Moderated Multiple Linear Regression in R

NOTE: An update to the lessR package has resulted in the following changes. First, the abbreviated output version of the Regression function is now called reg_brief (formerly reg.brief). Second, the argument to request standardized coefficients (based on z-score rescaling of the predictor and outcome variables) is now: new_scale="z" (formerly standardized=TRUE and rescale="z").
This tutorial demonstrates how to estimate a moderated multiple linear regression model in R, including how to assess and visualize differential prediction (predictive bias) by creating simple slopes plots.
As noted in the video [24:20], the jtools package has been replaced with the interaction package, and the probe_interaction function now belongs to the interactions package.
In terms of the context, I focus on the differential prediction of employee selection procedures (i.e., tools, assessments) and, in doing so, apply the three-step “step-up” process using moderated multiple linear regression models (Cleary, 1968; SIOP, 2003). This process uses what is referred to as hierarchical linear regression, as additional predictor variables are added at each subsequent step, and their incremental validity above and beyond the other predictors in the model is assessed; for an introduction to hierarchical linear regression, please see the following conceptual overview video ( • Hierarchical Linear Re... ) and tutorial ( • Hierarchical Linear Re... ).
Other videos from my "R Tutorials" playlist can be found here: • R Tutorials
The data file(s) referenced in this R tutorial (and other R tutorials on this playlist) can be downloaded as a compressed (zipped) folder by visiting the following link: github.com/davidcaughlin/R-Tu.... Click "Clone or Download" followed by "Download ZIP".