Updated Thu Aug 27 2020

- Added ANOVA_exact2 function as an extension of ANOVA_exact
- Now functional across all sample sizes but does not return a dataframe of afex aov object

- liberal_lambda arugment added: allows users to specify the type of lambda calcuations
- When liberal_lambda = TRUE; lambda = cohen_f^2 * (num_df + den_df + 1)
- When liberal_lambda = FALSE; lambda = cohen_f^2 * den_df

- Optimal alpha functions from JustifieR package added
- ANOVA_compromise function added which allows a compromise power analysis to be performed for all comparisons in a design
- ANOVA_design now returns as a class “design_aov” with specific print and plot methods see ?
`design_aov-methods`

- generate_cor_matrix function is now a non-exported function within the package (no longer contained within ANOVA_design)

- All simulation functions ANOVA_power, ANOVA_exact, and ANOVA_exact2 now returns as a class “sim_result” with specific print and plot methods see ?
`sim_result-methods`

- plot_power now has reduced sample size limtations -Option to use ANOVA_exact2 (exact2 argument) improves functionality (not limited to product of factors)
- Updated vignettes to include updated information on functions
- New vignette “Introduction to Justifying Alpha Levels”

- New Shiny App: justify
- Creates a UI for utilizing the ANOVA_compromise function via Shiny

- Superpower_options(“plot”) is now set to TRUE. Plots will, by default, be printed -Easily reset with Superpower_options(plot = FALSE)
- plot_power has new features -Plots now show desired power -min_n is now limited; smallest min_n allowed is equal to the product of the design (e.g., ’2b*2b’ has a smallest min_n of 4)
- Small update to plot_power to fix minor error in original code -Error resulted in power estimates being ~0.1-0.5% off actual power estimate

- Added emmeans_power function
- Documentation added to the vignette

- Small updates to the Shiny apps to fix typos

Unequal sample size in the design is now permitted -Limited to the ANOVA_design and ANOVA_power functions

Added estimated marginal means comparisons using

`emmeans`

R package.`emm = TRUE`

in the ANOVA_power, ANOVA_exact, and plot_power will result in emmeans being calculated- Default is all pairwise comparisons but this can be modified with
`contrast_type`

and`emm_comp`

options

Added global options

- Options that have crossover between functions can now be set globally for the package
- Includes: verbose, emm, emm_model, contrast_type, alpha_level, and plot
- These global options can be seen with Superpower_options()

Updated Shiny Apps

- Unequal n allowed for ANOVA_power
- Added numeric input for alpha level (no longer slider)
- Now includes emmeans options
- kableExtra, emmeans, magrittr, and dplyr packages now needed to knit markdown file in app.