Chi Square and Measure of Association

Question 1

What does a goodness of fit test tell you?

Answer 1

How well the observed and expected frequencies match, the value is proportional to size of differences.

Question 2

How can expected frequencies be distributed? 4

Answer 2

1. Equiprobably
2. In accord with theory
3. Previous observations
4. Normally distributed

Question 3

What does the test of independence tell us?

Answer 3

Whether two categorical variables are related or independent.

Question 4

what does it mean to reject Ho in the test of independence?

Answer 4

The two variables (or more) are considered likely to be related (not independent)

Question 5

What are the assumptions of chisquared tests?

Answer 5

1. Observation are independent (people contribute only one score) 
2. Size of expected frequency >5
3. Inclusion of non-occurrences

Question 6

What type of test would you use for any ordinal data?

Answer 6

Non-parametric

Question 7

When happens to chisquared values when you double the sample size?

Answer 7

It doubles the obtained value

Question 8

Which chisquare test do you use measure of association for?

Answer 8

Test of independence/two-way chisquare/contingency chisquare

Question 9

What is k in a cramer’s phi calculation?

Answer 9

The smaller of either r or c