POPH192 - Lecture 24

Question 1

what can p values do?

Answer 1

help decide if the study is due to chance or not

• the probability that the study result is wrong and is due to chance/sampling error

Question 2

definition of p values?

Answer 2

probability of getting a study estimate (or one further from the null), when there really is no association, just because of sampling error

Question 3

relationship between p value and chance?

Answer 3

the higher the p value, the more likely the association is due to chance

Question 4

what does a small p value (<0.05) mean?

Answer 4

small chance the association found is incorrect (due to chance) and there is no true association

Question 5

what does a large p value (>0.05) mean?

Answer 5

a larger chance the association found is incorrect (due to chance) and there is actually no association

Question 6

what is the p value if the 95% CI includes the null value?

Answer 6

p >0.05

Question 7

what is the p value if the 95% CI doesn’t include the null value?

Answer 7

p < 0.05

Question 8

what significance is it when p > 0.05?

Answer 8

not statistically significant

Question 9

what significance is it when p<0.05

Answer 9

statistically significant

Question 10

what are the 2 types of hypothesis of p values?

Answer 10

• null hypothesis (Ho)
• alternative hypothesis (Ha)

Question 11

what is the null hypothesis?

Answer 11

• there is no association between exposure and outcome

Question 12

what is the alternative hypothesis?

Answer 12

• truly is an association between exposure and outcome?

Question 13

what happens to hypothesis when p > 0.05?

Answer 13

• fail to reject null hypothesis
• reject alternative hypothesis
• not statistically significant

Question 14

what happens to hypothesis when p < 0.05?

Answer 14

• reject null hypothesis
• accept alternative hypothesis
• statistically significant

Question 15

useful flow chart :)))

Answer 15

Question 16

what is the interpretation of the study when the p value is given and is less than 0.05?

Answer 16

• since the p-value is less than 0.05 the association is statistically significant
• we reject the null hypothesis and accept the alternative hypothesis
• chance is unlikely an explanation of study finding

Question 17

what is the interpretation of study when you assume the p value from the 95% confidence interval and it is less than 0.05?

Answer 17

• this is a statistically significant association at the 0.05 level as the 95% confidence interval doesn’t include the null value of 1
• we reject the null hypothesis and accept the alternative hypothesis
• chance is an unlikely explanation of study finding

Question 18

what is the interpretation of the study when the p value is given and it is greater than 0.05?

Answer 18

since the p-value is greater than 0.05, the association is not statistically significant

• we fail to reject the null hypothesis and reject the alternative hypothesis
• chance is a likely explanation of study findings

Question 19

what is the interpretation of the study when you assume p value from the 95% CI, and it is greater than 0.05?

Answer 19

this is not a statistically significant association at the 0.05 level as the 95% confidence interval does include the null value of 1

• we fail to reject the null hypothesis and reject the alternative hypothesis
• chance is a likely explanation of study finding.

Question 20

what is type I error?

Answer 20

when the study finds a significant result when there truly isn’t one (null hypothesis is true)

Question 21

when is risk of type I error reduce?

Answer 21

when small p values are used

Question 22

what is type II error?

Answer 22

when you don’t find a statistically significant result when there truly is one

(alternative hypothesis is true and there is statistically significant association)

Question 23

when is type II error more likely?

Answer 23

with small sample size

Question 24

what type of clinical importance and statistical significance are the points labelled A-E on the graph?

Answer 24

A - not clincally important, not statistically significant
B - statistically significant, not clincally important
C - statistically significant, possibly clinically important
D - not statistically significant, possible clincally important
E - statistically significant, clinically important