Skip to contents

Plot Single Interaction Effect

Source: R/PlotInteractionEffectsContinuous.R
PlotInteractionEffectsContinuous.Rd

Creates a scatter plot with regression lines showing the interaction between a predictor and outcome variable, moderated by either a continuous or categorical variable.

Usage

PlotInteractionEffectsContinuous(
  data,
  interVar = NULL,
  outcome_var = NULL,
  predictor_var = NULL,
  covariates = NULL,
  n_lines = 3,
  alpha = 0.6,
  point_size = 2,
  Data = lifecycle::deprecated(),
  outcomeVar = lifecycle::deprecated(),
  predictorVar = lifecycle::deprecated(),
  covars = lifecycle::deprecated()
)

Arguments

data

A data frame containing the variables to be analyzed

interVar

Character string specifying the interaction variable (moderator)

outcome_var

Character string specifying the outcome variable

predictor_var

Character string specifying the predictor variable

covariates

Character vector of covariate names to include in the model

n_lines

For continuous moderators, number of lines to plot (default: 3 for low/med/high)

alpha

Transparency level for points (default: 0.6)

point_size

Size of points (default: 2)

Data

Deprecated (since 19.15.0). Use data instead.

outcomeVar

Deprecated (since 19.15.0). Use outcome_var instead.

predictorVar

Deprecated (since 19.15.0). Use predictor_var instead.

covars

Deprecated (since 19.15.0). Use covariates instead.

Value

A ggplot object showing the interaction effect

Details

For categorical moderators, separate regression lines are plotted for each category. For continuous moderators, regression lines are plotted at the mean and +/- 1 SD.

The subtitle shows the p-value for the interaction term. The caption lists any covariates included in the model.

Variable labels are used if available in the data frame.

Examples

data(SampleData)
data(SampleVariableTypes)

# Attach labels and factor levels for readable axis titles and legend
Labelled <- RevalueData(SampleData, SampleVariableTypes)$RevaluedData

# Categorical moderator: one regression line per Diagnosis group
PlotInteractionEffectsContinuous(
  Labelled,
  interVar = "Diagnosis",
  outcome_var = "AXL",
  predictor_var = "age"
)
#> `geom_smooth()` using formula = 'y ~ x'


# Continuous moderator: lines at mean and +/- 1 SD of the moderator
PlotInteractionEffectsContinuous(
  Labelled,
  interVar = "Adiponectin",
  outcome_var = "AXL",
  predictor_var = "age"
)
#> `geom_smooth()` using formula = 'y ~ x'