Statistics

Question 0

What is specificity?

Answer 0

True negatives/total # patients without disease.

True negatives/false positives + true negatives

Question 1

What is sensitivity?

Answer 1

True positives/total number of with disease.

True positives/true positives + false negatives

Question 2

What population does specificity take into account?

Answer 2

Only those without disease.

Question 3

What population does specificity take into account?

Answer 3

Only those patients without disease.

Question 4

What is prevalence?

Answer 4

The percentage of the population with disease.

Question 5

Both sensitivity and specificity are independent of what?

Answer 5

Prevalence

Question 6

Specificity = SPIN

Answer 6

SPecific test to rule IN a hypothesis.

Question 7

Sensitivity = SNOUT

Answer 7

SeNsitive to rule out a hypothesis

Question 8

What is positive predictive value?

Answer 8

The probability of disease in a patient with a positive test.

Question 9

Does positive predictive value reflect prevalence

Answer 9

Yes

Question 10

What is the formula to determine positive predictive value?

Answer 10

PPV = TP/(TP + FP)

Divide true positives by all who test positive.

Question 11

What is negative predictive value?

Answer 11

The NPV of a test is the probability of not having a disease if the test is negative.

Question 12

What is the formula for negative predictive value?

Answer 12

NPV = TN/(TN + FN)

Question 13

What is prevalence?

Answer 13

The fraction of the population that has disease.

Question 14

What does p-value mean?

Answer 14

P-value is the probability that an effect as large or larger than the observed effect would occur if in reality there was no true difference between the groups.

Question 15

If the p-value is .05 then the likelihood that the results are to chance is?

Answer 15

1 in 20 or 5%

Question 16

What is type 1 error?

Answer 16

Concluding there is a difference when there is actually NO difference.

Question 17

What is type 2 error?

Answer 17

Concluding there is no difference when one exists.

Question 18

What are confidence intervals?

Answer 18

Similar to p-values. For instance, a confidence interval (CI) of 95% is essentially p = < 0.05.

Question 19

If the CI in a meta analysis does not cross the vertical line used for signifance it means what?

Answer 19

The study shoes a significant response.

Question 20

What is meant by number needed to treat?

Answer 20

Number of people who need to be treated for a period of time to prevent one event.