Normal distribution
continuous probability distribution, bell shaped curve, symmetrical around the mean
binomial distribution
Discrete probability distribution/there is two outcome
poisson distribution
frequency of distribution of events that occur rarely and randomly /independent events occurring in fixed blocks of time or space
what are the 4 conditions for poisson?
1-probab. of two or more occurrences must be small
2-the probab. of occurrence of it in a sample is proportional to the size of that sample
3-the outcomes are independent
4- the probability of an occurrence is identical for all sample subdivisions
what distributions are used for categorical variables?
binomial and poisson
what are the two best way to test categorical variables?
analysis of their frequency,
goodness of fit, contingency analysis.
compares the observed frequency in different categories with the expected frequency under the null
Goodness of fit:
chi-square test, G test/ both look at ratio of observed and expected/ only need one categorical variable
G-test
log likelihood ratio test-more powerful than traditional chi square
contingency analysis
whether two categorical variables are associated with one another/ testing independence of categorical variables
extrinsic hypothesis
derived from information other than the data you are analyzing (the expected frequency is given)
intrinsic hypothesis
expected frequencies are derived from the data you are analyzing (we have to calculate the expected frequency based on our contingency table)
if G> X2
reject the null hypothesis
