coeff_power_plot_fun: Title: Coefficient of variation and Power plots for different...

In AshleyAkerman/mmReliability: What the Package Does (One Line, Title Case)

Description

NOTES: Function to generate plots for sample size estimations based on the coefficient of variation of two tools given a similar mean estimate NOTES: Or alternatively a power plot for sample size estimations based on effect size and power.

Usage

coeff_power_plot_fun(
  MDC_input = NA,
  grand_mean,
  fig_type,
  SESOI = c(1, 3, 5, 10, 20, 25, 30),
  CoV = seq(4, 50, 2),
  CV1 = NA,
  CV2 = NA,
  all_percent = FALSE
)

Arguments

MDC_input	vector of values referring to the minimum detectable change estimates. Only required for power curve
grand_mean	numeric value for the grand mean. If coefficient figure then this is the value consistent for both measurement tools. If a power curve, it is the grand mean of the one tool.
fig_type	string input taking form of "coefficient" or "power". Coefficient is to compare sample sizes across CV values. Power is to compare sample sizes across power for given effect sizes.
SESOI	vector of values for the smallest effect size of interest (default: 1, 3, 5, 10, 20, 25, 30%)
CoV	vector of values for the coefficient of variation range important for the "coefficient" plot. Default is a sequence from 4 to 50 with increments of 2.
CV1	numeric value for the coefficient of variation. If coefficient figure then the 1st CV would be the poorer tool, if power figure then CV for the only tool (and CV2 == NA)
CV2	numeric value for the Coefficient of variation. If coefficient figure this is required as the better of the two tools.
all_percent	boolean input. Default is FALSE. TRUE when the smallest effect size of interest and the MDC are the same unit (i.e., %), so don't need to transfer to absolute values.

Value

figure either for coefficient of variation or power curve and estimating sample size.

AshleyAkerman/mmReliability documentation built on Dec. 31, 2020, 9:51 a.m.