Hypothesis testing & P-values (6/12)

Question 1

What are the 2 basic approaches to statistical analysis:

Answer 1

○ Estimation (Confidence intervals)

○ Hypothesis testing (P-values) .

Question 2

How many steps are there to hypothesis testing

Answer 2

4

Question 3

what are the main steps to hypothesis testing

Answer 3

1. State your null hypothesis (H0) and your alternative hypothesis (HA).
2. Choose a significance level, α, for the test. (usually fixed: 0.05)
3. Obtain the probability of observing your results, or results more extreme, if the null hypothesis is true (calculating P-value).
4. Use your P-value to make a decision about whether to reject, or not reject, your null hypothesis.

Question 4

Define: P-value

Answer 4

the probability of observing your results, or results more extreme, if the null hypothesis is true

(a)

Question 5

If the Z value, when converted using the distribution table, is greater than the values on the table, what inference can be made?

Answer 5

Infer that the P-value is less than the number provided on the table ( e.g. Z=4.44 will lead to: 0.001 so… P < 0.001.)

Question 6

When comparing small means, which distribution table is used?

Answer 6

t-distribution table ( not the normal distribution)

Question 7

What us true of your results when the P-value is small (P<0.05)

Answer 7

results are UNLIKELY when the null hypothesis is true

Question 8

What us true of your results when the P-value is large

Answer 8

results are LIKELY when the null hypothesis is true

Question 9

P values range from

Answer 9

0 - 1

because they are a probability

Question 10

Define: False positive (with resect to the null hypothesis)

aka: a type … error

Answer 10

rejecting the null hypothesis when it is true

Type 1 error/ (a)

Question 11

Define: False negative (with resect to the null hypothesis)

Answer 11

not rejecting the null hypothesis when it is false

Question 12

Define: the Power of study

and equation

Answer 12

the probability of rejecting the null hypothesis when it is actually false

Power= 1-Beta

Question 13

P-values indicate whether the result is

Answer 13

statistically significant

Question 14

CI indicate whether the result is (2 things)

Answer 14

statistically significant ( does is cross 0?)
clinically relevant  (Does it cross the "clinically important" boundary?)

Question 15

P value = 0.05 is statistically significant , so this means…

Answer 15

there is sufficient evidence to reject the null hypothesis and accept the alternative hypothesis

Question 16

P value >/= 0.05 is not statistically significant , so this means…

Answer 16

there is insufficient evidence to reject the null hypothesis
(cannot conclude that the null hypothesis is true)

Question 17

strength of evidence: P-value >0.10

Answer 17

little or no evidence of a difference or a relationship

Question 18

strength of evidence: P-value 0.05 - 0.10

Answer 18

weak evidence of a difference or a relationship

Question 19

Strength of evidence: P-value 0.01 - 0.05

Answer 19

evidence of a difference or relationship

Question 20

strength of evidence: P-value < 0.01

Answer 20

strong evidence of a difference or relationship

Question 21

strength of evidence: P-value< 0.001

Answer 21

very strong evidence of a difference or relationship.

Question 22

Basic definition: Null hypothesis

Answer 22

There is no difference in outcomes between Group A and B

Question 23

Basic definition: Alternative hypothesis

Answer 23

There is a difference in outcomes between Group A and B

two tailed hypothesis - when you’re not sure if the change will be an increase or decrease