int.plot: Interaction plot

In Conigrave: Flexible Tools for Multiple Imputation

Description

Calculates a standardized two way or three way interaction and plots using ggplot2.

Usage

int.plot(data, outcome, predictor, moderator, y.lim = c(-1, 1),
  x.lim = c(-1, 1), x.lab = "auto", y.lab = "auto", title = "auto",
  title.size = 15, SDs = 1, legend.name = "auto",
  colour = "ghostwhite", show.points = FALSE, save = F,
  path = getwd())

Arguments

data	an object of class 'data.frame' or 'imputationList'.
outcome	a string with the name of the outcome variable.
predictor	a string with the name of the predictor variable.
moderator	a vector of the names of up to two moderating variables.
y.lim	vector of numerals indicating y axis bounds.
x.lim	vector of numerals indicating x axis bounds.
x.lab	a string with the label of the x axis.
y.lab	a string with the label of the y axis.
title	a string containing title text.
title.size	a numeral containing the font size of the title.
SDs	a numeral indicating the standard deviations of the moderators.
legend.name	a character string indicating the title of the legend.
colour	a character string containing the colour of the data points.
show.points	logical to determine whether or not to include points.
save	logical as to whether or not to save the plot.
path	string containing path of where to save plot. Defaults to working directory.

Value

A ggplot

Examples

carsdata<-mtcars
int.plot(carsdata,"mpg","disp","cyl", y.lim = c(-2.5,2.5))
int.plot(carsdata,"mpg","disp", c("cyl","am"), y.lim = c(-5.0,2.0))

Example output

Call:
lm(formula = scale(data[, outcome]) ~ scale(data[, moderator]) * 
    scale(data[, predictor]))

Residuals:
     Min       1Q   Median       3Q      Max 
-0.67711 -0.26637 -0.04891  0.17498  0.96203 

Coefficients:
                       Estimate Std. Error t value Pr(>|t|)    
(Intercept)            -0.50879    0.17692  -2.876 0.007619 ** 
scale(cyl)              0.07485    0.25034   0.299 0.767156    
scale(disp)            -0.97535    0.25180  -3.873 0.000589 ***
scale(cyl)*scale(disp)  0.58225    0.18171   3.204 0.003369 ** 
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 0.4413 on 28 degrees of freedom
Multiple R-squared:  0.8241,	Adjusted R-squared:  0.8052 
F-statistic: 43.72 on 3 and 28 DF,  p-value: 1.078e-10


Call:
lm(formula = scale(data[, outcome]) ~ scale(data[, moderator[1]]) * 
    scale(data[, moderator[2]]) * scale(data[, predictor]))

Residuals:
     Min       1Q   Median       3Q      Max 
-0.54507 -0.27873 -0.03879  0.22627  0.81471 

Coefficients:
                                 Estimate Std. Error t value Pr(>|t|)   
(Intercept)                       -0.5244     0.1858  -2.822  0.00944 **
scale(cyl)                         0.1553     0.2780   0.559  0.58159   
scale(am)                         -0.5448     0.1999  -2.725  0.01180 * 
scale(disp)                       -0.8752     0.3198  -2.737  0.01149 * 
scale(cyl)*scale(am)               0.7635     0.3073   2.485  0.02033 * 
scale(cyl)*scale(disp)             0.3522     0.2175   1.619  0.11854   
scale(am)*scale(disp)             -0.9613     0.3425  -2.806  0.00978 **
scale(cyl)*scale(am)*scale(disp)   0.6823     0.2147   3.178  0.00405 **
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 0.3903 on 24 degrees of freedom
Multiple R-squared:  0.882,	Adjusted R-squared:  0.8476 
F-statistic: 25.64 on 7 and 24 DF,  p-value: 1.144e-09

Conigrave documentation built on May 1, 2019, 7:31 p.m.