multiverse.plot {splithalf}

Visualising reliability multiverses

Description

This function allows the user to plot the output from splithalf_multiverse or testretest_multiverse. The plot includes an upper panel with all reliability estimates (and CIs) and a lower panel that indicates the data processing specifications corresponding to that reliability estimate. The (unofficial) function version name is "This function will make you a master in bird law"

This function examines the output from splithalf_multiverse or testretest_multiverse to extract the proportions of estimates above or below a set threshold (can be the estimate or the upper or lower CI estimates). The (unofficial) function version name is "This function will get you up to here with it"

Usage

multiverse.plot(
  multiverse,
  title = "",
  vline = "none",
  heights = c(4, 5),
  SE = FALSE
)

threshold(multiverse, threshold, use = "estimate", dir = "above")

Arguments

multiverse

multiverse object

title

string add a title to the plot? default is ""

vline

add a vertical line to the plot, e.g. use .5 for the median reliability estimate

heights

must be a vector of length 2, relative heights of plot panels. Defaults to c(4,5)

SE

logical includes an additional panel to plot the standard errors of the scores. Note: the heights parameter must be a vector of length 3, e.g. c(2,2,3). Defaults to FALSE

threshold

threshold to look for, e.g. 0.7

use

set to check the reliability "estimates", or the "upper" or "lower" CIs

dir

look "above" or "below" the 'use' at the set threshold

Value

Returns a visualization of a multiverse object

Examples

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

## example simulated data
n_participants = 60 ## sample size
n_trials = 80
n_blocks = 2
sim_data <- 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("A","B"),
                       each = n_trials,
                       length.out = n_participants * n_trials * n_blocks),
                       trial_type = rep(c("congruent","incongruent"),
                       length.out = n_participants * n_trials * n_blocks),
                       RT = rnorm(n_participants * n_trials * n_blocks,
                       500,
                       200),
                       ACC = 1)

## specify several data processing decisions
specifications <- list(RT_min = c(0, 100, 200),
                       RT_max = c(1000, 2000),
                       averaging_method = c("mean", "median"))
## run splithalf, and save the output
difference <- splithalf(data = sim_data,
                        outcome = "RT",
                        score = "difference",
                        conditionlist = c("A"),
                        halftype = "random",
                        permutations = 5000,
                        var.RT = "RT",
                        var.condition = "block_name",
                        var.participant = "participant_number",
                        var.compare = "trial_type",
                        var.ACC = "ACC",
                        compare1 = "congruent",
                        compare2 = "incongruent",
                        average = "mean")

## run splithalf.multiverse to perform the multiverse of data processing
## and reliability estimation
multiverse <- splithalf.multiverse(input = difference,
                                   specifications = specifications)

## can be plot with:
multiverse.plot(multiverse = multiverse,
                title = "README multiverse")


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

## example simulated data
n_participants = 60 ## sample size
n_trials = 80
n_blocks = 2
sim_data <- 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("A","B"),
                       each = n_trials,
                       length.out = n_participants * n_trials * n_blocks),
                       trial_type = rep(c("congruent","incongruent"),
                       length.out = n_participants * n_trials * n_blocks),
                       RT = rnorm(n_participants * n_trials * n_blocks,
                       500,
                       200),
                       ACC = 1)

## specify several data processing decisions
specifications <- list(RT_min = c(0, 100, 200),
                       RT_max = c(1000, 2000),
                       averaging_method = c("mean", "median"))
## run splithalf, and save the output
difference <- splithalf(data = sim_data,
                        outcome = "RT",
                        score = "difference",
                        conditionlist = c("A"),
                        halftype = "random",
                        permutations = 5000,
                        var.RT = "RT",
                        var.condition = "block_name",
                        var.participant = "participant_number",
                        var.compare = "trial_type",
                        var.ACC = "ACC",
                        compare1 = "congruent",
                        compare2 = "incongruent",
                        average = "mean")

## run splithalf.multiverse to perform the multiverse of data processing
## and reliability estimation
multiverse <- splithalf.multiverse(input = difference,
                                   specifications = specifications)


## the threshold function can be used to return the number of estimates
## above or below a certain threshold

threshold(multiverse = multiverse,
          threshold = 0.7,
          use = "estimate",
          dir = "above")

## End(Not run)