ANOVA and stuff

SSB

summation ni(ybar i - y bar..)²

SSTO

summation summation (yij - y bar ..)^2

SSW

summation (ni - 1) si²

df (SSW)

n. - I

df(SSB)

I - 1

df(SSTO)

n. - 1

F statistic

Fs = MSB/MSW

power

1- beta/alpha

beta

P(fail to reject H0|H0 false)

alpha

P(reject H0|H0 true)

sp²

variance pooled. (summation (ni-1) si²)/(n.- I). sqrt for SD

p~

y+2/n+4

assumptions for p~ confidence interval

y greater than or equal to 5, n-y is greater than or equal to 5. random sample is taken

confidence interval for p~

p~ ± Z a/2 sqrt ((p~(1-p))/n+4)

ANOVA confidence interval

y bar a - y bar b ± t (a/2k) sqrt (MSW(1/na) + (1/nb))