Tutorial 6
Question 1:
An FYP project requires students to survey 300 households around Castletroy to see
how many have an active subscription to Disney+. Of the households surveyed, 180
had Disney+ subscriptions.
a. Construct 95% and 99% confidence intervals, to 3 decimal places, for the
proportion of households with a Disney+ subscription.
b. What sample size is required to estimate this proportion to within 1% with a
confidence level of 95%, assuming no prior information about this proportion?
Answers:
n = 300, x = 180, 𝑝̂ = 180/300 = 0.6
a. 95% CI
%&(() %&) !.-(()!.-)
𝑝̂ ± 𝑧!.!#$ % ,
= 0.6 ± 1.96% .!!
= 0.6 ± 1.96(0.028) = 0.6 ± 0.055 =
(0.545, 0.655)
99% CI
%&(() %&) !.-(()!.-)
𝑝̂ ± 𝑧!.!!$ % ,
= 0.6 ± 2.58% .!!
= 0.6 ± 2.58(0.028) = 0.6 ± 0.072 =
(0.528, 0.672)
b. 1 − 𝛼 = 0.95, 𝛿 = 0.01
%&(() %&)
𝑧!.!#$ % ,
≤𝛿
!.$(() !.$)
1.96% ,
≤ 0.01 Use 𝑝̂ = 0.5 as this is the max on top after square root
0.5
1.96 ×
≤ 0.01
√𝑛
0.98
≤ 0.01
√𝑛
0.98 ≤ 0.01 √𝑛
0.98
≤ √𝑛
0.01
0.98 #
6 7 ≤𝑛
0.01
9604 ≤ 𝑛
Require a sample size of 9,604 to estimate this proportion to within 1% with a
confidence level of 95%
Question 2:
Mary O’Sullivan is a county councillor. With the upcoming local elections, she wants
to see the support for her political party in her constituency. In a random sample of
300 constituents, she finds 34% support her and her party. Construct a 95%
confidence interval for the support she in fact has.
Answers:
n = 300, 𝑝̂ = 0.34
%&(() %&) !../(()!../)
𝑝̂ ± 𝑧!.!#$ % ,
= 0.34 ± 1.96% .!!
= 0.34 ± 1.96(0.027) = 0.34 ± 0.053 =
(0.287, 0.393)
Question 3:
A masters student is conducting their research, where they survey 200 students and
asked them about the average time they spend on TikTok monthly. The mean time
spent on TikTok was 72 hours with a standard deviation of 3 hours.
a. Construct 90% and 97% confidence intervals for the average time spent on
TikTok in the whole population.
b. Find the sample size required to estimate the average time to within 1 hour
with a confidence level of 95%.
c. Is the sample representative of the population? Explain your answer.
Answers:
n = 200, 𝑥̅ = 72, s = 3
a. 90% CI
𝜎
𝑥̅ ± 1.64
√𝑛
3
72 ± 1.64 ×
√200
72 ± 1.64(0.212) = 72 ± 0.348 = (71.652, 72.348)
97% CI
𝜎
𝑥̅ ± 2.17
√𝑛
3
72 ± 2.17 ×
√200
72 ± 2.17(0.212) = 72 ± 0. 46 = (71.54, 72.46)
b. n = ?
3
1.96 × ≤1
√𝑛
5.88
≤1
√𝑛
5.88 ≤ 1 √𝑛
5.88
≤ √𝑛
1
5.88 #
6 7 ≤𝑛
1
𝑛 ≥ 34.57
𝑛 ≥ 35
c. Sample is not representative of the total population. The sample only contains
students. The average time a student spends on TikTok is far greater than the
average time a grandparent (excluding Grandad Joe) would spend on TikTok.
Question 4:
Guinness have their bottle filling machine set to put 330ml into each bottle. You
randomly take 5 bottles to sample the performance of the machine. The volume in
the 5 bottles is given below.
Bottle Volume (ml)
1 322
2 341
3 331
4 325
5 318
a. Calculate the mean volume in the sampled bottles.
b. Construct a 95% confidence interval, given that the standard deviation is
8.96ml.
c. From this confidence interval, does it appear that the machine is working
correctly?
Answers:
a. (322 + 341 + 331 + 325 + 318)/5 = 327.4ml
b. t = 𝑡!.!#$,/ = 2.776
1.2-
327.4 ± 2.776A B = 327.4 ± 2.776(4) = 327.4 ± 11.108 = (316.292, 338.508)
√$
c. We are 95% confident that the true value lies within the interval (316.292,
338.508). Since this interval contains 330, it appears that the machine is
working as expected.
Question 5:
Lidl recently opened a new store on the Dock Road in Limerick City. As part of the
new store opening, the management team want to estimate the average spend of
credit card users on their first visit. As the analyst, you’ve compiled the data and have
chosen 15 credit card transactions at random. From these transactions, you find the
mean spend in €100.55 with an associated standard deviation of €12.43. Assuming
the transaction value follow a normal distribution, construct a 95% confidence interval
for the average spend by credit card users on their first visit to Lidl’s new store.
Answers:
n = 15, since n < 30 we must use the t-tables. df = 15 -1 = 14 (degrees of freedom),
𝑡!.!#$,&' = 2.145 (table below).
𝑠
CI: 𝑋D ± 𝑡0.025,14 𝑛
√
95% CI: 100.55 ± 2.145 (12.43/Ö15)
100.55 ± 6.88 = (93.67, 107.43)
We are 95% confident that the true mean spend by first time credit card shoppers in
Lidl on the Dock Road in Limerick lies between €93.67 and €107.43.
Question 6:
The Irish marketing division of the shoe shop Schuh, ran two different advertising
campaigns at 20 different stores during the second quarter of the year. That is, 10
stores had one campaign and 10 stores had another. During this time, they measured
the average sales produced by each campaign at each store at the end of the quarter,
the results of which are in the table below.
Campaign A Campaign Z
Mean €160.25 €320.89
Standard Deviation €70.41 €60.29
Using this, construct a 95% confidence interval for the two campaigns. What
conclusions can you draw about the campaigns from the confidence intervals?
Answers:
𝑠
CI: 𝑋D ± 𝑡0.025,9 𝑛
√
Campaign A: 160.25 ± 2.262 (70.41/Ö10) = 160.25 ± 2.262(22.266) = 160.25 ± 50.37
= (109.88, 210.62)
Campaign Z: 320.89 ± 2.262 (60.29/Ö10) = 320.89 ± 2.262(19.065) = 320.89 ± 43.13
= (277.76, 364.02)
Since the intervals don’t overlap, we can say that at the 95% confidence level, there
is a difference between the two campaigns. From the data gathered, it would appear
that Campaign Z generates a higher valued sale than Campaign A.
