testretest.multiverse: Multiverse of data processing decisions on test retest...

In splithalf: Calculate Task Split Half Reliability Estimates

Multiverse of data processing decisions on test retest reliability estimates.

Description

This function enables the user to run a multiverse of data processing options and extract the resulting test-retest reliability estimates. The user specifies a set of data processing decisions and passes this to the function, along with specifying key variables within several "var." inputs (so that the function knows where to find your participant ids and RTs for example)

Usage

testretest.multiverse(
  data,
  specifications,
  test = "ICC2",
  outcome = "RT",
  score = "difference",
  var.participant = "subject",
  var.ACC = "correct",
  var.RT = "RT",
  var.time = "time",
  var.compare = "congruency",
  compare1 = "Congruent",
  compare2 = "Incongruent"
)

Arguments

data	dataset
specifications	list of data processing specifications
test	test retest statistic, "ICC2", "cor", "ICC3"
outcome	from splithalf() specifies the RT outcome - only "RT" available currently
score	currently only "difference" scores are supported
var.participant	= "subject",
var.ACC	= "correct",
var.RT	= "RT"
var.time	codes the time variable (currently only works for 2 timepoints)
var.compare	= "congruency" trial type used to create difference scores
compare1	specifies the first trial type to be compared (e.g. "Congruent" trials)
compare2	specifies the second trial type to be compared (e.g. "Incongruent" trials)

Details

The (unofficial) function version name is "This function will help you pay the troll toll"

Value

Returns a multiverse object containing the reliability estimates and dataframes from all data processing specifications provided

Examples

## Not run: 
## see online documentation for examples
https://github.com/sdparsons/splithalf
## also see https://psyarxiv.com/y6tcz

n_participants <- 80 ## sample size
n_trials <- 120
n_blocks <- 2

sim_data_mv <- data.frame(participant_number = rep(1:n_participants,
                                                   each = n_blocks * n_trials),
                          trial_number = rep(1:n_trials,
                                             times = n_blocks * n_participants),
                          block_name = rep(c(1,2),
                                           each = n_trials,
                                           length.out = n_participants * n_trials * n_blocks),
                          trial_type = rep(c("congruent","congruent",
                                             "incongruent","incongruent"),
                                           length.out = n_participants * n_trials * n_blocks / 2),
                          RT = rnorm(n_participants * n_trials * n_blocks,
                                     500,
                                     200),
                          ACC = c(rbinom(n_participants *
                                           n_trials *
                                           n_blocks / 6,
                                         1, .5),
                                  rbinom(n_participants *
                                           n_trials *
                                           n_blocks / 6,
                                         1, .7),
                                  rbinom(n_participants *
                                           n_trials *
                                           n_blocks / 6,
                                         1, .9),
                                  rbinom(n_participants *
                                           n_trials *
                                           n_blocks / 6,
                                         1, .5),
                                  rbinom(n_participants *
                                           n_trials *
                                           n_blocks / 6,
                                         1, .7),
                                  rbinom(n_participants *
                                           n_trials *
                                           n_blocks / 6,
                                         1, .9)))

specifications <- list(
ACC_cutoff = c(0, 0.5),
RT_min           = c(0, 200),
RT_max            = c(2000, 3000),
RT_sd_cutoff      = c(0, 2),
split_by          = c("subject", "trial"),
averaging_method  = c("mean")
)

icc2 <- testretest.multiverse(data = sim_data_acc,
specifications,
test = "ICC2",
score = "difference",
var.participant = "participant_number",
var.ACC = "ACC",
var.RT = "RT",
var.time = "block_name",
var.compare = "trial_type",
compare1 = "congruent",
compare2 = "incongruent")

multiverse.plot(icc2)


## End(Not run)

splithalf documentation built on Aug. 11, 2022, 5:10 p.m.