P-value function: T-test

These functions accept a data frame and perhaps test specific arguments (like whether or not the test will be asymptotic or simulation based). It produces a p-value.

Usage

pOneway(
  dat,
  fmla = YContNorm ~ trtF | blockF,
  simthresh = 20,
  sims = 1000,
  parallel = "no",
  ncpu = NULL
)

Arguments

dat

An object inheriting from class data.frame

fmla

outcome~treatment factor | block factor (following coin API).

simthresh

Below which number of total observations should the p-value functions use permutations rather than asymptotic approximations

sims

Either NULL (meaning use an asymptotic reference dist) or a number (meaning sampling from the randomization distribution implied by the formula)

parallel

Should the function use multicore processing for permutation based testing. Default is no. But could be "snow" or "multicore" following approximate in the coin package.

ncpu

is the number of workers (for "snow") or cores (for "multicore").

Value

A p-value

Examples

# Example using built-in data
data(example_dat, package = "manytestsr")

# Test for treatment effect on Y1 within a single block
single_block <- subset(example_dat, blockF == "B080")
p_val <- pOneway(single_block, Y1 ~ trtF | blockF, parallel = "no")
print(p_val)
#> [1] 0.2576562

# Test with permutation-based inference for small samples
p_val_perm <- pOneway(single_block, Y1 ~ trtF | blockF,
  simthresh = 100, sims = 500, parallel = "no"
)
print(p_val_perm)
#> [1] 0.2576562