R/effectsize_extractor.R

In fsingletonthorn/EffectSizeScraping: Effect size extractor

#' Extract cohen's ds, odds ratios, hazard ratios and partial eta sqaured tests from text
#' 
#' @param inputText input text 
#' @return A tibble contaitning all detected effect sizes in odds ratios, Hazard ratios, d and eta squared,  with columns identifying the type of statistic extracted, the reported effect size as reported in the text, an empty column for the degrees of freedom (for comparability with the other extractor functions), a column for reported p values, and the effect size value reported.
#' 
#' 
#' @examples
#' extractES("a cohen's d of .05, p < .01, and a partial eta squared of .1")
#' 
#' 
#' 
extractES <- function(input) {
  
  # Setting up all possible parts of the regex
  dRegex <-
    "\\b[dg]"

  ORRegex <- "(\\b((OR)|((?i)odds\\s{1,5}?ratio(?-i))))"
  HRRegex <- "(\\b((HR)|((?i)hazard\\s{1,5}?ratio(?-i))))"
  etaRegex <- "(?i)\\b(partial(-)?\\s*)?(\u03B7|(eta))\\s*p?\\s*2?(squared)?(?-i)"
  
  numbericRegex_commas <- "(((\\d{1,3}(?=,)(,\\d{3})*)|\\d+)?)"
  
  numbericRegex_commas_decimals <- "((((\\d{1,3}(?=,)(,\\d{3}){0,99})|\\d{1,99})(\\.\\d{1,99})?)|(\\.\\d{1,99}))"
  
  ofOrEqualsRegex <- "((of)|=|:|;)"
  
  numbericBelow1Regex <- "((?<![1-9])\\.\\d+|0(\\.\\d+)?|(1\\.0*(?!(0*[1-9])))|((?<![0-9\\.])1(?![\\.0-9])))"
  # additional p value detector
  # additional p value detector
  pValueRegex <- "((?i)((\\s*,?\\s*)(ns))|(p\\s*[<>=(ns):]\\s*[<>]?\\s{0,5}((ns)|(\\d?\\.?\\d+e?-?\\d*)|(\\.\\d+)))(?-i))"
  
  DPRegex <- paste0(
    dRegex,
    # Allowing spaces
    "\\s{0,5}",
    ofOrEqualsRegex,
    # Making degrees of freedom optional
    # Allowing spaces or negatives
    "\\s{0,5}",
    "\\-?",
    "\\s{0,5}",
    numbericRegex_commas_decimals,
    "\\s*",
    "\\;?",
    ",?",
    "\\s*",
    pValueRegex, 
    "?" # Making p values optional
  ) 
  
  
  ORPRegex <- paste0(
    "(",
    ORRegex,
    "|",
    HRRegex,
    ")",
    # Allowing spaces
    "\\s{0,5}",
    ofOrEqualsRegex, 
    "\\s{0,5}",
    numbericRegex_commas_decimals,
    "\\s{0,5}",
    "\\;?",
    ",?",
    "\\s*",
    pValueRegex, 
    "?" # Making p values optional
  ) 
  
  etaPRegex <- paste0(
    etaRegex,
    # Allowing spaces
    "\\s{0,5}",
    ofOrEqualsRegex,
    # Making degrees of freedom optional
    # Allowing spaces or negatives
    "\\s{0,5}",
    numbericBelow1Regex,
    "\\s*",
    "\\;?",
    ",?",
    "\\s*",
    pValueRegex, 
    "?" # Making p values optional
  )
  
  detected_d <- unlist(stringr::str_extract_all(
    input, DPRegex
  ))
  
  detected_OR <- unlist(stringr::str_extract_all(
    input, ORPRegex
  ))
  
  detected_eta <- unlist(stringr::str_extract_all(
    input, etaPRegex
  ))
  
  
  value_d <- stringr::str_extract(
    detected_d, numbericRegex_commas_decimals)
  
  ps_d <-
    stringr::str_trim(
      stringr::str_extract(
        detected_d,
        pValueRegex
      ))  
  
  value_OR <- stringr::str_extract(
    detected_OR, numbericRegex_commas_decimals)
  
  ps_OR <-
    stringr::str_trim(
      stringr::str_extract(
        detected_OR,
        pValueRegex
      ))
  
  value_eta <- stringr::str_extract(
    detected_eta,
    paste0("(?<=",
      ofOrEqualsRegex,
           "\\s{0,5})",
    numbericRegex_commas_decimals))
  
  ps_eta <-
    stringr::str_trim(
      stringr::str_extract(
        detected_eta,
        pValueRegex
      ))
  
  
  output <- 
    dplyr::bind_rows(
    list(d = tibble::tibble(raw = detected_d,  statistic = "d",  value = value_d, p = ps_d),
         OR = tibble::tibble(raw = detected_OR, statistic = "OR", value = value_OR, p = ps_OR),
         eta = tibble::tibble(raw = detected_eta, statistic = "eta", value = value_eta, p = ps_eta)
                        ))

  return(tibble::tibble(statistic = output$statistic,
                        reported = stringr::str_trim(output$raw), 
                        df1 = NA,
                        df2 = NA,
                        p = output$p,
                        value = as.numeric(output$value)
  ))
}