Statistics 1: Hypothesis Testing Using The Binomial Distribution
Statistics 1: Hypothesis Testing Using The Binomial Distribution
Statistics 1: Hypothesis Testing Using The Binomial Distribution
H0 : p = 0.25
H1 : p > 0.25
(ii) Let X be the number of young people in a sample of 30 who fail the test.
(iii) Let p be the probability that a young person fails the test.
H0 : p = 0.4
H1 : p ≠ 0.4
For the lower tail critical region, need the highest value of a such that
P(X ≤ a) < 0.025.
From tables, P(X ≤ 3) = 0.0160
P(X ≤ 4) = 0.0510
so the highest value of a is 3.
For the upper tail critical region, need the lowest value of b such that
P(X ≥ b) < 0.025 ⇒ 1 − P( X ≤ b − 1) < 0.025
⇒ P( X ≤ b − 1) > 0.975
From tables, P(X ≤ 11) = 0.9435
P(X ≤ 12) = 0.9790
The lowest value of b – 1 is 12, so the lowest value of b is 13.
30! 1 6 4 24
= (5 ) (5 )
6!24!
= 0.179 (3 s.f.)
= 0.999 (3 s.f.)
This probability is very nearly 1, so it is almost certain that at least one
packet contains a voucher.
H0 : p = 1
5
H1 : p ≠ 1
5
For the lower tail, need the highest possible value of a for which
P(Y ≤ a) < 0.025.
From tables, P(Y ≤ 0) = 0.0687
so the lower tail is empty.
For the upper tail, need the lowest possible value of b for which
P(Y ≥ b) < 0.025 ⇒ 1 − P(Y ≤ b − 1) < 0.025
⇒ P(Y ≤ b − 1) > 0.975
From tables, P(Y ≤ 4) = 0.9274
P(Y ≤ 5) = 0.9806
The lowest possible value of b – 1 is 5, so the lowest possible value of b
is 6.
Y = 0 which does not lie in the critical region, so accept H0. There is not
sufficient evidence to suggest that the proportion of packets containing
vouchers is different from 51 .
H0 : p = 0.2
H1 : p < 0.2
The alternative hypothesis takes this form because we are looking only at
a possible decrease in the proportion of vehicles with defective tyres.
(v) For H0 to be rejected, P(Y ≤ 1) must be less than the significance level. So
the significance level must be above 9.91%. So at the 10% level, H0 would
be rejected.