Tests for nominal data

Question 1

What does a binomial test tell us

Answer 1

Is the proportion in the data what we would expect by change?
- e.g. if you flip a coin 100 times you’d expect to get 50 heads and 50 tails and a binomial test tells us the probability of finding our observed proportion given the expected proportion (chance)

Question 2

what are the assumptions for a binomial test

Answer 2

• data are nominal
• data are independent
• random sample of population
• know the expected distribution score (chance level)
• single dichotomy ( one variable with 2 mutually exclusive outcomes)

Question 3

What is the Chi Square Test of independence

Answer 3

one sample vs another sample

Question 4

what are the assumptions for a Chi Square Test of Independence

Answer 4

• data are nominal
• two dichotomies: e.g. male/female and yes/no
• scores from a random sample
• data are independent
• N or over 40
• each category must have an expected N of 5 or above

Question 5

what’s the non parametric alternative for a chi square test

Answer 5

Fisher’s exact test

Question 6

What is Chi-square Goodness of fit test

Answer 6

similar to binomial
- how does an observed proportion differ from an expected proportion but this time there are more than 2 options

Question 7

what are the assumptions for chi square goodness of fit test

Answer 7

• data are nominal
• multiple levels of a single dependent variable (like location)
• scores are from a random sample
• independent
• each category has an expected N of 5 or above