Key rules and distributions for the three externals, condensed for revision.
Evaluating reports (91584)
Margin of error (95%) \mathrm{MoE}\approx\frac{1}{\sqrt n}\qquad \mathrm{CI}=\text{estimate}\pm\mathrm{MoE} Error types sampling (↓ with n) vs non-sampling (bias, not fixed by n) Causation needs a randomised experiment; beware lurking variables
Binomial P(X=r)=ⁿCᵣpʳ(1−p)^(n−r) ; mean np ; var np(1−p) Poisson P(X=r)=\frac{e^{-\lambda}\lambda^r}{r!}\qquad \mu=\sigma^2=\lambda Normal z=\frac{x-\mu}{\sigma}