STATISTICS AND PROBABILITY

LESSON: TESTS OF HYPOTHESIS

Learning Objectives:
1. Define the parameters used in statistical analysis.
2. Identify the parameter to be tested in a real-life problem.
3. Identify the notation to be used in formulating hypotheses.
4. Illustrate one-tailed and two-tailed tests.
5. Differentiate null and alternative hypotheses.
6. Formulate null and alternative hypotheses.
7. Define the statistical concepts related to test concerning means.
8. Identify the appropriate form of test statistics when: (a) the population variance is
assumed to be known; (b) the population variance is assumed to be unknown; and
(c) the Central Limit Theorem is to be used.
9. Apply the concepts of test statistic on real-life problems.

WEEK 2
IDENTIFYING PARAMETERS FOR TESTING IN GIVEN REAL-LIFE PROBLEMS
FORMULATING APPROPRIATE NULL AND ALTERNATIVE HYPOTHESES ON
A POPULATION MEAN
IDENTIFYING APPROPRIATE TEST STATISTICS INVOLVING POPULATION
MEAN

IDENTIFYING PARAMETERS FOR TESTING IN GIVEN REAL-LIFE PROBLEMS

Parameters in statistics are important component of any statistical analysis. In simple words,
a parameter is any numerical quantity that characterizes a given population or some of its
aspects. This means the parameter tells us something about the whole population.

However, the numerical measure that is calculated from the sample is called statistic.
Statistic is a known number and a variable that depends on the portion of the population.

A parameter denotes the true value that would be obtained if a census rather than a sample
was undertaken.

Examples of parameters are the measures of central tendency. These tell us how the data
behave on an average basis. For example, mean, median, and mode are measures of
central tendency that give us an idea about where the data concentrate. Meanwhile,
standard deviation tells us how the data are spread from the central tendency, i.e. whether
the distribution is wide or narrow. Such parameters are often very useful in analysis.

In the normal distribution, there are two parameters that can characterize a distribution - the
mean and standard deviation. By varying these two parameters, you can get different
kinds of normal distribution.

Different symbols are used to denote parameters.

IDENTIFYING PARAMETER TO BE TESTED
ILLUSTRATIVE Examples:
1. The average height of adult Filipinos 20 years and older is 163 cm for males.
Parameter: the average height of adult Filipinos 20 years and older
In hypothesis testing, the parameter will be translated into symbols such as 𝛍 = 𝟏𝟔𝟑
where 𝛍 is the symbol for mean/average and 163 is the value that pertains to the
average height.

2. A Grade 11 researcher reported that the average allowance of Senior High School
students is ₱100. A sample of 40 students has mean allowance of ₱120. At 𝛼 = 0.01
test, it was claimed that the students had allowance of ₱ 100. The standard deviation of
the population is ₱50.
Parameters: the average allowance of Senior High School students is ₱100 or 𝝁 =
₱𝟏𝟎𝟎
In this claim, there are different parameters used but the parameter to be tested in this
hypothesis would be the average allowance of Senior High School students since it
relates to the population, not in sample. Statistical hypothesis is a conjecture about the
population parameter that’s why you will look for the population mean, population
standard deviation, or population proportion but not sample mean.

3. According to a survey, 63% of the parents are willing to spend extra money for their
children’s health and education matters.
Parameter: the percentage/proportion of parents willing to spend extra money in
their children’s health and education matter or 𝒑 = 𝟎. 𝟔𝟑

To identify the parameters to be tested:

1. Just look for mean/average, standard deviation, variance, and proportion of
POPULATION.
2. Determine the value that pertains to the given parameter, then translate them in
symbols for hypothesis testing.

TOPIC 2: FORMULATING APPROPRIATE NULL AND ALTERNATIVE HYPOTHESES ON

A POPULATION MEAN
A statistical hypothesis is a statement about a parameter and deals with evaluating the
value of a parameter.

In statistical hypothesis testing, there are always two hypotheses: the null and the
alternative hypotheses. Below is a comparison between the two.
Null Hypothesis H o Alternative Hypothesis H a
 It states that there is no difference  It states that the population parameter
between population parameters (such has some statistical significance
as mean, standard deviation, and so on) (smaller, greater, or different than) with
and the hypothesized value. the hypothesized value.
 There is no observed effect.  There is an observed effect.
 The null hypothesis is often an initial  The alternative hypothesis is what you
claim that is based on previous might believe to be true or hope to
analyses or specialized knowledge. prove true.
To state the null and alternative hypotheses correctly:
1. Identify the parameter in a given problem.
2. Identify the claim to be tested that may show up in null or alternative hypothesis.
3. Translate the claim into mathematical symbols/notations.
4. Formulate first the null hypothesis ( H o ) then alternative hypothesis ( H a ) based on the
three different ways in writing hypothesis as illustrated below:
H o : μ=k Ho : μ ≤ k Ho : μ ≥ k
Ha : μ ≠ k H a : μ>k H a : μ<k

Hypothesis-Testing Common Phrases

¿  is equal to ≠  is not equal to
 is the same as  is not the same
 is exactly the same as  is different from
 has not changed from  has changed from
¿  is increased ¿  is decreased
 is greater than  is less than
 is higher than  is lower than
 is above  is below
 is bigger than  is smaller than
 is longer than  is decreased or reduced from
 is more than  is not more than
≥  is at least ≤  is at most
 is not less than  is not more than
 is greater than or equal to  is less than or equal to

ILLUSTRATIVE Examples:
1. The average weight of all Grade 11 students in Senior High School is 169cm. Is this
claim true?
Solution:
First, identify the parameter which is the mean height of all Grade 11 students. Since it is
a population mean, use the notation 𝝁. The claim in this example is that the average
weight is 169 cm which translates to 𝝁 = 𝟏𝟔𝟗 and is considered as null hypothesis. To
formulate the alternative hypothesis, write the complement/opposite of the null
hypothesis which is the average weight is not equal to 169 cm.
H o : The average weight of all Grade 11 students is 169 cm. / H o : 𝝁 = 𝟏𝟔𝟗 (claim)
H a : The average weight of all Grade 11 students is not 169 cm./ H a: 𝝁 ≠ 𝟏𝟔𝟗

2. The average price per square meter of a residential lot in an exclusive subdivision is
above ₱15,000. A buyer wants to test the agent’s claim.
Solution:
In this hypothesis, the parameter is the average. Therefore, you will use the symbol µ.
The claim is above ₱15,000 can be written as µ > ₱15,000 and greater than falls at
alternative hypothesis, H a : 𝝁> ₱𝟏𝟓, 𝟎𝟎𝟎. Since you have already formulated the
alternative, the null hypothesis will be H o : 𝝁 ≤ ₱𝟏𝟓, 𝟎𝟎𝟎 as a complement of >. You can
also write your null hypothesis as H o : 𝝁 = ₱𝟏𝟓, 𝟎𝟎𝟎.
H o : 𝝁 ≤ ₱𝟏𝟓, 𝟎𝟎𝟎 or H o : 𝝁 = ₱𝟏𝟓, 𝟎𝟎𝟎
 H a : 𝝁> ₱𝟏𝟓, 𝟎𝟎𝟎 (claim)

3. Holistic Fitness Center claims that its members reduced an average of 13 pounds after
joining the center. An independent agency that wanted to check this claim took a sample
of 40 members and found that they reduced an average of 12 pounds with the standard
deviation of 4 pounds. Determine the null and alternative hypotheses.
Solution:
In this example, the parameter to be tested is the average and the claim is reduced by
13 pounds. The claim that pertains to the parameter has the notation of (<). Therefore,
the claim is found at the alternative hypothesis and can be written as H a :𝝁<𝟏𝟑. The null
hypothesis would be H o :𝝁 ≥ 𝟏𝟑 or H o : 𝝁 = 𝟏𝟑.
H o : 𝝁 ≥ 𝟏𝟑 or H o : 𝝁 = 𝟏𝟑.
H a : 𝝁<𝟏𝟑 (claim)

ONE-TAILED AND TWO-TAILED TEST

The alternative hypothesis can take another form depending on the value of the parameter.
The parameter may increase, decrease, or changed from the null value. An alternative
hypothesis predicts not only the difference of sample mean from the population mean but
also how it would be different in a specific direction - lower or higher. This test is called a
directional or one-tailed test because the rejection region is entirely within one tail of the
distribution.

On the other hand, some hypotheses predict only that one value will be different from
another, without additionally predicting which will be higher. The test of such a hypothesis is
nondirectional or two-tailed because an extreme test statistic in either tail of the
distribution (positive or negative) will lead to the rejection of the null hypothesis of no
difference.

The table below shows the null and alternative hypotheses stated together with the
directional test.

ILLUSTRATIVE Examples:
1. Teacher A wants to know if mathematical games affect the performance of the students
in learning Mathematics. A class of 45 students was used in the study. The mean score
was 90 and the standard deviation was 3. A previous study revealed that 𝝁 = 𝟖𝟓 and the
standard deviation 𝝈 = 𝟓.

Solution:
 The parameter is the population mean 𝜇 = 85. You can write the hypotheses into
symbols: H o ∶𝜇 = 85 and H a ∶𝜇 ≠ 85. The phrase ‘affects performance’ has no clue of
the direction of the study, so it implies either an increase or decrease in performance.
This tells you that the test is a two-tailed test.
H o ∶𝜇 = 85 and H a ∶𝜇 ≠ 85 (two-tailed test)

2. A piggery owner believes that using organic feeds on his pigs will yield greater income.
His average income from the previous year was ₱120, 000. State the hypothesis and
identify the directional test.
Solution:
In this example, the null hypothesis is H o ∶𝝁 = 𝟏𝟐𝟎,𝟎𝟎𝟎. You may notice that the
hypothesis used the phrase ‘greater income’ that is associated with greater than.
Therefore, H a ∶𝝁>𝟏𝟐𝟎,𝟎𝟎𝟎. This hypothesis uses inequality symbol (>) so it is one-tailed
test and it uses greater than which specifically called for the right-tailed test.
H o ∶𝝁 = 𝟏𝟐𝟎,𝟎𝟎𝟎 and H a ∶𝝁>𝟏𝟐𝟎,𝟎𝟎𝟎(right-tailed test)

3. The average waiting time of all costumers in a restaurant before being served is less
than 20 minutes. Determine the hypotheses and the directional test.
Solution:
You may notice that the hypothesis used the phrase ‘less than’ which denotes that the
alternative hypothesis is H a ∶𝝁<𝟐𝟎. This hypothesis uses the inequality symbol (<) so it is
a one-tailed test and it used less than what is specifically called for the left-tailed test.
In this example, the null hypothesis H o ∶𝝁≥𝟐𝟎.
H o ∶𝝁≥𝟐𝟎and H a ∶𝝁<𝟐𝟎(left-tailed test)

TOPIC 3: IDENTIFYING APPROPRIATE TEST STATISTICS INVOLVING POPULATION

MEAN
Before moving on to different test statistics, define and review the following statistical terms.
A population includes all of the elements from a set of data.
A sample consists of one or more observations drawn from the population.
Sample mean( x ) is the mean of sample values collected.
Population mean( μ ) is the mean of all the values in the population. If the sample is
randomly selected and sample size is large, then the sample mean would be a good
estimate of the population mean.
Population standard deviation (𝝈) is a parameter which is a measure of variability with
fixed value calculated from every individual in the population.
Sample standard deviation (𝒔) is a statistic which means that this measure of variability
is calculated from only some of the individuals in a population.
Population variance( σ 2 ), in the same sense, indicates how the population data points
are spread out. It is the average of the distances from each data point in the population
to the mean, squared.

A test statistic is a random variable that is calculated from sample data and used in a
hypothesis test. You can use test statistics to determine whether to reject or accept the null
hypothesis. The test statistic compares your data with what is expected under the null
hypothesis.

To identify the test statistic, you must consider whether the population standard
deviation/variance is known or unknown. If the population standard deviation σ is known,
then the mean has a normal distribution. Use z-test. If the population standard deviation σ is
unknown, then the mean has a t- distribution. Use t-test. Instead of the population standard
deviation, use the sample standard deviation.
z-test
In a z-test, the sample is assumed to be normally distributed. A z-score is calculated with
population parameters such as “population mean” and “population standard deviation”. It is
used to validate a hypothesis that the sample drawn belongs to the same population. When
the variance is known and either the distribution is normal or sample size is large, use a z-
test statistic.

t-test
Like a z-test, a t-test also assumes a normal distribution of the sample. A t-test is used when
the population variance or standard deviation are not known. When the variance is unknown
and a sample size is less than 30, use a t-test statistic assuming that the population is
normal or approximately normal.

Central Limit Theorem

In Central Limit Theorem, if the population is normally distributed or the sample size is large
and the true population mean μ=μ o, then z has a standard normal distribution.

When population standard deviation σ is not known, we may still use z-score by replacing
the population standard deviation σ by its estimate, sample standard deviation s. Since
the sample is large the resulting test statistic still has a distribution that is approximately
standard normal.

When sample sizes are small, the Central Limit Theorem does not apply. You must then
impose stricter assumptions on the population to give statistical validity to the test procedure.
One common assumption is that the population from which the sample is taken has a normal
probability distribution to begin with. Under such circumstances, if the population standard
x−μo
deviation is known, then the test statistic σ still has the standard normal distribution.
√n
The table below shows what test statistic is appropriate when:
IDENTIFYING APPROPRIATE TEST STATISTIC

ILLUSTRATIVE Examples:
1. A manufacturer claimed that the average life of batteries used in their electronic games
is 150 hours. It is known that the standard deviation of this type of battery is 20 hours. A
consumer wished to test the manufacturer’s claim and accordingly tested 100 electronic
games using the battery. It was found out that the mean is equal to 144 hours.
Here, the sample size (n) is 100 (extremely large) and population standard
deviation (20 hours) is known, then the appropriate test statistic to be used is z-
test.
2. An English teacher wanted to test whether the mean reading speed of students is 550
words per minute. A sample of 12 students revealed a sample mean of 540 words per
minute with a standard deviation of 5 words per minute. At 0.05 significance level, is the
reading speed different from 550 words per minute?
The sample size (n) is 12 which is less than 30 and sample standard deviation (5
words per minute) was given. Therefore, the appropriate test is t-test.
3. A study was conducted to look at the average time students exercise. A researcher
claimed that in average, students exercise less than 15 hours per month. In a random
sample size n=115, it was found that the mean time students exercise is 𝑥̅= 11.3 hours
per month with s = 6.43 hours per month.
Since n=115, the sample size is large and variance is unknown. Hence, z-test is the
appropriate tool. (Central Limit Theorem)

NOTE: The examples above used standard deviations instead of variances. Variance is
the square of the standard deviation and conversely, the standard deviation is the square
root of the variance. Hence, if the standard deviation is known in the problem, then
basically, variance is also known.

UNDESTAND THE CONCEPT OF HYPOTHESIS TESTING AND THE REST OF THE

TOPICS . WE WILL HAVE ASSESSMENT @ FRIDAY MORNING [at least 20 points]