reappraised (version 0.1.1)

pval_cat_fn: Compares observed and expected distribution of p-values for categorical variables

Description

Creates plots of calculated p-value distribution and AUC (area under curve)

Usage

pval_cat_fn(
  df = pval_cat_data,
  seed = 0,
  sims = -1,
  btsp = 500,
  title = "",
  stat = "chi_midp",
  stat.override = "no",
  fisher.sim = "y",
  fish.n.sims = 10000,
  method = "mix",
  verbose = TRUE
)

Value

list containing 3 objects as described

• pval_cat_calculated_pvalues = plots of calculated p-value distribution and AUC

• pval_cat_reported_pvalues = plots of reported p-value distribution and AUC (if p-values were reported)

• all_results = list containing

  • pval_cat_baseline_pvalues_data = data frame of all results used in calculations

  • pval_cat_reported_pvalues= plot of reported p-value distribution

  • pval_cat_auc_reported_pvalues = AUC of reported p-values

  • pval_cat_calculated_pvalues = plot of calculated p-value distribution

  • pval_cat_auc_calculated_pvalues= AUC of calculated p-values

Arguments

df

data frame generated from load_clean function

seed

the seed to use for random number generation, default 0 = current date and time. Specify seed to make repeatable.

sims

number of simulations, default -1 = function selects based on number of variables.

btsp

number of bootstrap repeats used to generate 95% confidence interval around AUC

title

optional title for plots

stat

statistical test to be used 'chisq', 'fisher', 'midp' or 'midp.epitools' (from epitools package), 'midp.sas' (as calculated in SAS), or combinations -if chisq is not appropriate because expected cells<5, use second test: 'chi_fish', 'chi_midp' or 'chi_midp.epi','chi_midp.sas'

stat.override

if 'yes' then test specified in stat will be used rather than values for stat in data frame

fisher.sim

"yes" or "no" indicator whether to allow fisher test to simulate p-values for >2*2 tables

fish.n.sims

number of simulations to use in Fisher test, default 10,000

method

'sm', 'mix', or 'ind'. 'ind' does test on individual data, 'sm' summarises data and then does test on summary data, 'mix' does 'ind' for fisher and 'sm' for others. Duration varies with size of studies, test, and number of simulations. Experiment before running large simulations.

verbose

TRUE or FALSE indicates whether progress bar and comments show and prints plot

Details

See also Bolland MJ, Gamble GD, Avenell A, Grey A, Lumley T. Baseline P value distributions in randomized trials were uniform for continuous but not categorical variables. J Clin Epidemiol 2019;112:67-76.

Returns a list containing 3 objects and (if verbose = TRUE) prints the plot pval_cat_calculated_pvalues

Examples

# load example data
pval_cat_data <- load_clean(import= "no", file.cat = "SI_cat_all", pval_cat= "yes",
format.cont = "wide")$pval_cat_data

# \donttest{
# run function (takes a few seconds)
pval_cat_fn(seed=10, sims = 50, btsp = 100)$pval_cat_calculated_pvalues

# to import an excel spreadsheet (modify using local path,
# file and sheet name, range, and format):

# get path for example files
path <- system.file("extdata", "reappraised_examples.xlsx", package = "reappraised",
                    mustWork = TRUE)
# delete file name from path
path <- sub("/[^/]+$", "", path)

# load data
pval_cat_data <- load_clean(import= "yes", pval_cat = "yes", dir = path,
     file.name.cat = "reappraised_examples.xlsx", sheet.name.cat = "SI_cat_all",
     range.name.cat = "A:n", format.cat = "wide")$pval_cat_data# }