Confidence Interval Information
( \newcommand{\kernel}{\mathrm{null}\,}\)
Confidence Intervals if σ is KnownPoint estimate ± EBM (Error bound for a population mean)
*EBM is also known as the "Margin of Error"
CL=Confidence Level
We use the “Standard Normal Distribution” to calculate zα2
To find zα2 using Desmos:
inversecdf(normaldist(0,1), CL+ α2)
We are trying to capture the true population mean (μ, this is a parameter) with this confidence interval!
|
Confidence Intervals if σ is Not KnownUse the “sample standard deviation” or s instead. Because of this, we have to use t distributions.
ˉx±tα2(s√n)
Point estimate ± EBM (Error bound for a population mean)
*EBM is also known as the “Margin of Error” DF=Degrees of Freedom= n−1 CL=Confidence Level
To find tα2 using Desmos:
inversecdf(tdist(Degrees of Freedom), CL+ α2 )
We are trying to capture the true population mean (μ, this is a parameter) with this confidence interval!
|
Confidence Intervals for Proportionsˆp(p hat) OR p′(p prime)= sample proportion (think number of successes from Binomial Distributions)
If it’s wearing a “hat” it’s from a sample, not a population. No “hat” then it’s a population parameter!
ˆp=x(numberofsuccesses)n(samplesize)
ˆp±zα2√ˆpˆqn or ˆp±zα2√ˆp(1−ˆp)n
Where ˆq=1−ˆp
Point estimate ± EBP (Error bound for a population proportion)
*EBP is also known as the “Margin of Error”
CL=Confidence Level
We use the “Standard Normal Distribution” to calculate zα2
To find zα2 using Desmos: inversecdf(normaldist(0,1), CL+α2 )
We are trying to capture the true population proportion (p, this is a parameter) with this confidence interval! |
by Katryn Weston