Chapter 9 Hypothesis Testing

Question 1

What is the general question for classical hypothesis testing?

Answer 1

Given a sample and an apparent effect, what is the propbability of seeing such as effect by chance?

Question 2

Steps for hypotehsis testing?

Answer 2

1. Compute Test Statistic
2. Define the Null Hypothesis
3. Compute p-values, which is the probability of seeing the apparent effect if the null hypothesis is true
4. Interpret the Results (low p-value indicates that the effect is statistically significant, meaning that it is unlikely to occure by chance)

Question 3

Suggested interpretation of p-values?

Answer 3

Order of Magnitude:

< 1%, effect unlikely due to chance,

b/w 1% and 10%, boderline,

> 10%, effect due to chance likely

Question 4

When do you use a one-sided test over a two sided test?

Answer 4

If you are looking for any difference, then one sided. If you are looking for a difference in a perticular direction, then one-sided.

Question 5

When is the Chi-squared test used?

Answer 5

For tessting proportions

Question 6

Mathmatical definition of Chi-squared?

Answer 6

Where Oi are the observed frequencies and Ei are the expected frequencies

Question 7

What is a limitation of a Chi-squared test?

Answer 7

It indicates that there is a difference between two groups, but they don’t say anything specific about what that difference is.

Question 8

Define the power of a test?

Answer 8

the correct positive rate (sometimes called sensitivity)

• is dependent on the actual effect size

Question 9

What is an acceptable power rate?

Answer 9

80% (chance of yeilding a positive result)

Question 10

Methods for compensating for mutiple tests in hypothesis testing?

Answer 10

• Adjust the p-value (Holm-Bonferroni Method)
• Partioning the Data (Exploration and Testing)