R/wrap.t.one.R

In michaelkardas/behavioralwrappers: Wrappers and other functions for the behavioral sciences

#' One-sample t test
#'
#' @description Performs one-sample t tests. The function delegates the primary
#' operations to \code{\link[stats]{t.test}}.
#'
#' @param dv1 Column vector containing the dependent variable
#' @param mu Mean against which to compare responses on the dependent variable
#'
#' @seealso \code{\link[stats]{t.test}}
#'
#' @examples
#' wrap.t.one(dv1 = bdata$DV1, mu = 5)
#' @import effsize stringr stats
#' @export
wrap.t.one  <- function(dv1, mu = 0) {
  options(scipen=999)

  # Error checks
  if(is.null(dv1)) {return(paste("Cannot find the column vector inputted to parameter dv1."))}
  if(is.null(dv1)==F) {if(is.numeric(dv1)==F) {return("Argument dv1 must be numeric.")}}

  # Perform t test
  a <- t.test(dv1, NULL, alternative = c("two.sided"), mu, paired = FALSE, var.equal = TRUE,conf.level = 0.95)

  # compute Cohen's d
  b <- ((mean(dv1,na.rm=T)-mu)/sd(dv1,na.rm=T))

  # Produce output
  if(a$p.value < .001) {
    wrap.writeClipboard(paste("# one-sample t(",a$parameter,") = ",wrap.rd0(a$statistic,2),", p < .001", ", 95% CI = [",wrap.rd0(a$conf.int[1],2),", ",wrap.rd0(a$conf.int[2],2),"], d = ",wrap.rd0(b,2),sep=""))
    return(cat("# one-sample t(",a$parameter,") = ",wrap.rd0(a$statistic,2),", p < .001", ", 95% CI = [",wrap.rd0(a$conf.int[1],2),", ",wrap.rd0(a$conf.int[2],2),"], d = ",wrap.rd0(b,2),sep=""))
  }
  else {
    wrap.writeClipboard(paste("# one-sample t(",a$parameter,") = ",wrap.rd0(a$statistic,2),", p = ",wrap.rd(a$p.value,3),", 95% CI = [",wrap.rd0(a$conf.int[1],2),", ",wrap.rd0(a$conf.int[2],2),"], d = ",wrap.rd0(b,2),sep=""))
    return(cat("# one-sample t(",a$parameter,") = ",wrap.rd0(a$statistic,2),", p = ",wrap.rd(a$p.value,3),", 95% CI = [",wrap.rd0(a$conf.int[1],2),", ",wrap.rd0(a$conf.int[2],2),"], d = ",wrap.rd0(b,2),sep=""))
  }
}