In: Statistics and Probability
Lie Detector: Suppose a lie detector allows 20% of all lies to go undetected.
(a) If you take the test and tell 10 lies, what is the probability of having 5 or more go undetected? Round your answer to 3 decimal places.
(b) Would 5 undetected lies be an unusually high number* of
undetected lies? Use the criteria that a number (x) is unusually
large if P(x or more) ≤ 0.05.
-Yes, that is an unusually high number of undetected lies.
-No, that is not an unusually high number of undetected lies.
...............................................
*Unusually low and high number of successes: x successes among n trials is an unusually high number of successes if P(x or more) ≤ 0.05. x successes among n trials is an unusually low number of successes if P(x or fewer) ≤ 0.05.
Given that a lie detector allows 20% of all lies to go undetected
so p = 0.2
Here the number of undetected lies (X) follows a binomial distribution since there are only 2 outcomes possible here, either detected or undetected
Question (a)
If you take the test and tell 10 lies, what is the probability of having 5 or more go undetected
We need to find P(X 5)
The number of successes ''x'' in ''n'' trails in binomial distribution = * px* (1-p)p-x
= n! / [ x! (n-x)! ] * px * (1-p)n-x
Here n = 10, x = 5
p = 0.2
P(X 5) = P(X = 5) + P(X = 6) + P(X = 7) + P(X = 8) + P(X = 9) + P(X = 10)
= 10! / [ 5! (10-5)! ] * 0.25 * (1-0.2)10-5 + 10! /10! / [ 6! (10-6)! ] * 0.26 * (1-0.2)10-6 + [ 7! (10-7)! ] * 0.27 * (1-0.2)10-7 + 10! / [ 8! (10-8)! ] * 0.28 * (1-0.2)10-8 + 10! / [ 9! (10-9)! ] * 0.29 * (1-0.2)10-9 + 10! / [ 10! (10-10)! ] * 0.210 * (1-0.2)10-10
= 0.026424 + 0.005505 + 0.000786 + 0.000073 + 0.0000041 + 0.0000001
= 0.032793
= 0.033 rounded to 3 decimal places
If you take the test and tell 10 lies, what is the probability of having 5 or more go undetected = 0.033
Question (b)
Here P(X 5) = 0.033 which is less than or equal to 0.05
So Yes, that is is an unusually high number of undetected lies.