lecture 3

Question 1

What is the null distribution?

Answer 1

The distribution of t assuming that H0 is true

Question 2

What does the P value tell you?

Answer 2

The P value tells you how likely it s to get the results you got assuming that H0 is true.
(So, if 14 out of 18 toads prefer their right hand, the P value will be low, since the chance of this happening assuming H0 is true (purely due to chance) is small)

Question 3

If P < a then…
If P > a then…

Answer 3

If P < a, you reject H0
If P > a, you fail to reject H0

Question 4

What is a type 1 error?
What is a type 2 error?

Answer 4

Type 1 error: Reject H0 when it’s actually true. (False positive)
Type 2 error: Fail to reject H0 when in fact alternative hypothesis is true (false negative)

Question 5

What is the significance level alpha? What is the confidence level 1-alpha?

Answer 5

Significance level alpha is the probability of committing a type-1 error when H0 is actually correct.
Confidence level 1-alpha is the probability of not rejecting H0 when it’s correct.

Question 6

What are some factors that affect the power of a test?

Answer 6

• The effect size: how strong is the effect we’re looking for?
• Sample size: increased sample size leads to increased confidence
• Significance level: the lower the significance level, the harder it gets to reject H0.

Question 7

What does test statistic t tell you something about?

Answer 7

Test statistic t tells you how many standard errors away you are from the mean of the null hypothesis.