Lecture 6 REVISED

Question 1

hypothesis = ?

Answer 1

informed speculation to be tested

e.g., possible relationship between 2 or more variables

Question 2

association = ?

Answer 2

patterned/systematic variation

Question 3

tools for discrete variables in demonstrating an association?

Answer 3

descriptive statistics: contingency tables

inferential statistics: chi-squared test

Question 4

alpha = ?

Answer 4

threshold for statistical significance

usually 5%

Question 5

what are the necessary steps to establishing causality?

Answer 5

1. rationale linking cause and effect
2. demonstrate cause came before outcome
3. show association between cause and outcome
4. remove any other factors that could be related to the outcome

Question 6

marginal distribution = ?

Answer 6

focus on one variable at a time

Question 7

conditional distribution = ?

Answer 7

focusing on a variable and how it’s affected by another variable

Question 8

how do you calculate degrees of freedom?

Answer 8

n-1

Question 9

what is chi-squared (x squared) about?

Answer 9

comparing two differences

differences we observe in our sample vs differences we expect if null hypothesis were true

Question 10

how is chi-squared calculated?

Answer 10

(observed difference - expected difference) / expected difference

Question 11

p value

Answer 11

helps us weigh the inference value of descriptive evidence

between 0 and 1, tells us how probable the null hypothesis is

Question 12

high p value means…
low p value means…

Answer 12

high p value = sample data are compatible with true null hypothesis

low p value = the sample data aren’t compatible with true null hypothesis

Question 13

what do you do with your p value?

Answer 13

compare it to alpha/significance level to decide whether to reject or fail to reject null hypothesis

Question 14

does chi squared test work with all samples?

Answer 14

no, doesn’t work well with very small samples