Q4-M1 Stat
Q4-M1 Stat
Probability
Quarter 4 – Module 1:
Confidence Interval and Sample
Size
Lesson
Confidence Interval
1
Are you planning to run as a student leader? How are you going to
determine whether you can win or lose in your desired position? How
important is the margin of error in a survey for winning the election? Are
you familiar with the length of the confidence interval and how it is applied
in real-life situations? What is a margin of error?
What's New
Definition of Terms
The values of 𝒛𝜶⁄ (confidence coefficients) are listed below with the
𝟐
usual confidence level used in estimating population mean.
In the general formula for determining the interval estimate for the
𝜎
parameter 𝜇, the value 𝑥̅ − 𝑧𝛼⁄2 ( 𝑛) is called the lower confidence
√
𝜎
boundary or limit (LL) and the other value 𝑥̅ + 𝑧𝛼⁄2 ( 𝑛) is called the upper
√
confidence boundary or limit (UL).
𝜎
The term 𝑧𝛼⁄2 ( 𝑛) is called margin of error, denoted by E, which is
√
defined as the maximum likely difference between the observed sample
mean and the true value of the population mean 𝜇. Thus, another way of
writing the formula for finding the confidence interval for the population
parameter 𝜇 is:
̅−𝑬<𝝁<𝒙
𝒙 ̅+𝑬
𝝈
where 𝑬 = 𝒛𝜶⁄ ( 𝒏)
𝟐 √
Example 1:
Let’s see if we can get the same value of E if we use the formula for the
upper limit.
̅+E
UL= 𝑿
The confidence level is 90%. So, we are 90% confident that the mean
score lies between 77.18 and 78.82.
Note: Sometimes, you just need to convert the formula to find what is
missing.
Example 2.
Solution: From the table 1 (see above), for 95% level, 𝒛𝜶⁄ = 𝟏. 𝟗𝟔, thus, the
𝟐
margin of error is:
𝜎 10
𝑧𝛼⁄2 ( ) = 1.96 ( ) = 𝟑. 𝟗𝟐
√𝑛 √25
𝑬 = 𝟑. 𝟗𝟐
Example 3.
Solution:
𝜎 𝜎
Given: 𝑥̅ = 32 𝑥̅ − 𝑧𝛼⁄2 ( ) < 𝜇 < 𝑥̅ + 𝑧𝛼⁄2 ( )
𝜎 = 4.5 √𝑛 √𝑛
𝑛 = 100
4.5 4.5
From the table, for 97% level, 32 − 2.17 ( ) < 𝜇 < 32 + 2.17 ( )
𝑧𝛼⁄ = 2.17 √100 √100
2
32 − 0.98 < 𝜇 < 32 + 0.98
Substitute the values of 𝑥̅ , 𝑧𝛼⁄2 , 𝜎,
and 𝑛, then calculate. 𝟑𝟏. 𝟎𝟐 < 𝝁 < 𝟑𝟐. 𝟗𝟖
𝐸 = 0.98
Interpretation: Thus, we can say with 97% confidence that the interval between
31.02 miles-per-gallon and 32.98 miles-per-gallon contain the population mean
based on 100 units of the new model’s miles-per-gallon rate.
Example 5:
Solution:
Given: 𝑥̅ = 3 𝜎 𝜎
𝜎 = 0.5 𝑥̅ − 𝑧𝛼⁄2 ( ) < 𝜇 < 𝑥̅ + 𝑧𝛼⁄2 ( )
𝑛 = 50 √𝑛 √𝑛
From the table, for 95% level, 0.5 0.5
𝑧𝛼⁄2 = 1.96 3 − 1.96 ( ) < 𝜇 < 3 + 1.96 ( )
√50 √50
Substitute the values of 𝑥̅ , 𝑧𝛼⁄ , 𝜎,
2 3 − 0.14 < 𝜇 < 3 + 0.14
and 𝑛, then calculate.
𝟐. 𝟖𝟔 < 𝝁 < 𝟑. 𝟏𝟒
𝐸 = 0.14
Interpretation: Thus, we can say with 95% confidence that the interval
between 2.86 hours and 3.14 hours contain the population mean based on
50 five-year old children’s TV viewing time.
T-distribution
The concept of the degrees of freedom is used in the t-distribution.
The degrees of freedom, denoted by 𝒅𝒇, are the number of values that are
free to vary after a sample statistic has been computed, and they tell us the
specific curve to use when a distribution consists of a family of curves.
Thus, to compute for 𝒅𝒇 = 𝒏 − 𝟏 , where n is the sample size.
To determine the t-values, we use the t-table as shown below. Note:
𝜶 = 𝟏 − 𝒄𝒐𝒏𝒇𝒊𝒅𝒆𝒏𝒄𝒆 𝒍𝒆𝒗𝒆𝒍 (area on the right tail), for example, with 95%
confidence level, 𝜶 = 𝟏 − 𝟎. 𝟗𝟓, thus, the area on the right tail is 0.05.
Illustrative Example 1:
What is the values of 𝑡𝛼⁄2
(confidence coefficients) when
𝑛 = 22 at 99% confidence level?
𝛼⁄ = 0.005 and 𝑑𝑓 = 21
2
𝑠 5
𝐸 = 𝑡𝛼⁄2 ( ) = 1.753 ( ) = 𝟐. 𝟏𝟗
√𝑛 √16
𝑬 = 𝟐. 𝟏𝟗
Length of the confidence interval:
𝑳𝑪𝑰 = 𝟐𝑬
= 2(2.19)
= 𝟒. 𝟑𝟖
Example 7:
Solution: From the given, 𝑑𝑓 = 12 − 1 = 11, since it is 95% confidence level, 𝛼 = 0.05,
thus, 𝒕𝜶⁄ = 𝒕𝟎.𝟎𝟓⁄ = 𝒕𝟎.𝟎𝟐𝟓. From the t-table, for 𝒅𝒇 = 𝟏𝟏, 𝒕𝟎.𝟎𝟐𝟓 = 𝟐. 𝟐𝟎𝟏.
𝟐 𝟐
𝑠 𝑠
𝑥̅ − 𝑡𝛼⁄2 ( ) < 𝜇 < 𝑥̅ + 𝑡𝛼⁄2 ( )
√𝑛 √𝑛
9 9
27 − 2.201 ( ) < 𝜇 < 27 + 2.201 ( )
√12 √12
27 − 5.72 < 𝜇 < 27 + 5.72
𝟐𝟏. 𝟐𝟖 < 𝝁 < 𝟑𝟐. 𝟕𝟐
Example 8:
A printer manufacturer wants to estimate the pages-per-cartridge
print rate of their new printer model. A random sample of 10 units of the
new printer model showed a mean of 14000 pages-per-cartridge with a
standard deviation of 20. Construct a 98% confidence interval to estimate
the pages-per-cartridge rate of their new printer model.
Solution: From the problem, 𝑠 = 20, 𝑥̅ = 1400, and 𝑛 = 10. Therefore, 𝑑𝑓 =
10 − 1 = 9, and since it is 98% confidence level, 𝛼 = 0.02, so 𝑡𝛼⁄2 = 𝑡0.01 . From
the t-table, for 𝑑𝑓 = 9, 𝒕𝟎.𝟎𝟏 = 𝟐. 𝟖𝟐𝟏.
𝑠 𝑠
𝑥̅ − 𝑡𝛼⁄2 ( ) < 𝜇 < 𝑥̅ + 𝑡𝛼⁄2 ( )
√𝑛 √𝑛
20 20
1400 − 2.821 ( ) < 𝜇 < 1400 + 2.821 ( )
√10 √10
1400 − 17.84 < 𝜇 < 1400 + 17.84
𝟏𝟑𝟖𝟐. 𝟏𝟔 < 𝝁 < 𝟏𝟒𝟏𝟕. 𝟖𝟒
Interpretation: Thus, we can say with 98% confidence that the interval between
1382.16 pages and 1417.84 contains the true population mean pages-per-cartridge
based on the sample of 10 units of new printer model.
Example 9.
An admission officer of an educational institution wants to know the
mean age of all entering mathematics majors. He computed a mean age of
18 years and a standard deviation of 1.2 years on a random sample of 25
entering mathematics majors purportedly coming from a normally
distributed population. With 99% confidence, find the point estimate and
the interval estimate of the population mean.
Interpretation: Thus, we can say with 99% confidence that the interval
between 17.33 and 18.67 contains the true mean age the population of
entering mathematics majors based on the sample of 25 entering
mathematics majors.
Notes to the Students
Computing for the interval estimate of population proportion is not covered
here. Meanwhile, population proportion will be covered in hypothesis testing. A
population proportion is a fraction or a percentage of the population that has a
certain characteristic. The formula for computing the a large-sample confidence
interval for population proportion p is
𝑝̂ 𝑞̂ 𝑝̂ 𝑞̂
𝑝̂ − 𝑧𝛼⁄2 √ < 𝑝 < 𝑝̂ + 𝑧𝛼⁄2 √
𝑛 𝑛
where:
𝑝̂ = sample proportion
𝑞̂ = 1 − 𝑝̂
𝑛 = sample size
𝑧𝛼⁄2 = z value that leaves an area of 𝛼⁄2.
𝑝̂𝑞̂
From the formula, 𝑧𝛼⁄2 √ is the margin of error E.
𝑛
Lesson
Sample Size Determination
2
In the previous lessons, you learned how to identify and compute for
the confidence interval. This time, using the length interval, you will learn
how to compute for an appropriate sample size.
Your knowledge in determining the sample size is essential
particularly in conducting a research study. A researcher does not just
declare a number to represent the number of respondents he/she wants to
include in the investigation. An appropriate sample size and sampling
techniques must be implemented to come up with the best estimate of the
parameter. Therefore, it is one of the requirements so that the findings of the
study can be generalized to every member of the population.
What's In
𝐸 = margin of error
Since the value of 𝜎 is usually unknown, it can be estimated by the
standard deviation 𝜎 from a prior sample.
Factors Affecting Sample size Determination
There are three factors that influence the sample size determination;
(1) the level of confidence, (2) the population standard deviation, and (3) the
margin of error. Among these factors, only two are under the researcher’s
control.
These are:
• Margin of Error. Since no sample will be perfect, you need to decide how
much error to allow. The confidence interval determines this. The less error
you’re willing to accept, the bigger the sample size needs to be.
• Confidence Level. This determines how certain you want to be that the
parameter falls within your confidence interval. Most common confidence
intervals are 90%, 95% or 99%. The more confident you want to be, the
bigger the sample size needs to be.
EXAMPLE 1.
SOLUTION:
Let us identify the given in the problem:
Evaluate: 𝑛 ≈ 33.70
Round off to the nearest whole number: 𝑛 = 𝟑𝟒
Therefore, to be 99% confident that the estimate differs from the true
mean by 4, the minimum sample size needed is at least 34.
Note: When determining sample size, we always round off the resulting value
to the next whole number.
EXAMPLE 2.
SOLUTION:
To answer the question, let us identify the given:
Evaluate: 𝑛 ≈ 11.54
Round off to the nearest whole number: 𝑛 = 𝟏𝟐
Therefore, to be 95% confident that the estimate differs from the true
mean by 1.5, we need a sample size of at least 12.
EXAMPLE 3.
Evaluate: 𝑛 ≈ 46.98
Round off to the nearest whole number: 𝑛 = 𝟒𝟕
Therefore, to be 90% confident that the estimate differs from the true mean
by 6, we need a sample size of at least 47.
EXAMPLE 4.
𝑍𝛼⁄ 𝜎
Use the formula: 𝑛=( 2
)2
𝐸
(1.75)(25) 2
Substitute the given: 𝑛 =[ ]
6
Evaluate: 𝑛 ≈ 53.17
Round off to the nearest whole number: 𝑛 = 𝟓𝟒
Therefore, to be 92% confident that the estimate differs from the true mean
by 6, we need a sample size of at least 54.
The average percentage of people who are satisfied with the product is
_______________.
The lower limit of confidence interval is _________________.
The upper limit of confidence interval is _________________.
The confidence interval is _______________________.
2. You are planning to buy a cellphone with a good quality in order to attend
the online class. The average price of 50 cellphones is Php13,500 with a
margin of error of Php 273.65 and a confidence level of 99%.
3. A survey on 200 Senior High School learners on how much time they read
per week has a confidence interval of (13.5, 17.9) in hours at 95%
confidence level.
Activity 1.3. Compute the interval estimate given the following data:
1. Confidence Level =90%, 𝑥̅ = 42, 𝑠 = 10, 𝑛 = 20
2. Confidence Level =95%, 𝑥̅ = 142, 𝜎 = 9, 𝑛 = 25
3. Confidence Level =99%, 𝑥̅ = 28, 𝜎 = 12, 𝑛 = 60
4. To estimate the average salary of policemen, 64 of them were taken as
samples. If the mean salary of these samples is Php25,000 with
standard deviation of Php800, construct a 90% confidence interval
estimate for the average salary of all policemen.
5. The average weight of 15 adult Dagupan Bangus is 750grams with
standard deviation of 80grams. Construct a 98% confidence interval
estimate of the average weight of all adult Dagupan Bangus.
Activity 1.4. Determine the minimum sample size for each of the following: