powerANOVA: Power Analysis for Repeated Measures ANOVA.

powerANOVA is an R package intended as a companion to the article "A Tutorial on Using the Paired t-Test for Power Calculations in Repeated Measures ANOVA". Supplementary materials for the article can be found on osf.

1 Installation

powerANOVA is not on CRAN. The development version of powerANOVA can be installed directly from this GitHub repository using the additional package devtools.

# install devtools
install.packages("devtools")

# install subgroupsem
devtools::install_github("langenberg/powerANOVA")

2 Usage

First, load the package using the following command:

library(powerANOVA)

2.1 Graphical User Interface (GUI)

powerANOVA includes an easy to use GUI. The GUI is a shiny app which can be loaded using the following command:

power_gui()

The command will open a browser tab. The GUI is very self-explaining. Simply close the browser when you want to terminate the GUI.

2.2 Converters

• convert_cohens_d_petasq(cohens_d, n, population = TRUE)
• convert_petasq_cohens_d(p_eta_sq, n, population = TRUE)
• convert_petasq_f2(p_eta_sq)
• convert_F_petasq(Fval, df1, df2)
• convert_petasq_F(p_eta_sq, df1, df2)

2.3 Plots

• power_plot_cohens_d(n, cohens_d, alpha = 0.05)
• power_plot_mu_cov_contrast(n, mu, Sigma, contrast, alpha = 0.05)
• power_plot_p_eta_sq(n, p_eta_sq, df1, alpha = 0.05)

2.4 Power Calculators

• power_cohens_d(n, cohens_d, alpha = 0.05)

  E.g.:

   mu <- 58
   var <- 7200
   
   cohens_d <- mu/sqrt(var)
   
   power_cohens_d(n = 19, cohens_d = cohens_d)
  

• power_petasq(n, p_eta_sq, df1, alpha = 0.05)

  E.g.:

   p_eta_sq <- 0.3296746
   
   power_petasq(n = 19, p_eta_sq = p_eta_sq)
  

• power_mu_cov_contrast(n, mu, Sigma, contrast, alpha = 0.05)

  E.g.:

   means_dv <- matrix(c(492, 511, 483, 444), ncol = 1)
   names(means_dv) <- c("A1.B1", "A1.B2", "A2.B1", "A2.B2")
   
   # (co)variances of the dependent variables
   vcov_dv <- matrix(c(
       9000, 7200, 7200, 7200,
       7200, 9000, 7200, 7200,
       7200, 7200, 9000, 7200,
       7200, 7200, 7200, 9000
   ), ncol = 4, nrow = 4)
   
   colnames(vcov_dv) <- c("A1.B1", "A1.B2", "A2.B1", "A2.B2")
   rownames(vcov_dv) <- c("A1.B1", "A1.B2", "A2.B1", "A2.B2")
   
   # contrast vector
   contrast_vec <- matrix(c(1, -1, -1, 1), nrow = 1)
   
   colnames(contrast_vec) <- c("A1.B1", "A1.B2", "A2.B1", "A2.B2")
   rownames(contrast_vec) <- c("difference")
   
   power_mu_cov_contrast(n = 19, mu = means_dv, Sigma = vcov_dv, contrast = contrast_vec)