Sepideh P. answered • 06/24/21
Pretty Fly for a Data Sci(entist)
A good place to start is to review the definitions of sensitivity and specificity. Sensitivity, or true positive rate, is the proportion of positive cases that are correctly classified as positive: "how many of the positives are detected?" Specificity, or true negative rate, is the proportion of negative cases correctly classified as negative. I like to think of specificity as how "specific" the model is when it comes to correctly labeling an actual negative as a negative: "how many of the negatives are detected?"
If the test classifies everyone as positive (having the disease), this forcibly means anyone who is actually positive for the disease will be classified as positive. Think of how this would be represented as a proportion between 0 and 1 for sensitivity.
Note that specificity is a proportion involving cases that are classified as negative - are there any cases in this scenario being classified as negative? What is the numerator when you calculate specificity?
Hope this helps you get to the answer!
Shana C.
I also don’t understand how to set this up as a math equation…06/26/21
Sepideh P.
You can use the equations sensitivity = TP / (TP + FN) and specificity = TN / (TN + FP) where TP = true positive, TN = true negative, FP = false positive, FN = false negative. In this case, no samples are classified as negatives, so you have 0 true negative or false negative classifications.06/28/21
Sepideh P.
I added another way to think about it to my answer06/28/21
Shana C.
I am still kind of confused on this whole aspect06/26/21