Interpreting a CI of a Mean
You are 95% sure that the 95% CI includes the population mean. • By convention write like: • 36.75 to 36.89 • or [36.75, 36.89] • Not like “36.75-36.89”
Sample mean:
this is our
best estimate of the
population mean.
SD (s):
as SD increases, the
CI increases proportionally
Sample size:
the CI is inversely proportional to the square root of the sample size. • If the sample were 4 times larger, then the CI would be half as wide.
Degree of confidence:
you can generate a narrower
interval with lower degree of confidence.
• Wide: 90%, Wider: 95%, Even Wider: 99%
Assumptions: CI of a Mean
New one:
• The true population is at least approximately
Gaussian.
• Stay tuned for bootstraps and permutations that
don’t assume this.
df =
n - 1
How to Calculate the
CI of a Mean
W = (tcrit * s) / √n