dicho.t.z: t from dichotomizing/extreme group design to z

In phoebehlam/michaela: convert statistical metrics for meta-analytic purpose

t from dichotomizing/extreme group design to z

Description

t-statistics from dichotomizing/extreme group design to fisher's z
see pustejovsky, 2014. psychological methods.
returns z based on taylor series approximation, defaults 5th term (use dicho.d.z() if you'd like to adjust to lower term)
for unadjusted statistics only

Usage

dicho.t.z(t, totn, n1, n2, p1, p2, dir)

Arguments

t	t-statistics
totn	the total n
n1	cell size of group 1
n2	cell size of group 2
p1	cutoff percentile for group 1 (sample-based, please see other formulas for population-based in pustejovsky, 2014)
p2	cutoff percentile for group 2 (sample-based, please see other formulas for population-based in pustejovsky, 2014)
dir	if t-statistics is in absolute value, provide the empirical direction: +1 for positive associations (e.g., group 1 - group 2 >= 0) and -1 for negative associations (e.g., group 1 - group 2 < 0)

Examples

dicho.t.z(2.05, 15+15, 15, 15, 1/3, 1/3, -1)

phoebehlam/michaela documentation built on Oct. 23, 2024, 4:10 p.m.