dPrime: Calculating d-prime (d')

View source: R/dPrime.R

dPrimeR Documentation

Calculating d-prime (d')

Description

This function calculates the dimensionless statistic d-prime (d'). To do this, the (log-linear transformed if transformation = TRUE) percentage values are converted to z-values with the help of inverse cumulative distribution function (CDF) of the Gaussian distribution with a mean of 0 and a standard deviation of 1. Or as a equation:

d' = z(hit rate) - z(false alarm rate)

where z(p), p in [0,1] is the inverse Gaussian CDF.

Usage

dPrime(
  oldRating,
  ratings,
  oldNew,
  oldNewLevels = c(0, 1),
  transformation = TRUE
)

Arguments

oldRating

Values corresponding to 'old' responses.

ratings

A vector containing all responses.

oldNew

An vector containing whether an item was new/not-studied or old/studied.

oldNewLevels

An vector containing possible levels of old_newInformation. The first value or level is for new/not-studied items and the second value is for old/studied items. Defaults to c(0,1).

transformation

Logical indicating to choose a log-linear transformation following Snodgrass & Corwin (1988). Default is TRUE.

Value

The function returns a value for d'.

Author(s)

Joern Alexander Quent, alexander.quent@rub.de

References

Snodgrass, J. G., & Corwin, J. (1988). Pragmatics of measuring recognition memory: Applications to dementia and amnesia. Journal of Experimental Psychology: General, 117(1), 34–50. https://doi.org/10.1037/0096-3445.117.1.34


JAQuent/memoryROC documentation built on June 9, 2025, 1:43 a.m.