Business Statistics: Module 4.

Z-Test Page 1 of 7

Module 4. Fundamentals of Hypothesis Testing Tools (Z-test (one sample test;

one-tail and two-tail tests)

 Hypothesis – educated guess which needs to be test with the use of statistics.

 Hypothesis testing – usually starts with an assumption regarding a specific concern

of a population.

 Inferential statistics are used to test the hypotheses on the presence of significant
difference or relationship between and among variables or groups. As previously
mentioned in Module 1, inferential statistics are classified into:

 Parametric statistics – are used to test hypotheses if the data are numerical or
quantitative and normally distributed. Parametric statistics include z-test, t-test,
ANOVA, Pearson r correlation, and regression analysis among others.

 Non-parametric statistics – are used to test hypotheses if the data are categorical
or qualitative and non-normally distributed; as well as when the sample is small.
Non-parametric statistics are: Chi-square test, Spearman rho correlation,
McNemar test, Mann Whitney U test, Wilcoxon rank sum test, and Kruskal Wallis
test.

 Two types of hypothesis: null hypothesis and alternative hypothesis

 Null hypothesis (Ho) – assumptions, beliefs, supposition on a certain issue under

study. It is usually stated in negative sentence; but there are instances it is stated
positively but it means no difference or no relationship.

For a test of difference, these terms are usually use: there is no difference,
standard, current practice, norm, same, similar, equal. For example,
 The mean sales of the two branches is equal.
 The mean content of a soda can is 7.99 ounces.
 There is no significant difference on the academic performance of high school
graduates from public and private schools.

For a test of relationship: there is no relationship, independent. For example,

 There is no significant relationship between students’ study habits and their
academic performance.
 Career growth is independent of thee employees’ education background.

 Alternative hypothesis (H1 or Ha) – opposite of null hypothesis

For a test of difference: there is difference, different, not equal. For example:
 There is significant difference in the math anxiety levels of boys and girls.
 The average daily withdrawals from ATM is different from P35,000.
Business Statistics: Module 4. Z-Test Page 2 of 7

For a test of relationship: there is relationship, dependent. For example:

 Leadership skills of administrators is dependent on their gender.
 There is significant relationship between marketing and sales.

 Hypothesis can either be directional or non-directional

 One-tail hypothesis – the hypothesis has one direction, either it is going to the
left/negative tail or to the right/positive tail. There will be terms in the hypothesis
which denote its direction. Examples: less than, greater than, higher, lower,
below, at most, at least, reduced, and other similar words. For example:
 The average content of a one-liter soda is less than 1 liter. (left tail)
 Aggressive marketing leads to higher revenues. (upper tail)

 Two-tail hypothesis – the hypothesis has no direction. There are terms in the
hypothesis that do not shows direction. Examples: there is no difference, there is
difference, not different from, different from, same, similar, equal, not equal, there
is no relationship, there is relationship, independent, dependent. For example,
 There is significant difference on the average sales of the two branches.
 There is no significant relationship between age and reading comprehension
levels of the students.

 In rejecting or accepting the null hypothesis, two types of error may occur.

 Type I error – you commit the type 1 error when the null hypothesis is true and
should be accepted; but instead, you rejected it.

 Type II error – you commit the type II error when the null hypothesis is false and
should be rejected; but you accepted it instead.

For example: in any court hearing, the defendant is presumed innocent until
proven guilty. Based on this, the null and alternative hypotheses are as follow:
Ho: The defendant of a bribery case is not guilty.
H1: The defendant of a bribery case is guilty.

Adequate evidences proved that the defendant is not guilty of bribery and should
be acquitted. The judge, however, convicted the defendant. Type I error is
committed here.

Type II error occurred when there were enough evidences to prove that the
defendant committed bribery, and should be convicted. However, the judge
acquitted the defendant.

 Level of significance (α) – also known as alpha, margin of error; probability of

rejecting the null hypothesis. The alpha is usually given in the problem. If it is not
given and the confidence level is given, subtract the confidence level from 1 to get
Business Statistics: Module 4. Z-Test Page 3 of 7

the alpha. In addition, if the levels of significance and confidence are both not given,
then use the implied alpha, which is
.05.

 Confidence level – probability that

scores would be within the range of
confidence limits.

 Confidence limits – set the boundaries

between rejection and non-rejection
regions

 Rejection regions – shaded tails of the

normal curve; level of significance
http://pinkmonkey.com/studyguides/subjects/stats/c
 Non-rejection region – unshaded hap8/s0808701.asp
middle part, which is between the two
shaded tails of the normal curve

 Critical value (CV) – the basis for rejecting or not rejecting the null hypothesis

Z-Test (one sample test) – is used if population standard deviation (δ) is known

Z = (xbar - ɥ) ɥ = population mean xbar = sample mean

δ δ = population standard deviation
n

Problem illustration

1. The manager of a certain bank regularly monitors the ATM and stocks with sufficient
money so clients can withdraw money particularly during weekends. He would want
to make sure that the ATM will not run out of cash to meets the clients’ daily
withdrawal needs. Based on bank records, the average daily withdrawal of clients is
P188,000 with standard deviation of P9,875. For 30 days, he observed the daily
ATM withdrawals, which was noted at P193,000. At a .05 level of significance, is the
daily ATM withdrawals different from P188,000?

Let us analyze the problem. The focus of the problem is the daily ATM withdrawals.
The manager wants to find out if there is a changed in the amount of daily ATM
withdrawals, which is why he did the observation for 30 days. How do we know
which are the population mean and population standard deviation from the given
information? Usually, population mean and population standard deviation are based
on existing records or historical data. In this problem, population mean is P188,000
and population standard deviation is P9,875; which means P193,000 is the sample
mean from number of 30-day observations.
Business Statistics: Module 4. Z-Test Page 4 of 7

Now, we are ready to state the hypothesis. Your guide for writing your hypothesis is
the given question in the problem. That particular question is your alternative
hypothesis; you just need to state it in declarative form; and the population mean will
always appear in the hypothesis – be it null or alternative hypothesis. Let us
proceed.

Null hypothesis: The average daily ATM withdrawals is P188,000.

Alternative hypothesis: The average daily ATM withdrawals is different from
P188,000.

Note that both hypotheses show no direction, therefore, we have two tail hypothesis
in this problem. For a two tail hypothesis, we divide the given level of significance or
alpha by 2, to represent the two tails.

Level of significance: .05/2 = .025

Next, determine the critical z-value for a two-tail @ .025 level of significance from the
critical z-value table before the references section.

Critical z-value = + 1.96 Since its two tail hypothesis, you need to have
negative and positive critical values; hence, +1.96

Then, draw the normal curve and plot the critical z-values on the curve. Since it’s a
two tail hypothesis, you need to shade the two tails of the curve.

Normal curve
-1.96 1.96

After wards, write your decision rule.

Decision rule: Reject null hypothesis if Z < -1.96 or Z > +1.96; otherwise, do not
reject the null hypothesis

The underline terms/phrases in the decision rule are constant in every decision rule;
what will vary or change are the tool you use and the critical value/s.

Now, we are ready to calculate the z-value

Z = (xbar - ɥ) = (193,000 – 188,000) = 2.77

δ 9,875 _
n 30

After computing the z-value, you now have to state your conclusions; and give your
recommendations.
Business Statistics: Module 4. Z-Test Page 5 of 7

Conclusion: Reject the null hypothesis, 2.77 > 1.96; which means there is enough
evidence to support that daily ATM withdrawals is different from P188,000.

Recommendation: Based on the results, the manager needs to load more money in
the ATM than what they usually do so as to assure that it will not run out of cash
when clients make withdrawals from it. (You can give more recommendations aside
from the given example.)

2. A sales manager of a telecommunication reviewed the company’s records on

domestic call and learned that a subscriber spent a maximum average of 12 minutes
per domestic call with a standard deviation of 1.5 minutes. He would like the
subscribers to make longer domestic calls to generate more revenues. He promoted
this particular services to the public to encourage subscribers to more longer
domestic calls. After few months, a randomly select 35 domestic calls registered an
average of 13 minutes. At a .01 alpha, is there average length of domestic calls
longer than 12 minutes?

Note that the hypothesis is a one-tail because of the terms maximum and longer

Null hypothesis: The average length of domestic calls is less than or equal to 12
minutes
Alternative hypothesis: The average length of domestic calls is longer than 12
minutes.

Level of significance: .01 since it’s a one-tail hypothesis, no need to divide it by 2

Critical value: +2.33 the critical value is positive since its upper tail (longer)

Normal curve
2.33
Decision rule: Reject the null hypothesis if Z > 2.33; otherwise, do not reject the
null hypothesis.

We use the > symbol because of the word longer in the alternative hypothesis.

Computation: Z = (xbar - ɥ) = (13 – 12) = 3.94

δ 1.5 _
n 35

Conclusion: reject the null hypothesis, 3.94 > 2.33; which means that subscribers
make domestic calls longer than 12 minutes.

Recommendation: It seems that the promotional strategy has resulted to an

increased in length of domestic calls. The sales manager should continuously with
more and better promotions in order to entice more subscribers to spend longer
times on domestic calls.
Business Statistics: Module 4. Z-Test Page 6 of 7

End of Module Exercises

Read and analyze the problems carefully. Work on the problem and follow the step by
step procedure in solving it – from the null hypothesis to the recommendation.

1. A properly maintained and adjusted coffee machine dispenses 8 ounces of coffee

per cup with standard deviation of 0.30 ounce. The randomly select sample of 30
cups has mean content of 7.95 ounces. At a .05 alpha, is the coffee machine
dispensing less than 8 ounces of coffee per cup?

2. A group of female management students pooled their resources together and put up
a coffee shop last year; and its average daily sales was P15,650 with standard
deviation of P1,995. For the current year, the partners would want to know if their
daily sales have improved. They studied their sales for the past 30 days and noted
a mean daily sales of P16,250. At a .01 level of significance, is the average daily
sales of the coffee shop is higher than P15,650?

3. A local pizza shop delivers pizza to dormitories within the university belt to cater to
the needs of the students who reside in the dorms. Mean delivery time is 15
minutes with standard deviation of 2 minutes. However, the shop has been
receiving complaints of late deliveries lately. So the owner randomly selected 25
orders and recorded the average delivery time of 17 minutes. At a .05 level of
significance, is the average delivery time of pizza to dormitories longer than 15
minutes?

4. A copier machine operator claimed that he makes an average number of copies of at

least 5000 per day with standard deviation of 75 copies. To make sure that he has
enough supplies and understocking, he monitored the number of copies he
produced for 30 days which was noted at an average of 5105. At a .05 level of
significance, is the average number of copies produced per day is different from
5000?

5. A Social Media Agency reported that college students spent an average of 30 hours
per week on Facebook with standard deviation of 2.5 hours. To find out if this still
true, a new survey was conducted among 50 students and reported an average of
28 hours spent on Facebook. At a .01 level of significance, is there average number
of hours which college students spent on Facebook different from 30 hours?

Critical Z-values table

Level of significance One tail Level of significance Two tail
.01 2.33 .005 +2.575
.05 1.645 .025 +1.96
.10 1.28 .05 +1.645
Business Statistics: Module 4. Z-Test Page 7 of 7

References

Albright, S. et al. (2015). Business analytics: data analysis and decision making (5th
ed). Cengage Learning.
Anderson, D., Sweeney, D.J., et.al., (2018). Modern business statistics. Australia:
Cengage Learning.
Antivola, H. (2015). Business statistics: a modular approach. Books Atbp. Publishing.
Anywhere Math. (2016). Introduction to Statistics.
https://www.youtube.com/watch?v=LMSyiAJm99g.
Berenson, M.L., Levine, D.M., & Krehbiel, T.C. (2015). Basic business statistics:
concepts and applications. Pearson Education Sou7th Asia Pte. Ltd.
Bowerman, B. (2017). Business statistics in practice: using modeling, data, and
analytics (8th ed.). McGraw-Hill Education.
Jaggia, S. (2019). Business statistics: communicating with numbers (3rd ed.). McGraw-
Hill Education.
Lee, N. (2016). Business statistics: using excel & SPSS. Sage.
Simple Learning Pro. (2015). Mean, median, mode, range, and standard deviation.
https://www.youtube.com/watch?v=mk8tOD0t8M0.
Sharpe, N. (2015). Business statistics 3rd ed. Pearson Education.
Willoughby, D. (2015). An essential guide to business statistics. John Wiley & Sons.