Inferential statistics Hypothesis testing 1

PROPOSED EXERCISES
Taken from the book of Applied Statistics to Administration and Economics by Alfredo Díaz Mata

Hypothesis test for the mean.

1.- The production manager of a company knows that the average time taken for an operation is
5 minutes per person; for a performance evaluation, a random sample of 300 is taken.
operations and it is found that the average time is 4.95 min with a standard deviation
from 0.1 min. Determine if the established time for this process is being met with a
significance level of 0.04.

2.- The operations manual of a company has the rule that the number of products
The average daily defective pieces is 30. A random sample of 35 days yields a mean
of 29.5 defective products with a standard deviation of 2 products. Can it be considered
that the rule is not being complied with at a significance level of 0.06?

In a delegation, a campaign was implemented to reduce the daily water waste by

It is known that an average of 350 L is consumed. After carrying out the campaign, it was taken
a random sample of 1,500 people is found to have an average daily consumption of
345 L with a standard deviation of 13 L. Determine with a significance level of 0.01 if it has
has there been any variation in water waste.
Inferential statistics Hypothesis testing 2

4.- According to the latest results obtained, a restaurant has as data that its
customers spend an average of $185 per person, two months ago they changed the chef and the
Administration assures that diners are consuming less. A sample is taken.
random sample of 50 customers and it is found that the average consumption per customer is $140 with a
standard deviation of $25. Determine with a significance level of 0.01 if the management
you are right.

5.- As a security measure, a bank is promoting the use of cards to make purchases.
in a certain chain of stores, where there is an average record of $230 per day
payment with card. After the promotion, a random sample of 64 days is taken and it is found that
The average card purchases amount to 245 with a standard deviation of 28. Determine with
a significance level of 0.05 if there was a significant increase in the population average.

Hypothesis test for the difference of means

Inferential statistics Hypothesis testing 3

Large independent samples with unknown population variances and

considered different

6.- For the manufacture of a specific piece, 2 machines are used. A sample is taken from
40 pieces produced by both machines and it is found that the pieces produced by machine A
they have an average length of 83 mm with a standard deviation of 5 mm, while those of
Machine B has an average length of 82 mm with a standard deviation of 2 mm. Determine
if there is a difference in the length of the pieces manufactured by each machine, with a level of
significance of 5 percent.

7.- In a college, two degrees are offered, administration and accounting. 42 were asked.
administration students how many times they have consulted books in the library
during the last month, and it was found that on average there were 27 times with a standard deviation
of 4, while in the sample of 37 accounting students the average was 23 with a
standard deviation of 3. Check if there is a difference between the average of queries made
by the students of each degree with a significance level of 0.05.

8.- To test the combustion speed of 2 types of automotive oil, a sample was taken of
50 bottles of 2 different brands. In brand A, it was found that the average time of
combustion is 47.5 seconds, with a standard deviation of 3.2 seconds, and at mark B the time
The average is 49.4 seconds with a standard deviation of 3.7 seconds. Check if there is a difference.
between the combustion time of the 2 brands of oil with a significance level of 0.01.
Inferential statistics Hypothesis testing 4

Large and independent samples with unknown population variances

but considered equal
9.- A new organic fertilizer is being experimented with for wheat planting with the
purpose of testing whether the average growth of the plants in the plot sprayed with the product
it is the same or not as that of a plot that did not receive the new fertilizer. A sample of 45 was taken
sprouts that had contact with the fertilizer and 30 that did not. The data obtained is shown at
continuation. Test that hypothesis with a significance level of 5 percent:

Sample with fertilizer Sample without fertilizer

Sample size 45 plants 30 plants
media 65 cm. 60 cm
Standard deviation 8 cm. 4 cm.

10.- In a quality testing laboratory, it is desired to examine whether the average lifespan (in days)
the 2 brands of 100 watt bulbs are equal or not, with a significance level of 0.05. For this, it is
a sample of 30 bulbs from each brand was taken, the results are shown below:

Show brand A Show brand B

Sample size 30 30
media 2017 hours. 1950 hrs.
Standard deviation 173 hrs. 121 hrs.
Inferential statistics Hypothesis testing 5

11.- A government agency responsible for overseeing public health issues wishes to
to test whether the average nicotine content of cigarettes from a national brand and a foreign brand are
equal or not, with a significance level of 0.01. A sample of 50 cigars was taken from the
national brand and 30 of the foreign one. The obtained data is shown below.

Foreign brand National brand

Sample size 30 50
media 15.8 14.2
Standard deviation 2.4 1.8

Small and independent samples with unknown population variances

but considered equal
12.- A sample of 12 vacuum cleaners of the Bilmex brand showed that in their operation they consume

average of 44.84 kilowatt-hours with a standard deviation of 8.7 kilowatt-hours. While a

A sample of 17 vacuum cleaners from the brand Prix demonstrated that they use an average of 46.75 kilowatt-hours.

a standard deviation of 10.4 kilowatt-hours. If energy expenditure follows a normal distribution

Previous studies have shown that the variance of the 2 brands is equal, test the hypothesis.
that the energy consumption of Bilmex vacuum cleaners is lower than that of Prix vacuum cleaners
a significance level of 0.025.
Inferential statistics Hypothesis testing 6

13.- The personnel manager of a restaurant wants to evaluate whether the average amount of the
the tips that evening shift waiters receive are higher than those of the morning shift, for which I take
a sample of 5 waiters from each shift and record how much they received in tips daily
for a certain number of days. It was found that the morning shift waiters received in
an average of $632 with a standard deviation of $41.2; the evening shift waiters received
on average $685 with a standard deviation of $27.3. If tips follow a distribution
normal and it is known that the variances are equal, test the hypothesis at a level of
significance of 0.10.

14.- A purchasing manager is considering 2 types of lamps for the lighting of the plant.
assembler. She took a sample of 8 lamps of brand A and found that, on average, their
The duration time is 1,455 hours with a standard deviation of 87; of the lamp brand
B took a sample of 11 and found that on average they last 1,484 hours with a standard deviation
standard of 23. If the lifespan of the lamps follows a normal distribution and it is known that the
the variances are equal, test the hypothesis that lamp B lasts less than lamp A, with
a significance level of 0.05.
Inferential statistics Hypothesis testing 7

Small and independent samples with unknown population variances and

considered as different
15.- It is planned to use 2 types of batteries for a new model of mobile phone and it is desired
evaluate whether the duration of a charge of battery A is greater than that of battery B, for which
samples were taken with the following results:

A B
Sample size 11 13
media 38 months 42 months
Standard deviation 4.5 months 5.1 months

If it is not possible to assume that the variances are equal, demonstrate the hypothesis at a level of
significance of 0.025

16.- It is wanted to know if the average time that girls spend watching television daily is
older than the children. A sample of 10 boys and 10 girls was taken, the results are shown in
continuation:

Children Girls
Sample size 10 10
media 5.7 hours 4.8 hours
Standard deviation 2.2 hours 1.9 hours

If it cannot be assumed that the variances are equal, test the hypothesis at a level of
significance of 0.1.
Inferential statistics Hypothesis test 8

17.- In exercise 12, it was desired to test the hypothesis that energy expenditure in operation
the expense of Bilmex vacuum cleaners is less than that of Prix vacuum cleaners, and the obtained
following results of the samples:

Bilmex Price
Sample size 12 17
media 44.84 kilowatt-hours 44.75 kilowatt-hours
Standard deviation 8.7 kilowatt-hours 10.4 kilowatt-hours

If it is not possible to assume that the variances are equal, demonstrate the hypothesis that spending
the energy of Bilmex vacuum cleaners is lower than that of Prix vacuum cleaners with a level of
significance of 0.025.

Hypothesis testing for proportion

18.-A bottler has a rule that 25% of its containers must be made with
recycled material, so the production manager takes a random sample of 85 containers to
check that this requirement is being met, and find that 23.5% are made with pet
recycling. Taking into account a significance level of 0.03, determine if the rule is being
fulfilling in this company.
Inferential statistics Hypothesis testing 9

19.- The coordinator of the job placement office at a public university claims that at least 30% of
students who finish their studies get employed within 3 months. To prove this
affirmation, a sample of 50 students from that institution is taken and it is found that only 10
they found employment within the first 3 months after finishing their studies. Use
a significance level of 1% to test the hypothesis and say if it supports that statement.

20.- A teacher is preparing a group of students for a math marathon.

According to the curriculum, students must be in the workshop 20% of their school schedule;
A random sample of 55 students determines that of 8 hours of class, the students attend the workshop.
1 hr. Identify if the students are meeting the study time for preparation of the
marathon with a significance level of 0.01.

21.- A clothing factory normally buys fabric composed of 20% polyester and 80%
cotton to produce different garments; next month children's clothing will go on sale and it is
It is very important to reduce the percentage of polyester in the fabric, which is why they are changing suppliers.

A random sample of 300 m of fabric shows that its composition is 88% cotton and the rest
Inferential statistics Hypothesis testing 10

polyester. Was there a significant decrease in the amount of polyester in the fabric at a level of
significance of 0.01?

22.- An agricultural study shows that 75% of plant X grows in less than 2 months. It
began using a fertilizer with the intention of increasing this ratio. A sample
A random sample of 800 plants indicates that 420 grow in less than 2 months. Determine, with a level of
significance
of 0.01, if the percentage increased.

Hypothesis test for the difference of proportions

23.- A placement agency wants to analyze the attitudes of students from 2 universities
about the importance of salary in employment, for which a survey is conducted with 300 students
from a private university and 400 from a public one. In the private one, 280 respond that the most
important in employment is the salary, while in the public sector 240 respond that the salary is what
more important. Check the hypothesis that the proportions of students from the 2
universities that consider salary as the most important are the same, with a level of
meaning of 1 percent.
Inferential statistics Hypothesis testing 11

24.- As the election day approaches in an important city, a company

The polling company reviews the results of 2 surveys conducted on different dates and finds
that in the first survey where 500 citizens were interviewed, 51% of them expressed
to be in favor of the main candidate. In the subsequent survey, conducted among 800 voters, 47.5%
he expressed his support for that main candidate. With a significance level of 3%, it can be
to assert that the voters' preferences have changed?

25.- In an opinion poll conducted by a radio station, 60 out of 200 men said that
They liked a certain program, and 75 out of 300 women expressed the same opinion. With a
1% significance level, there is a difference between the proportions of men and women.
Do they like that radio program?

26.- A finance manager is analyzing the behavior of his accounts payable, he has
samples of accounts from the month of May of 2 consecutive years. In May of year 1, with
A sample of 1,300 accounts payable found 50 that had not been settled on time.
agreed, while in a sample of 1,000 accounts from May of year 2, there were 38 that did not
paid on time. With a significance level of 5%, can it be stated that there has been a
increase in the proportion of accounts payable that fall into delinquency?
Inferential statistics Hypothesis testing 12

27.- Changes were made in a production line to reduce the percentage of items.
defective. To evaluate whether the modifications were effective, samples of 300 were taken.
articles before and after making the changes, and it was found that before the percentage of
defective items were 4% and then turned out to be 3%. Check if the changes reduced the
proportion of defective items, with a significance level of 0.02.

Hypothesis test for variance

28.- The effectiveness in the bottling process of a beverage will soon be evaluated.
presentation of 1 L, the procedure has a variance of 0.000037 mL2. As a measure
Preventively, it is wanted to review that this indicator has not changed, so a sample is taken.
random sample of 81 bottles of the product and it is found that the variance is 0.000095 with a level of
Significance of 0.025, can it be assured that the variance is the same?

29.- In a distribution center, it was established that each worker had to pack 85 boxes in 90
minutes with a variance of 6; at the beginning of this period, 2 training courses were held
to reduce packaging time costs, for which it is expected that each worker
pack the same amount of boxes in less time, and as a consequence, the variance as well
decrease. A random sample of 65 workers indicates that on average they take 78 minutes
with a variance of 4 in packing the 85 boxes. Determine with a significance level of 0.10 if
the hypothesis that the variance in the times decreased should be accepted.
Inferential statistics Hypothesis testing 13

30.- A travel agency has a standard price to go to several destinations that are located at
less than 1,000 km from the capital, and that the variance of those trips is less than 64 km². One of the
new sellers assure that the agency has miscategorized the destinations, so it
Take a random sample of 35 trips that yields a variance of 65 km2. With a level of
significance of 0.10, should the seller's hypothesis be accepted, that is, recognize that the calculation
Is it done wrong?

31.- A company that specializes in the repair of electrical appliances recorded 6 months ago that,
on average, spends $120 on each job with a standard deviation of $5. The budget of
Next year it will be carried out by an independent firm and needs to verify the data, so take
a random sample of 40 jobs with a standard deviation of $4.90. With a level of
significance of 0.01, it can be accepted that the variance of the repair cost remains the
same?

32.- A bottled water distributor has established that the variance of the content of
Sodium in its 1.5 L presentation must be less than 0.0020 mg2. A random sample of 20
Inferential statistics Hypothesis testing 14

bottles of this presentation reflect that the variance is 0.0025. Determine at a level of
Significance of 0.005 if it is really being complied with as established.

Hypothesis testing for the relationship of variances

33.- A lighting manufacturer suspects that one of its production lines is manufacturing
products with too much variability in their shelf life. To assess the situation, he decides to compare the
variance of a sample of 25 of those lights against the variance of another sample of 30
coming from a production line that is not under suspicion. The
corresponding variances and it is found that the one from the suspicious production line is 900,
while the other is 350. It can be concluded, with a significance level of 1%, that the
variability of the useful life of the luminaires manufactured on the production line under suspicion is
greater than those that come from the production line that works properly?

34.- Twenty-one people aged between 20 and 35 who participated in a training course
based on a programmed text, they averaged 58.22 min to complete the course, with a variance
of 8.5 min. Meanwhile, in a simultaneous course for people over 35 years old, the average was
63.24 min with a variance of 9.2 min. Verify the hypothesis that the variance of the 2 courses
it is the same with a significance level of 5 percent.
Inferential statistics Hypothesis testing 15

35.- Experienced supervisors measure the variance of the cover thickness measurements.
41 plastic protectors recording that it is 1.39 cm. If a newly hired supervisor measures
a sample of 25 covers and finds that they have a variance of 1.47 cm2check with a
1% significance level the hypothesis that the new supervisor is making measurements that
they are not different from those of experienced supervisors.

36.- A random sample of 31 boxes of mangoes from a supplier's load yielded a variance
from 2.5 kg to its weight. A sample of 27 boxes of mangoes from another supplier reflected a
variance of 4.2 kg2Can the hypothesis that the variance of the weights of the 2 loads be accepted
Is it the same? Perform the test with a significance level of 1 percent.

37.- A sports commentator wants to know if the variance of the marathon completion time of the
Mexico City has changed since the last measurements which indicate, according to a sample
Random sample of 61 runners, a variance of one hour. A random sample of 30 was taken.
runners and a variance of 0.87 hours was observed. Conduct the test at a level of
significance of 0.05.
Inferential statistics Hypothesis testing 16

ANSWERS TO PROPOSED EXERCISES

Hypothesis test for the mean.

1.- H0: µ=5; H1µ≠5 z = -8.66 the null hypothesis is rejected and it is considered that the mean is not
it can be from 5.
3.- H0: µ=350; H1µ ≠ 350 z = -14.89 the null hypothesis is rejected and it is not considered that the
The average may be 350. If there is variation in water waste, (it has decreased).
5. H0: µ≤ 230; H1µ>230 z = 4.29, the null hypothesis is rejected, the mean is greater than 230.
Hypothesis testing for the difference of means.
7.- H 0:µ A=µc; H 1 :µA≠ µ c z = 5.06, the null hypothesis is rejected.
9.- H 0:µ c =µ s ; H1 :µc≠ µ s z = 3.16 the null hypothesis is rejected.
11. H:µ=µ
0 N ; H:µ≠µ
E 1 N E z = 3.38, the null hypothesis is rejected.
13. H:µ
0 ≤v µ;H:µ>
m 1µ v m t = 2.39, the null hypothesis is not rejected.
15. H:µ≤µ
0 B >µ
A ; H:µ 1 A B t = -2.04, the null hypothesis is not rejected.
17. H:µ≥µ0 B ; H:µ
P <1µ B P t = 0.025, the null hypothesis is not rejected.
Hypothesis testing for proportion. (P-value method)
19.- H0P = 0.30; H1P ≠ 0.30 z = -1.54, p = 0.123; the null hypothesis is not rejected.
21.- H0P ≥ 0.20; H1P < 0.20 z = -3.46, p=0; the null hypothesis is rejected.
Hypothesis testing for the difference of proportions. (P-value method)
23.- H 0:Ppr=P pu;H1 :Ppr≠ P pu z = 9.99, p = 0.00; the null hypothesis is rejected.
25. H 0:PH=P M;H1 :µHnotequaltoPM z = 1.23, p = 0.217; the null hypothesis is not rejected.
27.- H 0:PD≥PA;H1 :PD < P A z = -0.67, p=0.253; the null hypothesis is not rejected.
Hypothesis testing for variance. (P-value method)
2
29.- H0: σ≥6 ; H1: σ 2< 6 ; x2= 42.67, v=64, p=0.08; do not reject the null hypothesis.
31.- H0: σ 2=25 ; H1: σ≠25x 2
; 2= 37.46, v=39, p=0.919 fail to reject Ho.
Hypothesis test for the relationship of variances. (P-value method)
33.- H 0:σ 21≤ σ 22 ;H1:σ 12> σ 22 F=2.57 (V1=24, V2=29); p=0.008; the hypothesis is rejected
zero.
35.- H 0 :σ 21=σ 22; H 1 :σ 21 ≠ σ 22 F=1.057 (V1=24, V2=40); p=0.855; the null hypothesis is not rejected.
null hypothesis.
37.- H 0:σ 21=σ 22; H 1 :σ 21 ≠ σ 22 F=1.15 (V1=60, V2=29); p=0.695 the null hypothesis is not rejected.
null hypothesis.