quiz 3

Question 1

what is chi square test?

Answer 1

a group of test to find a relationship between categorical variables measured as frequencies.

Question 2

chi-square test of independence

Answer 2

• testing for independence between two or more categorical values

Question 3

chi-square goodness of fit test

Answer 3

comparing observed frequencies to explicitly stated expected frequencies

Question 4

what are counts?

Answer 4

how often a particular event occurs

Question 5

if i flipped a coin 50 times, what would the expected be?

Answer 5

25 heads & 25 tails

Question 6

were males more likely to go to college than females? what would the null hypothesis be?

Answer 6

the rate of college attendance is independent from gender.

Question 7

If chi square is greater than critical value, what can we assume?

Answer 7

we can assume that the p-value is smaller than alpha and the null hypothesis can be rejected.

Question 8

if chi square is less than the critical value, what can we assume?

Answer 8

we can assume that the p-value is bigger than alpha and the null hypothesis cannot be rejected.

Question 9

problems with chi-square test

Answer 9

• chi square is very sensitive to low values and zeros
• Chi square is an approximation of the differences
• approximation gets worse with small sample sizes
• unreliable with d.f.=1 (would get a high false positive)

Question 10

what do i do if i want to compare frequencies with small sample sizes?

Answer 10

• you can collect more data or you can use Fisher’s exact

Question 11

yates correction

Answer 11

makes a more accurate p-value