4.17; Underage drinking pt. 1: Data collected by the Substance Abuse and Mental HEalth Service Administration (SAMSHA) suggests that 69.7% of 18-20 year olds consumed alcoholic beverages in any given year.
The binomial distribution is appropriate for this calculation because the sample of 18-20 year olds is fixed (10), and the probability that they’ve consumed alcohol is a success or failure event (consumed (P = 0.697) or not consumed (Q = 0.303))
Binomial formula: (N choose K) * P^K * Q^(N-K) N = 10 P = 0.067 K = 6 Q = 0.303
choose(10,6) * (0.697)^6 * (0.303)^4
## [1] 0.2029488
choose(10,4) * (0.303)^4 * (0.697)^6
## [1] 0.2029488
Idea: P(k = 0, or 1, or 2,…., or n) = P(k=0) + P(k=1) + … + P(k=n) = P(k = “at most”) Where P is calculated with the binomial formula:
(N choose K) * P^K * Q^(N-K) N = 5 P = 0.697 K = 0,1,2 Q = 0.303
p_zero <- choose(5,0) * (0.697)^0 * (0.303)^5
p_one <- choose(5,1) * (0.697)^1 * (0.303)^4
P_two <- choose(5,2) * (0.697)^2 * (0.303)^3
at_most_two <- p_zero + p_one + P_two
at_most_two
## [1] 0.1670716
Sum of all probabilities: P(k=1) + P(k=2) + P(k=3) + P(k=4) + P(k=5)
p_one <- choose(5,1) * (0.697)^1 * (0.303)^4
P_two <- choose(5,2) * (0.697)^2 * (0.303)^3
P_three <- choose(5,3) * (0.697)^3 * (0.303)^2
p_four <- choose(5,4) * (0.697)^4 * (0.303)^1
p_five <- choose(5,5) * (0.697)^5 * (0.303)^0
at_least_one = p_one + P_two + P_three + p_four + p_five
at_least_one
## [1] 0.997446
Via the compliment of nobody drinking
p_at_least_one_comp <- (1 - choose(5,0) * (0.697)^0 * (0.303)^5)
p_at_least_one_comp
## [1] 0.997446
As we compare both answers and methods above, we see it is much easier to find the compliment of nobody drinking, which is the same as asking/solving for the probability of at least one drinking alcohol.