Statistical inference may be divided into two major areas: estimation and tests of

hypothesis. In this module, we will focus on the tests of hypothesis.

Similar to estimation, the purpose of hypothesis testing is to make inferences about a
population based on the data contained in the sample drawn from the same population. To
ensure that these inferences are valid, we assume that the population is normally distributed
and a random sample is drawn from that population.

LESSON 1: BASIC CONCEPTS IN HYPOTHESIS TESTING

At the end of the lesson, the learner is able to:

• illustrate a statistical hypothesis

• differentiate a null hypothesis from an alternative hypothesis

• differentiate Type I from Type II error

• illustrate consequences of committing errors.

 TOPIC OUTLINE:

1. Definition of Statistical Hypothesis

2. The difference of null hypothesis from alternative hypothesis

3. Consequences of making a decision

4. Two possible errors in a test of hypothesis

Consider the pronouncement made by the government concerning the El Nin̄o

phenomenon:

”The country will experience El Nin̄o phenomenon in the next few months.”

Observe that

• the pronouncement is a claim that may be true or false. Such claim could be referred
to as a statistical hypothesis.

• There are two possible actions that one can do with the statement. These actions are
either to accept the statement or to reject it.

• Actions made in response to the statement have consequences.

Possible consequences of accepting that the statement is true include:
(a) increase the importation of rice in anticipation of supply shortage;
(b) buy materials for water storage; (c) use drought-resistant varieties of rice;
(d) invest in programs to make Filipinos ready; and the like.

On the other hand, when the statement is rejected because we think it is false, possible
consequences are
(a) we are not prepared for rice and water shortage;
(b) farmers experience great loss on production; or
(c) we do not do anything.
 • Some of the consequences are tolerable while other consequences are severe. Experi-
encing a few days of water shortage is tolerable but having rice shortage for a month or
two is unbearable. The degree of the possible consequence is the basis in making
the decision. If the consequences of accepting the claim that El Niño phenomenon is
going to happen are tolerable, then we may not reject the pronouncement. However,
if the consequences are severe, then we reject the claim.

• The validity of the statement could be tested based on some empirical facts. We may
also say that the validity of this statement can be accessed through a series of steps
known as test of hypothesis

A statistical hypothesis is a claim or a conjecture that may either be true or false.

The claim is usually expressed in terms of the value of a parameter or the distribution
of the population values.

test of hypothesis is a procedure based on a random sample of observations with a

given level of probability of committing an error in making the decision, whether the
hypothesis is true or false.

TYPES OF STATISTICAL HYPOTHESIS

1. Null Hypothesis
The null hypothesis, denoted by Ho , specifies the statement to be tested. It is
assumed to be true throughout the statistical analysis.
A null hypothesis is generally a statement of no change. Thus, a statement of
equality or one which involves the equality is usually considered in the null hypothesis.
Possible forms of the null hypothesis include (a) equality; (b) less than or equal; and
(c) greater than or equal.

2. Alternative Hypothesis
An alternative hypothesis, denoted by Ha , is the claim that is accepted in case the
null hypothesis is rejected.

NOTE: The null and alternative hypotheses are complementary and must not overlap. The
usual pairs are as follows:
(a) Ho : P arameter = V alue versus Ha : P arameter 6= V alue;

(b) Ho : P arameter = V alue versus Ha : P arameter < V alue;

(c) Ho : P arameter = V alue versusHa : P arameter > V alue;

(d) Ho : P arameter ≤ V alue versus Ha : P arameter > V alue;

(e) Ho : P arameter ≥ V alue versus Ha : P arameter < V alue

 TYPES OF ERRORS

The following table shows the two actions that one can make on the hypothesis and
the corresponding consequences in taking such actions:

Action Hypothesis is TRUE Hypothesis is FALSE

Reject the hypothesis Error Committed No error committed
Fail to reject(accept) the hypothesis No error committed Error committed

The table shows that there are no errors committed when we reject a false hypothesis
and when we fail to reject a true hypothesis. On the other hand, an error is committed
when we reject a true hypothesis and such error is called a Type I error. Also, when we
fail to reject (accept) a false hypothesis, we are committing a Type II error.

As mentioned earlier, for every action that one takes, there are consequences. When we
commit an error, there are consequences, too. Since it is an error in decision making, the
consequences may be tolerable or too severe, severe enough to cause lives. In Statistics, we
measure that chance of committing the error so we will have a basis in making a decision.

EXAMPLES In each of the following, state the null and alternative hypothesis and describe
the Type 1 and Type II errors.

1. A manufacturer of IT gadgets recently announced they had developed a new battery

for a tablet and claimed that it has an average life of at least 24 hours. Would you
buy this battery?
Answer:
Ho: The average life of the newly developed battery for a tablet is at least 24
hours.
Ha: The average life of the newly developed battery for a tablet is less than 24
hours.
Type I error is committed when you did not buy the battery and a possible
consequence is you lost the opportunity to have a battery that could last for at least
24 hours. On the other hand, Type II error is committed when you did buy the
battery and found out later that the battery’s life was less than 24 hours. A possible
consequence of this Type II error is that you wasted your money in buying the battery.
2. A teenager who wanted to lose weight is contemplating on following a diet she read
about in the Facebook. She wants to adopt it but, unfortunately, following the diet
requires buying nutritious, low calorie yet expensive food. Help her decide.
Answer:
Ho:The diet will not result to a change in her weight.
Ha: The diet will induce a reduction in her weight.
Type I error is committed when the teenager did follow the diet and a possible
consequence is that she spent unnecessarily for a diet that did not help her reduce
weight. On the other hand, Type II error is committed when the teenager did not
follow the diet. A possible consequence of this Type II error is that the teenager lost
the opportunity to attain her goal of weight reduction.

3. Alden is exclusively dating Maine. He remembers that on their first date, Maine told
him that her birthday was this month. However, he forgot the exact date. Ashamed
to admit that he did not remember, he decides to use hypothesis testing to make an
educated guess that today is Maine’s birthday. Help Alden do it.
Answer:
Ho:Today is Maine’s birthday
Ha:Maine’s birthday is on another day and not today.
Type I error is committed when Alden’s guess of Maine’s birthday is not on this
day and a possible consequence is that Alden failed to greet or give Maine a birthday
gift today. On the other hand, Type II error is committed when Alden guessed that
today is Maine’s birthday. A possible consequence of this Type II error is that Alden
made the mistake of greeting Maine a happy birthday on that day.
 LESSON 2: STEPS IN HYPOTHESIS TESTING

At the end of the lesson, the learner is able to:

• identify the steps in hypothesis testing

• illustrate the level of significance and corresponding rejection region

• calculate the probabilities of committing an error in a test of hypothesis

TOPIC OUTLINE:

1. Introduce the steps in hypothesis testing procedure.

2. Define level of significane and its role in hypothesis testing.

3. Illustrate the corresponding rejection region based on a given level of significance.

4. Compute the probabilites of committing an error in a test hypothesis.

A test of hypothesis is a series of steps that starts with the formulation of the null and
the alternative hypotheses and ends with stating the conclusion. The following is a sum-
mary of these steps:

1. Formulate the null and the appropriate alternative hypotheses.

2. Identify the test statistic to use. With the given level of significance and the distribu-
tion of the test statistics, state the decision rule and specify the rejection region.

3. Using a simple random sample of observation, compute for the value of the test
statistic.

4. Make a decision whether to reject or fail to reject (accept) Ho.

5. State the conclusion.

 REMARKS ON THE STEPS OF HYPOTHESIS TESTING:

1. When formulating the hypothesis to be tested, remember that we do not know the true
state of nature of the hypothesis, that is, whether the hypothesis is true or false. Take
into consideration the two types of hypothesis to state namely:the null hypothesis and
the alternative hypothesis.

2. The level of significance, usually denoted as α in doing the test of hypothesis, is the
same α that we encounter in the discussion of the (1−α)% confidence interval estimate.
confidence interval estimate.

A level of significance, denoted by α, is the probability of rejecting a true

null hypothesis or committing a Type I error in the test of hypothesis, that is

P(Type I Error) = P(reject Ho / Ho is true) = α

We also have

P(Type II Error) = P(accept Ho / Ho is false) = β

Since it is a probabilty of committing an error, it is usually a small value and it is

between 0 and 1.

3. The test statistic is a standardized expression of the point estimator of the parameter
identified in the hypothesis. The value of the test statistic is based on the random
sample of observations gathered or collected for the purpose of the test of hypothesis. A
critical value identifies the value of the test statistic that would lead to the rejection of
Ho at the chosen level of significance. It ia also important to note that the distribution
of the test statistic is also needed to be specified.

4. Part of the decision rule is the specification of the rejection region or critical
region. The rejection region is a part of the set of all possible values of a test statistic
for which Ho is rejectedand it is bounded by the critical value or values of the test
statistic.
With the computed value of the test statistic, the next step is to use the decision rule
to make a decision whether to reject or fail to reject (accept) the null hypothesis.

5. Conclusions made are in relation to the purpose of the test of hypothesis.

 ILLUSTRATION. Consider a decision rule which is stated as follows:

“At a given α = 0.05, we reject Ho if the computed test statistic (denoted as tc ) is greater
than a tabular value of the t-distribution with n − 1 degrees of freedom. Otherwise, we fail
to reject Ho .”

In this decision rule, the level of significance is set is set at α equal to 0.05 and the
test statistic is denoted by tc which is assumed to follow the Student’s t-distribution with
n − 1 degrees of freedom. The rejection region is the area to the right of the tabular value
obtained from the Student’s t-distribution with n − 1 degrees of freedom. Such rejection
region is illustrated in the following figure.

EXAMPLES

1. In testing the null hypothesis

Ho :“The average daily number of text messages that a Grade 11 student
sends is equal to 100”
against an alternative hypothesis stated as
Ha : “The average daily number of text messages that a Grade 11 student
sends is greater than 100”,
a random sample of 16 students were selected and interviewed. The daily number of
text messages she sends is obtained.
(a) The null hypothesis is said to be rejected if the sample mean is at least 102,
otherwise the null hypothesis will be accepted or we fail to reject Ho. It is assumed
that the number of text messages that a Grade 11 student sends in a day follows a
normal distribution with standard deviation equal to 5 text messages. Compute for
the probability of committing Type I error.
 Solution Computing for the probability of committing Type I error, we have

α = P[Committing a Type I error] = P[RejectHo |Ho is True]

= P [X̄ ≥ 102|µ = 100]
= P X̄−100 ≥ 102−100
 
√5 √5
16 16

= P [Z ≥ 1.60]
= 1 - P [Z < 1.60]
= 1 - 0.9452 = 0.0058.

Thus, we say the probability of rejecting a true null hypothesis is 0.0058 or we say
that on the average, we are assured with 94.52% (1-0.0058 = 0.9452) confidence that
we are making a correct decision in accepting a true null hypothesis.

(b) The alternative hypothesis is stated as “The average daily number of text mes-
sages that a Grade 11 student sends is greater than 100.” If we assume that the true
distribution of the number of text messages that a Grade 11 student sends in a day
follows a normal distribution with a mean of 103 and a standard deviation equal to 5
text messages, then compute the probability of Type II error.
Solution Computing for the probability of committing Type II error, we have
β = P[Committing a Type II error]
= P[Accept Ho |Ho is False]=P[Accept Ho |Ha is True]
= P [X̄ < 102|µ = 103]
= P X̄−103 < 102−103
 
√5 √5
16 16

= P [Z < −0.80]
= 1 - P [Z < 0.80]
= 1 − 0.7881 = 0.2119.
In this case, the probability of accepting a false null hypothesis or accepting Ho given
that the average number of text messages that a Grade 11 student sends in a day is
indeed 103 (greater than 100) is computed as 0.2119.

2. The Graduate Record Exam (GRE) is a standardized test required to be admitted

to many graduate schools in the United States. A high score in the GRE makes
admission more likely. According to the Educational Testing Service, the mean score
for takers of GRE who do not have training courses is 555 with a standard deviation
of 139. Brain Philippines (BP) offers expensive GRE training courses, claiming their
graduates score better than those who have not taken any training courses. To test
the company’s claim, a statistician randomly selected 30 graduates of BP and asked
their GRE scores.
 (a) Formulate the appropriate null and alternative hypotheses.
Answer. Ho: Graduates of BP courses did not score better than 555, while
Ha : Graduates of BP courses did score better than 555.
(b) Identify situations when Type I and Type II errors are committed and state their
possible consequences.
Answer. Type I error is committed when we declare that the company’s claim is true
where in fact BP graduates do not perform better than 555 and a possible consequence
is that the tuition fee paid for the training is wasted. On the other hand, Type II is
committed when we declare that the BP’s claim is false when in fact BP graduates do
score better than 555 and a possible consequence is that opportunity to score better
than 555 is lost.

Reference: Teaching Guide for Senior High School, General Mathematics by the Com-
mission on Higher Education in Collaboration with the Philippine Normal University