Slides 4: Null Hypothesis Testing

null hypothesis

nothing new/different/special going on in the sample

null hypothesis equation

H₀ is μ₁= μ₂, always population

if enough evidence against H_0…

…reject it and accept the alternative

p value

the probability of getting results as extreme as what you got assuming the null is true

equation of the exact probability of x happening

binomial equation p(x) = k!/(x!(k-x)!

alpha

the cutoff value of how strong evidence has to be/how low p needs to be to conclude H₀ is not true

What do you do if you want to accept less/weaker evidence?

raise alpha value (e.g. alpha of 0.1)

What do you do if you want to require very strong evidence?

lower alpha value (e.g. alpha of 0.01)

if p < alpha then…

…reject H₀ (null) and accept H_A (alternative hypothesis)

if p > alpha then…

fail to reject H₀ (null)

Type I error

when the null is incorrectly rejected when it is actually a true null

Type II error

when a false null hypothesis is not rejected