Formula for sensitivity:
TP / (TP+FN)
Formula for specificity
TN / (TN+FP)
Formula for PPV
TP / (TP+FP)
(all are "positive" test results)
Formula for NPV:
TN / (TN+FN)
What does the null hypothesis state?
That there is NO difference between to groups.
What is type 1 or α error? How does it relate to the null hypothesis?
αlpha error is αrrogant: assumes a difference between 2 populations when there is none: incorrectly rejects null hypthesis.
What is type II or β error?
β error is humβle: assumes no difference in populations when there is one: incorrectly accepts null hypothesis.
What is the significance of p < .05?
95% certainty that the observed difference between two groups is real (5% chance that it is only due to chance)
How can you use a 95% confidence interval to determine whether a difference is real?
The further the CI is from 1.0, the more likely it is real. If it "crosses" 1, it is NOT statitstically significant.
most frequently occurring number
the middle number
3 types of QUANTITATIVE variables:
1. Students t-test
2. Paired t-test
What is the ability to detect disease in a population?
What is the ability to state that disease is not present?
What is the likelihood that a positive result is real?
What is the likelihood that negative result is negative?
Formula for accuracy:
(TP + TN) / (TP + TN + FP + FN)
What is the impact of prevalence on Sensitivity?
What is the impact of prevalence on PPV?
As prevlance goes up, so does PPV.
What is the formula for the false positivity rate?
1 - specificity
What is the formula for the false negative rate?
1 - sensitivity
Statistics for the ABSITE