chi-square analysis Flashcards
assumptions chi-mutually exclusive.
each group 1 cell= must be independent
mutually exclusive= individual can not simutanouly belonhg one than one cell because events do not occur simultaniously eg you toss a coin cannot be heads and tails
assumptions chi-large enough expectant count
frequency each less than 20% of expected count 5 or greater .
random selection
ensure represent population conduct multiple study like test- re test
one way chi
one variable and several levels
two way chi
two categorical variable
assumptions- (observe)
observe frequency count and expected frequency.=n/k
total number divided number of v =MEAN?
to decide statistic sig
difference between observed and expected frequency, square so n neg and divide by expercted value
larger x2 mean
larger disprecency between observed and expected
-predict null.
if large accept to reject null as by chance
formula of chi
x2=3(0-E)squared/E
if broke assumptions
fisher exact if expectancy count less than 5
if sig use phi-coefficient
0.10- small effect
0.30 medium
0.50 large