This is a draft cheat sheet. It is a work in progress and is not finished yet.
Sampling Distribution
Inference: statistic (sample, x̅) --> parameter (population, μ) |
CLT: large n --> x̅~N(μ,σ/√n) - as n increases, x̅ approaches μ |
Sample Proportion: mean = p and s.d. = √[p(1-p)/n] |
Z-statistic: when n>30, z=(x̅-μ)/(σ/√n) --> z-table |
T-statistic: when n<30, d.f.=n-1 --> t-table |
|
|
Statistical Inference
Inference |
Confidence Interval |
Tests of Significance |
- probability --> trustworthy? |
estimate+/- MoE |
H0: "no effect" |
- sample --> population |
x̅ +/- z∗(σ/√n) |
Ha: what we are testing |
- based on sampling distribution |
MoE ↓ = σ↓ = n↑= confidence ↓ |
Assume H0 --> what is the P of a result as/more extreme than statistic --> reject if P≤α |
|
