L10-recruitment

L10-recruitment - Recruitment Example Predicting...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
Recruitment Example Predicting Money-Makers Using Aptitude Tests TP = True Positive, FP = False Positive, TN = True Negative, FN = False Negative Historic Population Current Population True and False Positives and Negatives Made Money? Is… 1 Yes Yes TP 1 ? ? Yes Yes TP 2 Yes Yes TP ? ? Yes No FP 3 Yes Yes TP ? ? No Yes FN 4 Yes Yes TP ? ? No No TN 5 Yes No FP ? ? To determine "TN", "FP", "FN", or "TN", do this: 6 No No TN ? ? 7 No No TN ? ? So, first letter is "T" if test = reality, and is "F" otherwise. 8 No Yes FN ? ? ? ? 100 ? ? The aptitute test predicts if person is money-maker or not. STEP 1. Confusion Matrix Count true- and false- positives and negatives. Yes No Totals "Yes" 4 1 5 "No" 1 2 3 Totals 5 3 8 STEP 2. Historic Population Proportions STEP 4. Estimated Current Population Proportions Computed from counts & totals in Confusion Matrix Yes No Totals Yes No "Yes" 50% 13% 63% "Yes" ? ? "No" 13% 25% 38% "No" ? ? Totals 63% 38% 100% Totals 10% 90% 100% Prior Probabilities are P(Reality) for each state of reality. STEP 3. Reliabilities / Likelihoods STEP 5. Joint Probabilities (Historic Accuracy of our Tests) Probability we said it AND it was so. Yes No Yes No "Yes" 80% 33% "Yes" 8.0% 30.0% "No" 20% 67% "No" 2.0% 60.0% 100% 100% 10% 90% 100% Reliabilities are P(Information | Reality) Employee Number Passed Test? Made Money? Candidate Number Passed Test? Made Money? Passed Test? Test is T RUE if and only if test and reality are the same. Test is P OSITIVE if and only if it answered Yes . So, second letter is "P" if test = Yes , and is "N" otherwise. "Yes" = We predicted Yes "No" = We predicted No Yes = It actually was Yes No = It actually was No Reality : Made Money Information : Passed Test To understand it best, you should consult this spreadsheet in conjunction with the lecture slides on Probability. Totals are subjective Prior Probability guesstimates Reality : Made Money Reality : Made Money Reality : Made Money Reality : Made Money Joint Probabilities are P(Information Reality).
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 04/19/2008 for the course OPIM 101 taught by Professor Lee during the Spring '08 term at UPenn.

Page1 / 8

L10-recruitment - Recruitment Example Predicting...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online