Statisticsprobability11 q4 Week5 v4
Statisticsprobability11 q4 Week5 v4
PROBABILITY 11
4th Quarter
Week 5
Learner’s Tasks
Lesson Overview
This lesson still discusses the concepts of hypothesis testing. In the previous
weeks, you have learnt about hypothesis testing on the population mean. It includes
stating the null and alternative hypotheses, determining the level of significance and
the rejection region based on the computed test-statistic, and the types of errors
involved. You have also drawn conclusion about the population mean based on the
test-statistic value and the rejection region. In here, discussions and examples on
solving problems involving population mean are designed to deepen your
understanding about the lesson.
Let us start with problem solving using z-test on the comparison between the
population mean and sample mean.
z-test on the comparison between the population mean 𝝁 and sample mean
̅ is used when the population mean 𝝁 and the population standard deviation 𝝈 are
known, and 𝝁 is compared to a sample mean ̅ It is also important to identify the
parameters or statistics involve in the problem to determine the appropriate formula.
1
Solution: Given: sample size 𝑛 = 200
Parameter: Population mean 𝜇 = 5
Population standard deviation 𝜎 = 2
Statistic: sample mean = 3
4. Establish the critical region/s: Based on 𝐻1, this is a one-tailed test (left-tailed test)
with critical regions 𝑧 < −𝑧𝛼 where −𝑧𝛼 = −𝑧0.05 = −1.645.
The critical region (shaded part) is illustrated below:
This test will be used when the population standard deviation 𝜎 is unknown
but the sample standard deviation 𝑠 is known or can be computed. The t-test will
replace z-test. The degrees of freedom df is the number of pieces of independent
information available for computing variability. The required number of degrees of
freedom varies depending on the size of the distribution. Hence, for a single group of
population 𝑑𝑓 = 𝑛 − 1 and for two groups 𝑑𝑓 = 𝑛1 + 𝑛2 − 1 for t-test where 𝑛, 𝑛1, and 𝑛2
are the sample sizes.
2
Example2. Solve the given problem.
The mean number of hours a student consumes to finish answering the
Statistics Module is 5 hours. A random sample of 15 students was asked and found
that their mean number of hours to finish answering the Statistics module is 6 hours
with a standard deviation of 2 hours. Test the hypothesis that the mean number of
hours to finish answering the module is greater than 5 hours using a level of
significance 𝜶 = 0.05.
3
Let us define the following:
1. The conditions for a binomial experiment are met. That is, there is a fixed number
of independent trials with constant probabilities and each trial has two outcomes
that we usually classify as “success” and “failure”.
2. The condition 𝑛 ≥ 5 and 𝑛𝑞 ≥ 5 are both satisfied so that the binomial distribution
with 𝜇 = 𝑛 and 𝜎 = √𝑛 𝑞.
If all the above conditions are met, the test statistic is the z-test statistic for
proportions. The formula for computing this value is:
𝒛
Test statistic: z =
̂
z=
̂
where: ̂ ,
is the hypothesized population proportion
𝑞
However, we use 𝜎 ̂ √ in computing the value of the test statistic z because
we are going to use the hypothesized value Note that: 𝑞 .
̂
Test statistics z for proportion is: z = .
√ 𝑛
4
In general,
for a one-tailed test:
𝐻0:
𝐻1: > and the rejection region is z > +za
or (𝐻1: < ) and the rejection region is z < -za
For the following examples 3 and 4, we will solve the problem using only the first
three (3) steps in hypothesis testing involving population proportion.
Example 3: Using the 0.05 level of significance run a z-test given the following:
n=74 ; ̂= ;
Solution: Note that the parameter of interest is the population proportion .
1. Formulate the null and alternative hypotheses:
𝐻0: = 0
𝐻0: = 0.10
𝐻1: ≠ 0.10
2. Since = 𝟕𝟒, the Central Limit Theorem applies. And the significance level is
𝛼 = 0.05, two-tailed (since the problem does not suggest direction.)
3. Since it is in the given to run a z-test, thus,
̂
z= ; ̂= 𝟎𝟎 ; 0= 0.10 ; 𝑞0 = 1 − 0 = 1- 0.10 = 0.9
√ 𝑛
Example 4: Mr. Sy asserts that fewer than 5% of the bulbs that he sells are defective.
Suppose 300 bulbs are randomly selected and tested and 10 defective bulbs are
found. Does this provide sufficient evidence for Mr. Sy to conclude that the fraction
of defective bulbs is less than 0.05? Use 𝛼 = 0.01.
5
Now, let us do the following activities. Use a separate sheet of paper for your
answer.
Situation: A rural health unit conducted a survey on the heights of the male aged 18
to 24 years old. It was found out that the mean height of male aged 18 to 24 years
old was 70 inches. Test the hypothesis that the mean height of the male aged 18 to
24 years old is not 70 inches if a random sample of 20 male aged 18 to 24 years old
had a mean height of 65 inches with a standard deviation of 3. Use 1% level of
significance.
Formative Test
Let us see how far you have learned about our lesson. Consider the problem below
and answer the questions that follow. Write the letter corresponding to your answer
in a separate sheet of paper.
1. It is thought that 12% of all students taking BSED course received a grade of A. In a
sample of 155 students, it is found that 21 made an A. State your null and alternative
hypotheses for testing the true proportion in order to conclude that the number of
students taking BSED course who received a grade of A is greater than 12%.
A. 𝐻0: = 0.12 B. 𝐻0: < 0.12 C. 𝐻0: = 0.12 D. 𝐻0: = 0.12
𝐻1: < 0.12 𝐻1: 0.12 𝐻1: 0.12 𝐻1: 0.12
2. A school canteen manager claims that less than 50% of the students of the school
are dissatisfied by the community cafeteria service. Test this claim by using sample
6
data obtained from a survey of 500 students of the school where 54% indicated their
dissatisfaction of the community cafeteria service. Use 𝛼 . What is the test
statistic when CLT is to be used?
A. one-tailed t-test C. two-tailed t-test
B. one-tailed z-test D. two-tailed z-test
For nos. 3-4. Ms. Josie claims that 10% of the beauty products that she sells produce
positive effects. Suppose 50 products are randomly selected and 10 were found to
have good results.
4. Using the 0.05 level of significance, which of the following gives the correct test
statistic value to support Ms. Josie‟s claim?
A. z = 0.58 B. z = 1.78 C. z = 2.38 D. z = 3.18
5. The owner of the perfume factory claims that the average content of their
product is 250 ml. To test the claim, a group of consumer gets a sample of 100
such bottles, calculates the capacity of each bottle and then finds the mean
content to be 248 ml. The standard deviation s is 5 ml. Use 95% confidence level.
Solve for the test statistics value.
A. z = -2 B. z = -3 C. z = -4 D. z = -5
Answer Key
Activity 1
1. 𝐻0: = 0.60 2. 𝐻0: = 0.50 3. 𝐻0: = 0.10
𝐻1: < 0.60 𝐻1: > 0.50 𝐻1: 0.10
Activity 2
1. z = -2.89 2. z = 1.79 3. z = -0.93
Activity 3
Solution: Given: = 65, = 70, 𝑠 = 32, = 20, and 𝛼 = 1% = 0.01
1. Hypotheses: 𝐻0: 𝜇 = 70 and 𝐻1: 𝜇 ≠ 70
2. Significance level: = = 0.005
̅
3. Test statistic: t = ( ) (√𝑛) = ( ) (√ ) = -0.70 and df =19
4. Critical region: Since ≠ is used in 𝐻1, 𝑡 < −𝑡 𝛼 /2, df or 𝑡 > 𝑡 𝛼 /2, df will be used. Now using
the student t-distribution table, the value of − 𝑡 𝛼 /2,df is − 𝑡 𝜶/2,df = −𝑡 0.005, 19 = −2.861
and 𝑡 0.005, 19 = 2.861
5. Decision: We failed to reject 𝐻0, since −0.70 < −2.861 is FALSE and −0.70 > 2.861 is
also FALSE.
7
6. Conclusion: There is no sufficient evidence to conclude that the mean height of
male aged 18 to 24 years old was 70 inches.
References
Belecina, R. R.et.al. (2016). Statistics and Probability. First Edition. Rex Book Store, Inc.
McLeod, S. A. (2019, Nov 25). What is central limit theorem in statistics? Simply
psychology: Retrieved May 27, 2020 from https://www.simplypsychology.org
/central-limit-theorem.html
Stephanie. (2020). Statistics How To. Retrieved May 27, 2020 from
https://www.statisticshowto.com/probability-and-statistics/hypothesis-testing/
8
9
10