effect_sizes: Effect sizes conversions

In jmbarbone/jordanExtra: Extra Functions for the mark Package

Effect sizes conversions

Description

Calculate effect sizes and conversions

Usage

odds_ratio(a, b = NULL, c = NULL, d = NULL, type = "hits_misses")

odds2d(odds, var = FALSE)

odds2r(odds, n1 = 4, n2 = n1, var = FALSE)

r2cohend(r, var = FALSE)

cohend2r(cohend, n1 = NULL, n2 = n1, var = FALSE)

cohend2odds(cohend, var = FALSE)

Arguments

a, b, c, d	Values in the contingency table; valid if either 'a', 'b', 'c' and 'd' are the same length or if 'a' has length of '4' and 'b', 'c', and 'd' are 'NULL'.
type	Type of entry for [odds_ratio()]
odds, r, cohend	A vector of effect sizes
var	Logical, if 'TRUE' converts the variance instead
n1, n2	The ns of the groups

Details

'odds' are log odds

Examples

x <- c(26, 13, 5, 6)
odds_ratio(x)           # 2.4
odds2r(2.4)             # 0.2346004
odds2d(2.4)             # 0.4826712
r2cohend(0.2346004)    # 0.4826712
cohend2odds(0.4826712) # 0.8754687
cohend2r(0.4826712)    # 0.2346004
exp(0.8754687)          # 2.4

jmbarbone/jordanExtra documentation built on April 13, 2025, 3:28 a.m.