STATISTICS and PROBABILITY

Chapter 4
ESTIMATION OF PARAMETERS
LESSON 1

Computing the Point Estimate of a Population Mean

Lesson Objectives:
 Understand the concept of estimation;
 Distinguish between point estimate and interval estimate; and
 Find the point estimates of population means and proportions.

In real life, most of the information we deal with are numerical. When we wake up in the morning, our first
concern is the time, be it a number (e.g., 6 o’clock) or a word (e.g., early morning). Between the two
expressions, which would you prefer?

Task:
Collect information about the wake-up time of every person in your class during
school days. Represent each information by a tally mark. Then, count the tally
marks and write the corresponding decimal numeral under the heading
Frequency.

In the following activity, you will find a number that will describe an entire set of data reporting and
average wake-up time.
Activity 1:
Wake – Up Time
Prepare a tally board with the following headings and entries.
Time Tally Frequency
5:00
5:30
6:00
6:30
7:00
7:30
Total
Average
1. Fill up the table using tally marks. A tally mark is a vertical stroke.
2. After the last student has entered a tally mark, count the tally marks for every row and put this number
under the heading frequency.
3. Insert other times if necessary.
4. For times in between, agree on rounding rule.
5. Write an equation to determine the average.
6. Discuss the number operations involved.
7. What is the average wake up time?

The arithmetic average computed from the table is also known as the mean. The class constitutes a sample.
If we repeat the activity to, say, ten random classes, then we obtain ten arithmetic averages or means. Suppose
we proceed to compute the mean of the means for all ten (10) classes. The final result is a number that is called
point estimate of the mean μ of the population where the samples came from. In symbols, we write.
Xx= μ
This expressions is read as “The mean of the means is equal to the population mean μ (read myu).”

An estimate is a value or a range of values that approximate a parameter. It is

based on sample statistics computed from the sample data.

Estimation is the process of determining parameter values.

Activity 2:
Restaurant Service Quality

Task:
Read the following situation carefully. Then, perform the appropriate
operation to answer the questions.

Norma, a TLE researcher, looked at the average time (in minutes) it takes a
random sample of customers to be served in a restaurant. From 40 customers, the
following information was obtained. What is the average wait time?
______________________

8 8 10 18 10 13 8 10 8 10
12 10 16 16 12 15 12 12 9 15
10 20 20 12 10 10 16 10 18 12
15 12 15 14 15 16 15 12 8 8

1. What is the mean of the sample? ____________minutes.

2. This value is called _________________ of the population mean μ.
3. Based on the observation of the mean, would you patronize the restaurant?

The mean of the sample is an estimate of the population parameter μ, the “true” average time it takes to
be served in the restaurant. The number is used to describe a particular characteristics, wait time, of the
population. If, indeed, this is the population. An infinite population is a hypothetical collection of elements such
as all the results of a coin tossing experiment to determine the probability of getting heads or tails.

Population parameters are usually unknown fixed values. But, there is a way to determine them. In the
example, there are two ways to report the results:
1. Report a number that describes the average wait time. This number is called point estimate.
In this case, the mean is the best estimator. Why not the median or the mode? Among the three
measures of central value, the mean is the most reliable measure of center. In a mean, any change
in a value affects the result. Means of samples vary less than the median or the mode.
2. Report a range of values that contains the number that truly describes the wait time. This s called the
interval estimates.

Point estimate is a specific numerical value of a population parameter. The

sample mean x is the best point estimate of the population mean.

Interval estimate is a range of values that may contain the parameter of a

Estimates provide limited information. It does not tell much about the information, and it does not tell much
about the possible size of the error. We supplement the information by stating that the sample is a random
sample. In our example, we have a random sample whose mean is x = 12.5 minutes and whose standard
deviation s is 3.41 minutes. The sample standard deviation an serve as the estimate of σ .

How confident are we about our estimations? In order to feel confident about our estimators as
approximations of the true parameter values, we take as as many random samples as possible from the
population of interest, compute the sample statistics, and then carefully compare the results. Then, we
formulate conclusions and based on these, we make particular decisions. A good method of estimating a
population parameter is described as one where the estimates from many samples are equal to the true
population parameter. In such case, we can say that the sample statistic is an unbiased estimate. If across
many sample statistic departs from the true population value, then the estimate is said to be biased.

A good estimator has the following properties:

1. When the mean of a sample statistic from large number of different random samples equals the true
population parameter, then the sample statistic is an unbiased estimate of the population parameter.
 2. Across the many repeated samples, the estimates are not very far from the true parameter value.
The following figures illustrate bias where the vertical line represents the population mean and the dots
represent sample means from the x sampling distribution.

σ
. . = 0.025 ...
2
. . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . .
μ μ μ
Negative bias Unbiased Positive bias
(Under estimate) (On target estimate) (Over estimate)

Activity 3.
Coco Juice
Study the following situation and do the task. Then, answer the following questions or supply the
missing information.

Mr. Santiago’s company sells bottled coconut juice. He claims that a bottle contains 500 ml of such
juice. A consumer group wanted to know if his claim is true. They took six random samples of 10 such bottles
and obtained the capacity, in ml, of each bottle. The result is shown as follows:
Sample 1 500 498 497 503 499 497 497 497 497 495
Sample 2 500 500 495 494 498 500 500 500 500 497
Sample 3 497 497 502 496 497 497 497 497 497 495
Sample 4 501 495 500 497 497 500 500 495 497 497
Sample 5 502 497 497 499 496 497 497 499 500 500
Sample 6 496 497 496 495 497 497 500 500 496 497
Assuming that the measurements were carefully obtained and that the only kind of error present is the sampling
error, what is the point estimate of the population mean?

Solutions:
Method 1: Using the EXCEL Program

A sample of the worksheet for sample 1 is shown as follows.

A B C D E F
1 500
2 498
3 497
4 503
5 499
6 497
7 497
8 497
9 497
10 495

Steps:
1. Encode the data in the cells preceding from cell A1 to cell A10.
2. Select the data.
3. Click the insert function fx. The insert function dialogue box appear.
4. In the insert function dialogue box, click the arrow to select category. A drop window will appear. In this
window, click statistical.
 5. In the insert function dialogue box, there is another window labeled Select a function. Click
AVERAGE. At the bottom of the box, click OK.
6. Another dialogue box will appear. This is the function arguments box. There are two smaller windows in
the box labelled Number 1 and Number 2. In the window number 1, write A1:A10. The computer reads
the numbers encoded in Column A Row 2, and so on to Column A Row 10.
7. Copy the formula result in the dialog box. For the encoded data of sample 1, the result is 498. Then click
OK.
8. Repeat the procedure for the other samples.
9. Compute the mean of the means also called overall means. This is the point estimate of the μ.

Alternatively, in MS Excel Program, the values may simply be encode one after the other from Row 1 to
Row 60. Next, select the entries. Then repeat steps 3 to 7. With the proliferation of the statistical
software, you can choose the most convenient model to compute statistical values for statistical analysis.

Method 2: Manual Estimation of the Population Mean

When dealing with a large number of values, the mean of small samples may be obtained. These means
constitute a sampling distribution of means. To find the overall mean, simply find the sum of the men values.
Then, divide this sum by the total number of sample means.

For example, let us consider the six sample rows of the 60 bottles as excellent samples. Next, we
compute the row mean.

For example mean of row 1,

500+498+ 497+503+ 499+ 497+497 +497+ 497+ 495

X r1 =
10

4980
=
10

= 498

Activity 4:

Coco Juice Revisited

1. Proceeding in the same manner, compute the mean for each remaining row.
2. Compute the sum of the means.
3. Enter the values of the following table.
4. Divide the sum by the number of means. The result is called overall mean.
5. The overall mean is the point estimate of the population mean. What is this number?
Discuss your procedure with your classmates.

Sample Row Sum of Scores Mean

1 4980
2 4984
3 4972
4 4979
5 4984
6 4971
Overall Mean
Activity 5:

More on Coco Juice

Task:
1. Look at the 60 bottles of coconut juice as consisting of 10 columns and 6 rows.
2. Compute the means of the column samples.
3. What is the overall mean? This value is also an estimate of the population mean μ.

Sample Sample Sample Sample Sample Sample Sample Sample Sample Sample
1 2 3 4 5 6 7 8 9 10
500 498 497 503 499 497 497 497 497 495
500 500 495 494 498 500 500 500 500 497
497 497 502 496 497 497 497 497 497 495
501 495 500 497 497 500 500 495 497 497
502 497 497 499 496 497 497 499 500 500
496 497 496 495 497 497 500 500 496 497

Activity 6:

Computing More Means

1. Compute the remaining column means.
2. Compute the mean of the means.
3. Enter your values in the following table.
4. Based on the 10 random samples, what is the point estimate of the population mean?

Column Sum of Scores Mean

1 2996
2 2984
3 2987
4 2984
5 2984
6 2988
7 2991
8 2988
9 2987
10 2981
Overall Mean
Based on the ten random samples, the point estimate of the population parameter is ______ml.

Notice that regardless of how we randomly elect our samples, thus forming a sampling distribution of
means, the mean of the means is equal to the population mean. We now feel confident in saying that if more
random samples are obtained from the population, and the means of these random samples are computed, then
the mean of the means of the samples is the point estimate of the population mean.

Finally we interpret the result by saying that the claim is different from the computed value. That is, the
computed mean _____ based on the sample is (equal to, slightly less than, slightly more than) the claimed value
of 500 ml.

Now based on the sample result, what action (e.g., patronage of the product) would you make? Write
this on a piece of paper. Discuss your answer with your seatmate.

Other than the mean, we also compute the standard deviation s to describe the dispersion of the scores in
a distribution. The standard deviation is the square root of the variance. So, we compute the variance first. Then,
we extract the square root of the result to get the value of the standard deviation.
Activity 7:
Bottled Drinks
Task: Complete the computation of the standard deviation of the 60 bottled drinks.

2
Recall: Variance (s ) : s 2
= ∑ ( X−X )
2

n−1

√ ∑ (X −X )
2
Standard Deviation (s): s =
n−1

Where:
∑ = ‘sum of’
X = column mean
X = Overall mean
N = number of cases

In the following procedure, the values are expressed up to two decimal places. Note that that we are on
the column means.
Steps Solutions
1. Find the sum of the
s 2
= ∑ ( X−X ) 2

values in the numerator n−1

and divide it by 9 (from ( 499.93−497.83)
2 2
+(497.33−497.83) +…+(496.33−497.83)
2
n-1 = 10 – 1 = 9). s2 =
10−1

s2 = __________________

This result is called variance.

2. Extract the square root

of the variance. s = ____________________

This value is called ________________

3. Describe/interpret the
result (Applying the obtained value of s.)
On the average, the sample mean is ________________ away from the mean
of the ten values.

Activity 8:
Truck Ban
Study the next example and do the indicated tasks to provide the missing
information in the solution.
Mr. Domingo conducted a survey among ten random samples of people who are
in favor of truck ban in a section of metropolitan area. He determined the percentages
of those who are in favor of the ban. Assuming that the only error present is the
sampling error, he wanted to determine the point estimate of the population mean
percentage and the standard deviation based on 500 observations. The following
numbers represent the percentages of the ten surveys.

47.0 56.4 50.1 60.2 48.0 55.3 60.0 59.5 63.0

57.5

Solutions:
Steps Solutions
1. Find the mean of the
percentages.
Formula: X =
∑X where X is percentage.
n
 47.0+56.4 +50.1+…+57.5
X=
10
X = ____________

2. Give the estimate of the

population mean The estimate of the population mean percentage is _______________
percentage.

Compute the standard deviation.

Steps Solutions
1. Find the sum of the
values in the numerator
s2 = ∑ ( X−X )2
and n-1. n−1

2. Divide the sum by 9 s2 = ----------------------------------------------------------------

s2 = ______________

the value is called ___________________

3. Extract the square root

of s2 S = ______________________ this value is called ___________________

4. Describe/interpret the
result _________________________________________

Prepared by:

DOMINIC T. PAYYAC, MAEd, LPT

Teacher III