DMAR (version 1.0.0)

plot_R2: Visualize the Proportion of Variance Explained (\(R^2\))

Description

Creates a horizontal bar chart showing the observed \(R^2\) as a proportion of total variance, with an optional confidence interval displayed beneath the bar and sample size / predictor-count annotations.

Usage

plot_R2(
  R2,
  N = NULL,
  p = NULL,
  conf_level = 0.95,
  show_ci = TRUE,
  show_n = TRUE,
  random_predictors = TRUE,
  title = NULL,
  palette = "okabe_ito",
  colors = NULL
)

Value

A ggplot2 object.

Arguments

R2

The observed squared multiple correlation coefficient (\(0 \le R^2 \le 1\)).

N

Total sample size.

p

Number of predictors.

conf_level

Confidence level for the confidence interval (default 0.95).

show_ci

Logical. If TRUE (the default), a confidence interval is shown beneath the proportion bar. Requires both N and p.

show_n

Logical. If TRUE (the default), N and p are annotated on the plot.

random_predictors

Logical. Whether the predictors are random (TRUE, the default) or fixed. Passed to ci_R2.

title

Optional plot title.

palette

Character string naming the color palette used for the “Explained” portion of the bar when colors is NULL. Defaults to "okabe_ito", base R's colorblind-safe Okabe-Ito palette; "tableau" is also available.

colors

Optional character vector of length 2: the first color fills the “Explained” portion of the bar, the second the “Unexplained” portion. When NULL (the default), the “Explained” portion uses the first color of palette and the “Unexplained” portion a neutral light gray.

Author

Ken Kelley [email protected]

References

Kelley, K. (2008). Sample size planning for the squared multiple correlation coefficient: Accuracy in parameter estimation via narrow confidence intervals. Multivariate Behavioral Research, 43, 524--555. tools:::Rd_expr_doi("10.1080/00273170802490632")

Maxwell, S. E., Delaney, H. D., & Kelley, K. (2027). Designing experiments and analyzing data: A model comparison perspective (4th ed.). Routledge. (See Chapter 3 on \(R^2\) as a model comparison effect size.)

See Also

ci_R2, ci_R, plot_ci, plot_smd

Other plotting: plot_cfa_k(), plot_ci(), plot_equivalence(), plot_forest(), plot_irt_information(), plot_mediation_mbco(), plot_randomization_test(), plot_regions_of_significance(), plot_smd(), plot_trajectories(), plot_trajectories_fitted(), power_equivalence_md_plot()

Examples

# The default display: the observed R2 fills its share of the bar, the
# confidence interval sits beneath it, and N and p are annotated.
plot_R2(R2 = 0.25, N = 100, p = 5)

# With fixed predictors and a 90% confidence interval. The interval
# beneath the bar is now the fixed predictor interval from ci_R2, and
# its label reads 90% rather than 95%.
plot_R2(R2 = 0.35, N = 200, p = 3, conf_level = 0.90,
        random_predictors = FALSE)

# Without the interval or the sample size annotation, the bar stands
# alone, which is the display to use when N and p are not known.
plot_R2(R2 = 0.10, show_ci = FALSE, show_n = FALSE)