Skip to content

Instantly share code, notes, and snippets.

@jirilukavsky
Last active August 10, 2016 07:56
Show Gist options
  • Save jirilukavsky/e8745c5a78a1d2f674d364e157852818 to your computer and use it in GitHub Desktop.
Save jirilukavsky/e8745c5a78a1d2f674d364e157852818 to your computer and use it in GitHub Desktop.
Quick power analysis
# Power analysis examples to answer quick questions.
# - using pwr package in R
# - see also http://www.statmethods.net/stats/power.html
# --------------------------------------------------
# setup
library(pwr)
p <- 0.05
power <- 0.80
# Summary of possible errors
cat("For given p/power, you have\n ", p * 100, "% chance to find an effect which is not there (Type 1)\n ",
100 * (1 - power), "% chance to miss an effect which is there (Type 2)")
# Two-sample t-test
# How many people for a given effect (in _each_ group)
d <- 0.2
pwr.t.test(d = d, sig.level = p, power = power, type = "two.sample")
# Paired t-test
# How many people for a given effect
d <- 0.2
pwr.t.test(d = d, sig.level = p, power = power, type = "paired")
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment