001).

Code for the simulation studies reported in the paper is available at: SimulationCode.R.

Version History

Version 0.1.3 (released TBD):

• Fixed App crashes when estimating power of parameter with value = 1 (issue #5)
• Fixed App crashes when automatically setting residual variances (issue #6)

Version 0.1.2 (released February 23, 2021):

• Fixed an error where residual variances could not be calculated even when user-entered values were on standardized metric and produced a positive definite model-implied covariance matrix
• Fixed an error in power estimation where the population values were misrepresented internally, resulting in inaccurate power estimates

Version 0.1.1 (released March 31, 2020): Initial release