comparison_test_treatment: Performs between- or within-individual experiments comparing...
In ben18785/chronodoseresponse: melluxdrc

Description Usage Arguments Details Value Examples

For a between-individual comparison, a t-test comparing the melatonin suppression level at a single lux is conducted on two subgroups (untreated and treated) comprising different sets of individuals.

comparison_test_treatment(
  is_between,
  lux,
  n,
  population_treated_df,
  is_treated_higher,
  nreps = 1
)

`is_between`	a Boolean indicating whether experiment is within or between type
`lux`	the lux value at which to conduct the comparison
`n`	number of individuals in sample
`population_treated_df`	a virtual population data frame with individual who have been treated and those who have been untreated (where treatment is indicated by a Boolean column named "treated")
`is_treated_higher`	Boolean indicating whether treated group has a higher response
`nreps`	integer indicating number of test replicates (which defaults to 1)

For a within-individual comparison, a paired t-test comparing the melatonin suppression level at a single lux level for a set of individuals before and after they are treated is conducted.

This function requires based an (ideally large) simulated population of individual dose-response data.

a dataframe where each row corresponds to a test result in a replicate; note that 'result' indicates that the difference was of the correct sign

library(melluxdrc)
library(purrr)
library(dplyr)

# create a population where half are treated; half not
n <- 200
population_df <- virtual_treatment_experiment(n, treated_ed50_multiplier=0.5)

# carry out a within-study test at 10 lux for 30 people
# note that, because the ed50 multiplier for the treated group was
# <1, this group should have a higher response
is_treated_higher <- TRUE
comparison_test_treatment(FALSE, 10, 30, population_df, is_treated_higher)