Automated generation, running, and interpretation of moderated nonlinear factor analysis models for obtaining scores from observed variables. This package creates 'Mplus' input files which may be run iteratively to test two different types of covariate effects on items: (1) latent variable impact (both mean and variance); and (2) differential item functioning. After sequentially testing for all effects, it also creates a final model by including all significant effects after adjusting for multiple comparisons. Finally, the package creates a scoring model which uses the final values of parameter estimates to generate latent variable scores.
| Version: | 0.1 |
| Depends: | R (≥ 3.1.0) |
| Imports: | grDevices, graphics, stats, utils, ggplot2, MplusAutomation, reshape2, gridExtra, stringr, plyr, devtools |
| Published: | 2018-04-12 |
| Author: | Veronica Cole [aut, cre], Nisha Gottfredson [aut], Michael Giordano [aut], Tim Janssen [ctb] |
| Maintainer: | Veronica Cole <vcole at email.unc.edu> |
| License: | GPL-2 |
| NeedsCompilation: | no |
| CRAN checks: | aMNLFA results |
| Reference manual: | aMNLFA.pdf |
| Package source: | aMNLFA_0.1.tar.gz |
| Windows binaries: | r-devel: aMNLFA_0.1.zip, r-release: aMNLFA_0.1.zip, r-oldrel: aMNLFA_0.1.zip |
| OS X binaries: | r-release: aMNLFA_0.1.tgz, r-oldrel: aMNLFA_0.1.tgz |
Please use the canonical form https://CRAN.R-project.org/package=aMNLFA to link to this page.