JNplots (version 0.1.2)

jnt_cont: Calculation and visualization of regions of non-significance to assess the influence of continuous moderators

Description

Produces a plot showing how changes in the moderator affect the slope and significance of the relationship between the dependent variable and the predictor.

Usage

jnt_cont(
  X,
  Y,
  m,
  data,
  alpha.sig = 0.05,
  correlation = NULL,
  res = 100,
  xlab = X,
  ylab = Y,
  col.gradient = TRUE,
  sig_color = "lightblue",
  nonsig_color = "grey",
  max_col_grad = "red",
  min_col_grad = "blue",
  legend = TRUE
)

Value

List with six elements: (1) results from the linear model, (2) lower and (3) upper limits of (non)significance in the moderator, (4) lower and (5) upper data limit in the data, and (6) a graphical output.

Arguments

X

A character string defining the name of the covariate (e.g., size in an allometry analysis). Must be the same as the name of the variable in the dataset (see argument “data”).

Y

A character string defining the name of the dependent variable. Must be the same as the name of the variable in the dataset (see argument “data”).

m

A character string defining the name of a continuous moderator. Must be the same as the name of the variable in the dataset (see argument “data”). The variable must be continuous.

data

A dataframe containing the variables in the model.

alpha.sig

A value representing the significance value (alpha) to be considered.

correlation

an optional corStruct object describing the within-group correlation structure. See the documentation of corClasses for a description of the available corStruct classes. If a grouping variable is to be used, it must be specified in the form argument to the corStruct constructor. Defaults to NULL, corresponding to uncorrelated errors.

res

A numerical value that aids in the plotting of regions of (non)significance. Default=100, higher numbers increase the number of fitted regression lines plotted (N=res-1).

xlab

A title for the X axis. Defaults to the name of the predictor variable in the data.

ylab

A title for the Y axis. Defaults to the name of the dependent variable in the data.

col.gradient

A logical indicating whether the significant regression lines should be plotted with a gradient of colors representing moderator values. Defaults to 'TRUE'.

sig_color

If col.gradient = FALSE, a character string indicating the color of the significant regression lines. Defaults to 'lightblue'.

nonsig_color

If col.gradient = FALSE, a character string indicating the color of the non-significant regression lines. Defaults to 'grey'.

max_col_grad

If col.gradient = TRUE, a character string indicating the maximum color of the gradient.

min_col_grad

If col.gradient = TRUE, a character string indicating the minimum color of the gradient.

legend

A logical indicating whether a legend should appear on top of the plot. Defaults to 'TRUE'.

References

Toyama, K. S. (2024). JNplots: an R package to visualize outputs from the Johnson-Neyman technique for categorical and continuous moderators, including options for phylogenetic regressions. Evolutionary Ecology 38: 371–385.

Examples

#### non-phylogenetic model ####
data(lizard_home_range)
jnt_cont(X='PHR95_overlap_z', Y='hrsize95', m='degree_z',
data=lizard_home_range, xlab = 'home range overlap 95',
ylab='home range size 95')

#### phylogenetic model ####
jnt_cont(X='bio12', Y='back_bright', m='bio1', data=bird_colors,
correlation=corPagel(1, tree_Furnariidae),xlab='precipitation (mm)',
ylab='back brightness (scaled)',res=200)