analysis of nominal data Flashcards
what test is used for nominal data
chi squared
chi squared test
to work out if observed frequency differs from expected frequency
uses of chi squared
goodness of fit and test for independence
goodness of fit
does the observed frequency of one nominal level variable differs from the expected frequency of that variable?
test for independence
is there an association between two nominal level variables?
how does chi squared work
compares the Observed data to the frequency distribution of the null hypothesis
critical value
when we compare the found value to the chi squared if the null hypothesis is true
chi squared test of independence
tests whether two categorical variables are related
expected values
E = [(row frequency)x(column frequency)] / N
df for a 2x2 test
1
df
df = (number of columns - 1) x (number of rows - 1)
limitations of chi squared 1
Should have at least 20 observations
limitations of chi squared 2
Should be an expected value of no less than 5 in any individual cell
limitations of chi squared 3
Observations must be unique to one cell
limitations of chi squared 4
Only frequencies can ever appear in a cell
