277

Chapter 8 Testing Other Hypothesis

COURSE OUTLINE

8.1 Introduction
8.2 Testing difference between two large sample means
8.3 Testing difference between two small independent sample means
8.4 Testing difference between two dependent sample means
8.5 Testing the difference between two proportions

6.1 Introduction

Presented in the previous chapter some types of statistical tool that are appropriate
concerning the difference between sample mean and population mean, sample variance and
population variance, and sample proportion and population proportion which is called one-sample
case are presented.
This chapter is concerned with the characteristics of two random samples which are
called the two-sample case and the statistical tools which are appropriate to the test concerning
two samples. The two-sample case requires that the samples be selected independently. This
means that one of the samples cannot affect the probability that any particular case for the other
sample. Also in the two-sample case the null statement for two sided hypothesis is still a
statement of “no difference”. However, instead of saying that the population from which a sample
is drawn has a certain characteristics, it will say that the population are the same.

6.2 Testing difference between two large sample means

To illustrate the procedure for testing difference between two large sample means,
consider a problem concerning the number of slaughtered cattle in Ilocos Region. The sample is
divided by semester. The sample statistics are computed for first semester and second semester.
The number of slaughtered cattle yields interval-ratio-level data, assuming that the distributions
of the variables are approximately normal, a test for the significance of difference in sample can
be conducted.
Since the sample sizes are large (both sample sizes are greater than 30) and the distribution
of the differences in sample means will be normal, then the test statistics 𝑧𝑥̅1−𝑥̅2can be computed
by the usual formula:
 278

(1)
Sample outcome (𝑥1̅ − 𝑥̅2) minus the expected outcome(𝜇1 − 𝜇2), all divided by the standard
deviation of the sampling distribution of the differences in sample means.
The standard deviation of the sampling distribution of the difference in sample means is
defined as

(2)

Substituting to (1)

(3)

The population standard deviation is rarely given, to estimate them you must use the sample
standard deviation, corrected for bias. Equation (2) became,

(4)

Equation 4 is called a pooled estimate since it combines information from both the samples.
Substituting to (1), we have the statistical tool for the test of difference between two sample
means

(5)

Population means usually are not given in the problem. In testing hypothesis, it is usually
hypothesized that the difference between the sample means are equal to zero. If the null
hypothesis is true then the difference between the two populations means would be zero or very
close to zero.

Hypothesis Testing procedure for testing the difference between two large sample
means
 279

1. State the hypotheses.

State null (𝑯𝟎) and alternative (𝑯𝟏) hypotheses.
2. Determine the critical value.
Identify 𝜶, use 𝜶 𝐨𝐫 𝜶/𝟐 for one-tailed or two-tailed test respectively, then
determine the critical value using the z table.
3. Compute test statistics. Use the formula below,

4. Interpret the result.

a. Make decision.
For two tailed test, accept the null hypothesis if −𝒛𝜶⁄𝟐 ≤ 𝒛 ≤ 𝒛𝜶⁄𝟐.
For one tailed test, accept the null hypothesis if;
𝒛 ≥ −𝒛𝜶 left tailed test
𝒛 ≤ 𝒛𝜶 right tailed test
b. Make conclusion.

Example 8.2.1
A vendor claims that the production of Habitchuelas (Snap beans) during rainy season is
higher than during summer season. The table below shows the production of Habitchuelas in
metric tons on the month of March to May (summer season) and June to October (rainy season) in
Pangasinan from 2002 to 2014. Can a vendor conclude, at 𝛼 = 0.05, that the production of
habitchuelas during rainy season is higher than during summer season?

Year Summer Season Rainy Season

March April May June July August Sept. Oct.

2002 17.49 25.11 28.4 18.76 26.62 27.45 22.28 17.38
2003 16.57 15.31 23.87 39.94 28.3 36.64 24.52 22.87
2004 19.28 17.67 22.27 24.82 31.38 20.88 29.95 31.37
2005 20.2 23.52 15.9 19.14 24.78 23.22 27.01 34.04
2006 28.02 22.52 22.93 27.34 33.25 32.4 28.08 26.96
2007 19.49 19.55 24.98 24.95 23.87 25.26 33.56 34.64
2008 22.99 21.24 38.39 42.19 39.17 37.74 33.56 34.64
2009 28.13 22.75 35.21 44.67 33.16 33.27 34.47 70.12
2010 26.92 18.93 25.44 27.35 27.29 16.2 15.62 48.1
2011 18.99 23.82 48.32 49.34 35.83 26.33 35 104.81
2012 33.69 34.04 36.77 25.94 23.55 59.38 46.38 21.45
2013 28.8 28.93 27.65 29.36 27.55 28.34 44.35 41.47
2014 25.65 28.83 33.31 29.75 32.09 30.28 27.15 25.11
Source: Philippine Statistics Authority
 280

Solution
Step 1. State the hypotheses
𝐻0: The production of Habitchuelas during rainy season is lower than or equal to during
summer season. 𝜇1 ≤ 𝜇2.
𝐻1: The production of Habitchuelas during rainy season is higher than during summer
season. 𝜇1 > 𝜇2. Claim Step 2. Find the critical value
The significant level is 𝛼 = 0.05, from the z table the critical value is 1.64.
Step 3. Compute the test statistics.
a. Compute the test mean and variance for each season
,
𝑠12 = 65(18.76 2 2 2
+26.52 +⋯+2565.11∙ 64)−(18.76+26.52+⋯+25.11)

= 177.154
17 49 25 11 33 31

∙
= 49.05
b. Compute for 𝑧.

( 35 25 )

Step 4. Interpret the result

The computed test statistics (3.43) is greater than the critical value (1.64). This
implies that the null hypothesis must be rejected. Therefore, there is enough evidence to
accept the claim that the production of Habitchuelas during rainy season is greater than
during summer season.

Example 8.2.2
A supervisor wishes to determine whether the teacher-student ratio in the Private Higher
Education in Region1 and Region 2 is the same. The results are shown here. Is there enough
evidence to support the claim of the supervisor that the teacher-student ratio is the same? Use 𝛼 =
0.10

Region 1 Region 2
𝑥̅1 = 0.1 𝑥̅2 = 0.15
𝑠1 = 0.09 𝑠2 = 0.26
𝑛1 = 81 𝑛2 = 49
Source: Private Higher Education Institutions
 281

Step 1. State the hypotheses

𝐻0: The teacher-student ratio in the Private Higher Education in Region1 and Region 2 is
the same. 𝜇1 = 𝜇2. Claim.

𝐻1: The teacher-student ratio in the Private Higher Education in Region1 and Region 2 is
not the same. 𝜇1 ≠ 𝜇2.

Step 2. Find the critical value

The level of significant is 𝛼 = 0.10. Using , using the z-table the critical
values are 1.64 and -1.64.

Step 4. Compute the test statistics.

Step 5. Decision
The computed test statistics falls between -1.64 and 1.64. This implies that the
test value is on the critical region. Thus, the null hypothesis must be accepted. Therefore,
there is enough evidence to accept the claim that the teacher-student ratio in both Regions
is the same.

Technology Application: Using Microsoft Excel Software Step by Step

Example 8.2.3
Use excel to carry out test of differences between the production of Habitchuelas during summer
and rainy season.

Solution
Step 1. In Cell A1 enter: summer. In cell B1 enter: rainy. Enter the volume of production of
habitchuelas during rainy and summer season data in the respective columns.
 282

Step 2. Find the variance of each sample data.

Step 3. Select Data Data Analysis z Test: Two Sample Means. Click OK.

Step 4. For Variable 1 Range enter: $A$2:$A$40. For Input Variable 2 Range enter:
$B$2:$B$66. For Hypothesized Mean Difference enter: 0. Enter the computed variance of
variable 1 and 2 from step 5. For Alpha enter: 0.05. Lastly, for the output range enter $D$2.
Click OK

z-Test: Two Sample for Means

Variable Variable
1 2
32.34877 25.43282 Mean
Known Variance 177.154 49.0506
Observations 65 39
 283

Hypothesized Mean
Difference 0
Z 3.465279 z - value
P(Z<=z) one-tail 0.000265 z Critical one-
tail 1.644854
P(Z<=z) two-tail 0.00053 p - value
z Critical two-tail 1.959964

The computed z-value give the same result with the test of differences between two large
samples Using Excel.

Exercise 8.2
Exercise 1 through 5. Test the claim about the mean of the difference of two populations at the given
level of significance. Use a z-test for independent samples. Assume the populations are normally
distributed.
1. 𝐶𝑙𝑎𝑖𝑚: 𝜇1 ≠ 𝜇2, 𝛼 = 0.05. Sample statistics: 𝑥1̅ = 78.32, 𝑠1 = 4.56, 𝑛1 = 45 and 𝑥̅2 =
64.23, 𝑠2 = 12.56, 𝑛2 = 34.

2. 𝐶𝑙𝑎𝑖𝑚: 𝜇1 > 𝜇2, 𝛼 = 0.10. Sample statistics: 𝑥1̅ = 17.12, 𝑠1 = 3.13, 𝑛1 = 76 and 𝑥̅2 =
15.23, 𝑠2 = 2.12, 𝑛2 = 112.
3. 𝐶𝑙𝑎𝑖𝑚: 𝜇1 = 𝜇2, 𝛼 = 0.05. Sample statistics: 𝑥1̅ = .83, 𝑠1 = .23, 𝑛1 = 36 and 𝑥̅2 = 1.32, 𝑠2 =
0.31, 𝑛2 = 40.
4. 𝐶𝑙𝑎𝑖𝑚: 𝜇1 ≥ 𝜇2, 𝛼 = 0.01. Sample statistics: 𝑥1̅ = 156.17, 𝑠1 = 25.5, 𝑛1 = 56 and 𝑥̅2 =
122.45, 𝑠2 = 30.5, 𝑛2 = 48..
5. 𝐶𝑙𝑎𝑖𝑚: 𝜇1 < 𝜇2, 𝛼 = 0.01. Sample statistics: 𝑥1̅ = 36.78, 𝑠1 = 4.72, 𝑛1 = 47 and 𝑥̅2 =
40.23, 𝑠2 = 8.92, 𝑛2 = 59.
Exercise 6 through 11. Each table shows the average and standard deviation of the monthly consumer
price index in the provinces of Ilocos Region from January 2012 to August 2014. At 𝛼 = 0.05, verify that
the monthly consumer price index in each paired provinces are the same. Source: Philippine Statistics
Authority
6. Pangasinan and La-Union
Mean Standard Deviation
Pangasinan 124.86 3.28
La-Union 133.60 2.94

7. Pangasinan and Ilocos Sur

Mean Standard Deviation
Pangasinan 124.86 3.28
Ilocos Sur 130.85 4.64
 284

8. Pangasinan and Ilocos Norte

Mean Standard Deviation
Pangasinan 124.86 3.28
Ilocos Norte 123.32 3.10

9. La-Union and Ilocos Sur

Mean Standard Deviation
La-Union 133.60 2.94
Ilocos Sur 130.85 4.64

10. La-Union and Ilocos Norte

Mean Standard Deviation
La-Union 133.60 2.94
Ilocos Norte 123.32 3.10

11. Ilocos Sur and Ilocos Sur

Mean Standard Deviation
Ilocos Sur 130.85 4.64
Ilocos Norte 123.32 3.10

Exercise 12 to 15. A research director believes that the retail price per kilogram of cabbage and native
pechay is the same. The data in each table is the average standard deviation of the monthly retail price
of cabbage and native pechay from January 2010 to May 2015. At 𝛼 = 0.05, is there enough evidence
to support the research directors claim in each provinces. 12. Pangasinan
Mean Standard deviation
Cabbage 38.46 10.23
Pechay native 31.46 8.35

13. La-Union
Mean Standard deviation
Cabbage 34.49 8.46
Pechay native 53.62 12.15

14. Ilocos Sur

Mean Standard deviation
Cabbage 32.11 11.98
Pechay native 34.58 8.30

15. Ilocos Norte

Mean Standard deviation
Cabbage 32.16 14.84
Pechay native 42.29 12.76
 285

Exercise 16 to 19. The tables below reveal the monthly retail price of the poultry products in the
provinces of Ilocos Region from January 2010 to May 2015. Show that monthly retail price of each type
of poultry products is the same in the
province.
Mean Standard
deviatio
n
Pangasina 131.96 16.87
n
Ilocos 140.09 17.41
Norte

Mean Standard
18. deviatio
n
La-Union 130.92 17.61
Ilocos Sur 152.39 18.87

Chicken egg (other breed). Use 𝛼 = 0.05

19. Chicken egg (other breed). Use 𝛼 = 0.10

Mean Standard
deviation
Iloc
os
Nort
e 4.87

8.3 Testing difference between two small independent sample means

When the sample sizes are small (less than or equal to 30) and the population standard
deviation is unknown, the z distribution cannot be used. When both samples are independent and
taken from two normally or approximately normally distributed population, the appropriate
statistical tool that can be used is t distribution
 286

Testing difference between two sample variances

If in your study you are concerned with the uniformity of the two populations, this means
that you are interested in testing hypotheses concerning the uniformity of one population
comparing to the uniformity of the other population. Then there are statistical tools that are used
to compare the two sample variances. The statistical tool presented in this section is the F-test.
The test concerning one sample variance you will use chi-square; for the test concerning two
sample variances you will use F test. F-test is simply the ratio of the variance of two sample data.
The variance of the sample data with larger variance should be placed in the numerator.

Definition 8.3.1 F-test

If 𝒔𝟐𝟏 and 𝒔𝟐𝟐 be the variance of the two sample data, then the test statistic to compare two
variances is,

where 𝒔𝟐𝟏 > 𝒔𝟐𝟐, with degrees of freedom 𝒏𝟏 − 𝟏 and 𝒏𝟐 − 𝟏. 𝒏𝟏 − 𝟏is the degrees of freedom
for numerator and 𝒏𝟐 − 𝟏 is the degrees of freedom for the denominator.

Assumptions and requirements for using F test

1. Each sample is distributed normally.
2. The two sample data are selected independently.
3. Level of measurement is interval-ratio

Figure 8.3.1 shows the family of F distribution based on the degrees of freedom.

Figure 8.3.1

Hypothesis Testing procedure for testing the difference between two variances

1. State the hypotheses.

State the null (𝑯𝟎) and alternative (𝑯𝟏) hypotheses.
2. Determine the critical value.
a. Identify the alpha value and find the degrees of freedom for the first
sample (𝒅. 𝒇. 𝒏𝟏) and degrees of freedom for the second sample
(𝒅. 𝒇. 𝒏𝟐).
𝒅. 𝒇. 𝒏𝟏 = 𝒏𝟏 − 𝟏
 287

𝒅. 𝒇. 𝒏𝟐 = 𝒏𝟐 − 𝟏
b. Locate the critical value using the F-table. Use 𝜶 and 𝜶⁄𝟐 for one –
tailed and two-tailed respectively. The critical value is found where
the row and column intersect in the column.
3. Find test statistics. Compute,

4. Interpret the result.

a. Decision
Accept the null hypothesis if,
𝑭 < 𝑪. 𝑽.
b. Conclusion

Example 8.3.1
A researcher claims that the variance of the monthly inflation rates in Pangasinan and
Ilocos Norte are the same. The data below shown the monthly inflation rate from August 2013 to
August 2014. At 𝛼 = 0.05 level of significance, is there enough evidence to accept the claim of
the researcher?
Pangasinan 4 4 4.1 5.2 4.8 5.3 2.1 2.3 1.7 1.5 3.2 3.5 3.8
Ilocos
4 4.5 4.6 3.9 4.6 4.3 2.6 3.1 2.9 2.8 3.2 3.6 4
Norte
Solution
Step 1. State the hypotheses.
𝐻0: The monthly inflation rates in Pangasinan and Ilocos Norte are the same.
𝐻1: The monthly inflation rates in Pangasinan and Ilocos Norte are not the same.
Step 2. Determine the critical value.
a. Identify the alpha value and find the degrees of freedom for the first sample (𝑑. 𝑓. 𝑛1)
and degrees of freedom for the second sample (𝑑. 𝑓. 𝑛2).
𝛼 = 0.05,
𝑑. 𝑓. 𝑛1 = 𝑛1 − 1 = 13 − 1 = 12
𝑑. 𝑓. 𝑛2 = 𝑛2 − 1 = 13 − 1 = 12
b. Locate the critical value using the F-table. Since the problem is a two tailed, thus the
𝛼⁄2 = 0.05⁄2 = 0.025 table must be used. Hence, the critical value is 3.28. Step
3. Find test statistics.
 288

Step 4. Interpret the result.

The computed 𝐹 test (3.20) is less than the critical value (3.28). This implies that the null
hypothesis must be accepted. Therefore, there is enough evidence to support the claim that
the monthly inflation rate in Pangasinan and Ilocos Norte are the same.

Example 8.3.2
A consumer price index (CPI) measures changes in the price level of a market basket of
consumer goods and services purchased by households. A director believes that the variance of
the monthly consumer price index in Pangasinan is greater than in La-Union. You randomly
select 10 months in Pangasinan and 8 months in La-Union from January 2012 to August 2014.
At 𝛼 = 0.01 level of significance, is there enough evidence to accept the claim of the director?
Source: Philippine Statistics Authority

Pangasinan 128.7 128.4 129.2 123 123.3 126.3 126.9 121.4 120.9 122.8
La-Union 137.4 132.2 131.4 134.5 135.7 129 129.1 131.6

Step 1. State the hypotheses.

 289

𝐻0: The monthly consumer price index in Pangasinan is less than in La-Union.
𝐻1: The monthly consumer price index in Pangasinan is greater than in La-Union.

Step 2. Determine the critical value.

a. Identify the alpha value and find the degrees of freedom for the first sample (𝑑. 𝑓. 𝑛1)
and degrees of freedom for the second sample (𝑑. 𝑓. 𝑛2).
𝛼 = 0.01,
𝑑. 𝑓. 𝑛1 = 𝑛1 − 1 = 10 − 1 = 9
𝑑. 𝑓. 𝑛2 = 𝑛2 − 1 = 8 − 1 = 7
b. Locate the critical value using the F-table. Since the problem is a one-tailed, thus the
𝛼 = 0.01 table must be used. Hence, the critical value is 5.61.
Step 3. Find test statistics.
a. Square the values in each sample.
b. Take the sum of each column.

𝒙 𝒙𝟐 𝒚 𝒚𝟐
128.7 16563.69 137.4 18878.76
128.4 16486.56 132.2 17476.84
129.2 16692.64 131.4 17265.96
123 15129 134.5 18090.25
123.3 15202.89 135.7 18414.49
126.3 15951.69 129 16641
126.9 16103.61 129.1 16666.81
121.4 14737.96 131.6 17318.56
120.9 14616.81
122.8 15079.84
∑ 𝒙 = 1250.9 ∑ 𝒙𝟐 =156564.7 ∑ 𝒚 = 1060.9 ∑ 𝒚𝟐 = 140752.7

c. Compute 𝑆1 and 𝑆2.

√
= 3.16

√
= 3.02
 290

Step 4. Interpret the result.

The computed 𝐹 test (1.09) is less than the critical value (5.61). This implies that the null
hypothesis must be accepted. Therefore, there is no enough evidence to support the claim that
the variance of monthly consumer price index in Pangasinan is greater than in La-Union.

Testing difference between two small independent sample means

To do computation for t, you need to know first if the variances of each sample are equal
or not equal. The degrees of freedom is based on the result if the variances are equal or not equal.
You need the value of degrees of freedom in order to use the t table (table 3 appendix A). To
determine whether two sample variances are equal, you can use the F test. There are two formulas
when we compute for t: t distribution for unequal variances and t distribution for equal variances.

Definition 8.3.1
Let 𝒙̅𝟏, 𝒙̅𝟐 and 𝒔𝟏, 𝒔𝟐 be the mean and standard deviation of the random samples with 𝒏𝟏 and 𝒏𝟐
sizes, if the variances of the samples are assumed to be unequal, then

Where the degree of freedom is equal to the smaller of 𝒏𝟏 − 𝟏 or 𝒏𝟐 − 𝟏.

If the variances of the samples are assumed to be equal,

then

Where the degrees of freedom are equal to 𝒏𝟏 + 𝒏𝟐 − 𝟐

Hypothesis Testing procedure for testing the difference between two small sample means

(𝒙
̅ 𝟏 −𝒙
̅ 𝟐 )−(𝝁𝟏 −𝝁𝟐 )
𝒕=
(𝒏𝟏 −𝟏)𝒔𝟐 𝟐
𝟏 +(𝒏𝟐−𝟏)𝒔𝟐 𝟏 𝟏
√ √ 𝒏 +𝒏
𝒏𝟏 +𝒏𝟐 −𝟐 𝟏
 291

2. State the hypotheses.

State null (𝑯𝟎) and alternative (𝑯𝟏) hypotheses.
3. Determine if the variances are equal or not equal. Compute F.
4. Determine the critical value.
a. Identify 𝜶.
b. Determine the degrees of freedom.
For unequal variance, the degrees of freedom is equal to the smaller of 𝒏𝟏 − 𝟏
or 𝒏𝟐 − 𝟏.
For equal variance, the degrees of freedom is equal to 𝒏𝟏 + 𝒏𝟐 − 𝟐.
c. Determine the critical value using the t-table.
5. Find test statistics. For unequal variance use,

For equal variance use,

6. Interpret the result.

c. Make decision.
For two tailed test, accept the null if −𝒕𝜶⁄𝟐 ≤ 𝒕 ≤ 𝒕𝜶⁄𝟐.
For one tailed test, accept the null if;
𝒕 ≥ −𝒕𝜶 left tailed test
𝒕 ≤ 𝒕𝜶 right tailed test
d. Interpret the decision.

Example 8.3.3
From example 10.3.1, a researcher also claims that the monthly inflation rates in
Pangasinan and Ilocos Norte are the same. At 𝛼 = 0.05 level of significance, is there enough
evidence to accept the claim of the researcher?

Solution
1. State the hypotheses.
𝐻0: The monthly inflation rates in Pangasinan and Ilocos Norte are the same. 𝜇1 = 𝜇2. 𝐻1:
The monthly inflation rates in Pangasinan and Ilocos Norte are the same. 𝜇1 ≠ 𝜇2.
2. Determine if the variances are equal or not equal.
From problem 1, the variances of the monthly inflation rates in Pangasinan and La-Union
are the same.
3. Determine the critical value.
The level of significance is 𝛼 = 0.05. The problem is two-tailed, thus you will use
0.025. Since the variances are equal, you will use the degrees of freedom 𝑛1 + 𝑛2 − 2 =
13 + 13 − 2 = 24. Hence, from t table, the critical values are -2.064 and 2.064.
4. Find test statistics.
 292

5. Interpret the result.

Since computed 𝑡 falls between -2.064 and 2.064, then the null hypothesis must
be accepted. Therefore, there is enough evidence to accept the claim of the researcher that
the inflation rate in Pangasinan is the same as in La-Union.

Example 8.3.4
Example 10.3.2, shows that the monthly consumer price index in Pangasinan is lower
than in La-Union. Use 𝛼 = 0.01 level of significance.
 293

Solution
Step1. State the hypotheses.
𝐻0: The monthly consumer price index in Pangasinan is lower than in La-Union. 𝜇1 <
𝜇2. Claim
𝐻1: The monthly consumer price index in Pangasinan is equal to or greater than in La-
Union. 𝜇1 ≠ 𝜇2.
Step2. Determine if the variances are equal or not equal.
From problem 2, the variances of the monthly consumer price index in Pangasinan and
La-Union is the same.
Step 3. Determine the critical value.
The problem is a one-tailed test with significant level 𝛼 = 0.01. Since the variances are
equal, you will use the degrees of freedom 𝑛1 + 𝑛2 − 2 = 10 + 8 − 2 = 16. Hence, from t table, the
critical values is -2.583. Step 4. Find test statistics.
 294

= −8.9997 𝑜𝑟 − 9

Step 5. Interpret the result.

Since the computed 𝑡 = −9 is less than -2.583, this implies that the null
hypothesis must be accepted. Therefore, there is enough evidence to accept the
claim that the monthly consumer price index in Pangasinan is lower than in
LaUnion.

Example 8.3.5
An agriculturist claims that the monthly retail price of cabbage in Ilocos Norte is less
than in La-Union from January 2010 to May 2015. You randomly select 9 months in Ilocos Norte
and 12 months in La-Union. At 𝛼 = 0.05, can you support the claim of the agriculturist?

Mean Standard
Deviation
Ilocos Norte 29.26 15.04
La-Union 34.80 6.58

Solution
Step1. State the hypotheses.
𝐻0: The monthly retail price of cabbage in Ilocos Norte is less than in La-Union. 𝜇1 < 𝜇2. Claim

𝐻1: The monthly retail price of cabbage in Ilocos Norte is greater than or equal to in La-Union.
𝜇1 ≥ 𝜇2.

Step2. Determine if the variances are equal or not equal.

a. Determine the critical value.
At 𝛼 = 0.05, with 𝑑. 𝑓. 𝑛1 = 𝑛1 − 1 = 12 − 1 = 11 and 𝑑. 𝑓. 𝑛2 = 𝑛2 − 1 =
9 − 1 = 8 the critical value is 2.95.

b. Compute F.

The computed (5.22) is greater than 2.95, this implies that the variance
of monthly retail price of cabbage in Ilocos Sur is higher than in La-Union.

Step 3. Determine the critical value.

The problem is a one-tailed test with significant level 𝛼 = 0.05. Since the variances are
unequal, you will use the degrees of freedom 𝑛1 − 1 = 8. Hence, from t table, the critical
value is -1.860.

Step 4. Find test statistics.

Compute t. Use unequal variance formula.
 295

Step 5. Interpret the result.

Since the computed 𝑡 = −1.14 is greater than −1.860, this implies that the null
hypothesis must be accepted. Therefore, there is no enough evidence to support the claim of the
agriculturist that the monthly retail price of cabbage in Ilocos Norte is less than in La-Union.

Technology Application: Using Microsoft Excel Software Step by Step

Example 8.3.6
Use excel to carry out test of differences between the variance of the monthly inflation rate in
Pangasinan and Ilocos Norte.

Solution

Step 1. In Cell A1 enter: Pangasinan. In cell B1 enter: Ilocos Norte. Enter the monthly inflation
rate data in Pangasinan and Ilocos Norte in the respective columns.
Step 2. Select Data→ Data Analysis →F-Test Two-Sample for Variances. Click OK.

Step 3. For Variable 1 Range enter: $A$2:$A$14. For Input Variable 2 Range enter: $B$2:$B$14.
For Alpha enter: 0.025. Lastly, for the output range enter $D$4. Click OK
 296

F-Test Two-Sample for Variances

Variable Variable
1 2
Mean 3.5 3.7
Variance 1.625 0.51
Observations 13 13
Df 12 12
F 3.186275
P(F<=f) one-tail 0.027698
F Critical one-tail 3.277277

The computed F-value give the same result with the test of variances between two
samples Using Excel.

Example 8.3.7
Use excel to carry out test of differences (small sample) between the means of the monthly
inflation rate in Pangasinan and Ilocos Norte.

Solution
Step 1. In Cell A1 enter: Pangasinan. In cell B1 enter: Ilocos Norte. Enter the monthly inflation
rate data in Pangasinan and Ilocos Norte in the respective columns.
Step 2. Select Data Data Analysis t-Test Two-Sample Assuming Equal Variances. Click OK.

Step 3. For Variable 1 Range enter: $A$2:$A$14. For Input Variable 2 Range enter: $B$2:$B$14.
For Alpha enter: 0.05. Lastly, for the output range enter $D$4. Click OK
 297

t-Test: Two-Sample Assuming Equal Variances

Variable 1 Variable 2
Mean 3.5 3.7
Variance 1.625 0.51
Observations 13 13
Pooled Variance 1.0675
Hypothesized Mean
Difference 0
Df 24
t Stat -0.49352
P(T<=t) one-tail 0.313066
t Critical one-tail 2.063899
P(T<=t) two-tail 0.626133
t Critical two-tail 2.390949

The computed t-value give the same result with the test of differences between two
sample means (small samples) using excel.

Exercise 8.3
Exercise 1 to 5. Test the claim using t-test of independent about the mean of the difference of two
populations at the given level of significance. Assume that the data are normally distributed.
1. 𝐶𝑙𝑎𝑖𝑚: 𝜇1 ≠ 𝜇2, 𝛼 = 0.01. Sample statistics: 𝑥1̅ = 12.12, 𝑠1 = 2.31, 𝑛1 = 23 and 𝑥̅2 =
64.23, 𝑠2 = 14.67, 𝑛2 = 26. Assume 𝜎12 ≠ 𝜎22.
2. 𝐶𝑙𝑎𝑖𝑚: 𝜇1 > 𝜇2, 𝛼 = 0.10. Sample statistics: 𝑥1̅ = 37.8, 𝑠1 = 7.13, 𝑛1 = 16 and 𝑥̅2 =
29.23, 𝑠2 = 5.28, 𝑛2 = 12. Assume 𝜎12 = 𝜎22.
3. 𝐶𝑙𝑎𝑖𝑚: 𝜇1 = 𝜇2, 𝛼 = 0.05. Sample statistics: 𝑥1̅ = .92, 𝑠1 = .31, 𝑛1 = 14 and 𝑥̅2 = 1.18, 𝑠2 =
0.22, 𝑛2 = 12. Assume 𝜎12 = 𝜎22.
4. 𝐶𝑙𝑎𝑖𝑚: 𝜇1 ≥ 𝜇2, 𝛼 = 0.05. Sample statistics: 𝑥1̅ = 66.37, 𝑠1 = 16.25, 𝑛1 = 20 and 𝑥̅2 =
62.54, 𝑠2 = 12.05, 𝑛2 = 18. Assume 𝜎12 ≠ 𝜎22.
5. 𝐶𝑙𝑎𝑖𝑚: 𝜇1 < 𝜇2, 𝛼 = 0.01. Sample statistics: 𝑥1̅ = 34.78, 𝑠1 = 3.72, 𝑛1 = 22
and 𝑥̅2 =
40.23, 𝑠2 = 7.23, 𝑛2 = 26. Assume 𝜎12 ≠ 𝜎22.
Exercise 6 to 15, Perform the hypothesis testing procedure for z test large sample for two populations.
6. The standard deviations of the registered sales and service Filipino emigrant workers are 612.50
and 473.73 respectively (Source: http://www.cfo.gov.ph/). At 𝛼 = 0.10, verify that the variances
of the two occupation are not significantly different from each other.
7. The variance of the percentage of board passers in electrical and mechanical engineering at Don
Mariano Marcos Memorial State university is 167.41 and 66.22 from 2010 to 2012 respectively
(Source: Philippine Regulatory Commission). Suppose that the data are approximately normal.
 298

Using 𝛼 = 0.05, is the variance of electrical engineering board passer is different from the
variance of mechanical engineering?
8. The table below shows the blood glucose level of mice 15 days after alloxan administration. The
mice are treated with Syzygium cumini (Duhat) leaf extract and metformin (standard drug).
Syzygium cumini 163.5 159.5 158
Metformin 152.5 147.5 147
Source: Anne Lorraine J. Pineda (2014). Hyperglycaemic effects of three indigenous Philippine
plants to alloxan diabetic mice.
At 𝛼 = 0.01, is the blood glucose level of the mice treated with Syzygium cumini is comparable
with the mice treated with metformin?
9. The data below shows the amount of rainfall in millimetre during August and September in
Dagupan, Pangasinan. At 𝛼 = 0.05, determine if the amount of rainfall is significantly different
from each other.
Aug Sep
1998 188 1,063
1999 1,071 272
2000 649 322
2001 329 328
2002 217 329
2003 1,030 391
2004 1,184 112
2005 333 325
2006 547 274
2007 710 197
2008 615 371
2009 394 532
2010 490 164
2011 827 419
2012 994 387
Source: Philippine Atmospheric, Geophysical and Astronomical Services Administration (PAGASA)
10. The table below shows the number of live births by gender in Ilocos Region. At 𝛼 = 0.01,
determine if the number male live births is significantly higher than the number of female live
births.

Year Male Female

2011 46,469 42,957
2010 48514 46613
2009 48948 46160
2008 51243 48183
2007 50069 46399
2006 47926 44919
2005 48592 45480
2004 48166 45465
2003 46463 44295
2002 48007 45210
 299

2001 46969 46120

2000 44129 42387
1999 45613 43563
1998 44726 42358
1997 43694 42494
1996 45017 43327
Source: Field Health Service Information Service (FHSIS), DOH

Each table from exercises 11 to 16 show the annual per capita consumption of agricultural commodities
by socio-economic class of households in Ilocos Region. In each exercise, determine if there is a
significant difference between the data according to socio-economic class of households. Source: Bureau
of Agricultural Statistics
11. Rice. Use 𝛼 = 0.10.

Upper Class 120.43

Lower Class 119.55

12. Camote. Use 𝛼 = 0.05.

1.92
1.87

13. Pechay Use 𝛼 = 0.05

Upper Class 0.5

7

0.47
Middle Class 0.6 0.57
2

14. Tomato. Use

Mean Standard N
 300

Deviation
Middle Class 0.66 0.12 3
lower Class 0.52 0.23 3

15. Squash. Use 𝛼 = 0.01

Mean Standard N
Deviation
Middle Class 3.604 1.065 3
lower Class 2.09 0.27 3

16. Mongo. Use 𝛼 = 0.10

Mean Standard N
Deviation
Upper Class 1.27 0.34 3
Middle Class 1.62 0.06 3

17. A researcher believes that the monthly purchasing power of peso in La-Union and Ilocos Sur is
the same. The data below show the randomly selected months in La-Union and in Ilocos Sur. Is
there enough evidence to support the claim of the researcher? Use 𝛼 = 0.1.
La-Union 0.74 0.72 0.76 0.75 0.78 0.77 0.77 0.7 0.7 0.7
6 6 6
Ilocos 0.73 0.78 0.78 0.78 0.76 0.75 0.75 0.7
Sur 4
Source: National Statistics Office

18. Francis claims that the yearly number of Filipino emigrants in the municipality of Paoay Ilocos
Norte is greater than in the municipality of Piddig Ilocos Norte. The data below show the number
of Filipino emigrants from 1988 to 2012 in the said municipality. At 𝛼 = 0.05, is there enough
evidence to support the claim of Francis?
Paoay Piddig

1 5
8 7 2 3 4 4 6 7 7 8 9
3 3 0 2 5 7 7 9
7 2 1 0 0 4 9
6 3 2 0 0 0 0 5 2 3 3 5 6
9 6 4 4 6 1 4
8 5 3 7 1 4
7 4 2 8

19. Dominic believes that the yearly number of Filipino emigrants in the municipality of Luna and
Naguilian La-Union is the same. The data below show the number of Filipino emigrants from
1988 to 2012 in the said municipality. At 𝛼 = 0.05, verify the claim of Dominic.
 301

Luna Naguilian

8 3 7
9 8 5 2 0 4 0 2 3 5
9 8 7 6 2 2 1 0 5 0 1 1 2 2 6 9 9
8 1 0 6 1 2 3 4 5 5 5 8
6 3 7 4
7 4 8 8
0 2 9
3 0 10
11 0

8.4 Testing difference between two dependent sample means

Two samples are said to be dependent on each other when the elements of one are related to
those of the other in any significant or meaningful manner. In fact the two samples consist of
observations made of the same objects, individuals or more generally, on the same selected
population elements. The ' t ' test is often used to compare ' before ' and ' after ' scores in
experiments for the determination of the significant change that has occurred. For example you
may perform t test to determine: (1) the effect of newly discovered teaching strategy on the
student performance (2) the efficacy of two drugs, (3) the accuracy of the two instruments, and so
on.
To compute the test statistics you will use the following formulas:
1. You will find the difference 𝑑 for each data pair:
𝑑 = 𝑥𝑖 − 𝑦𝑖
2. Determine the mean 𝑑̅ of these differences.

3. The standard deviation of the differences between the paired data entries.

4. Compute the test statistics

The distribution for 𝑑̅ is approximated by a t-distribution when the following requirements are
met:
1. The samples must be randomly selected.
2. The samples must be dependent (paired).
3. Both populations must be normally distributed.
 302

Definition 8.4.1 mean and standard deviation of the Paired Differences for Two Samples
Let 𝒅𝒊 be the distribution of the paired difference for two samples, then the mean and standard
deviation of the distribution 𝒅𝒊 we have,

To make an inference about 𝜇𝑑 you will use the parameter 𝑑̅. The parameter 𝑑̅ is equal to

𝑥̅ − 𝑦̅ while the parameter 𝜇𝑑 is equal to 𝜇1 − 𝜇2.

Definition 8.4.2 Test Statistic for paired differences for two samples
If the sample size is large (i.e 𝒏 ≥ 𝟑𝟎) or the sample sizes is small but normally distributed,
then the t distribution is used to make inference which is calculated as

Most of the research problems 𝜇𝑑 is not known. To solve for 𝑡 you will assumed that the
null hypothesis will be true, then 𝜇1 − 𝜇2 = 0.

Hypothesis Testing procedure for testing the difference between two dependent (paired)
sample means
1. State the hypotheses.
State null (𝑯𝟎) and alternative (𝑯𝟏) hypotheses.
2. Determine the critical value/s.
Two tailed
Divide the alpha value by 2 and then locate the z score value (large
sample) or t value (small sample) that corresponds to the quotient
with the given degrees of freedom.
One tailed
Locate the z score value (large sample) or t value (small sample)
corresponds to the alpha value with the given degrees of freedom.
3. Find test statistics. Compute,

4. Interpret the result. Two tailed test

If 𝒕−𝜶⁄𝟐 < 𝒕 < 𝒕𝜶⁄𝟐 accept the null hypothesis, otherwise reject.
One – tailed test. Accept the null hypothesis if,
𝒕 < 𝒕𝜶 right tailed test
𝒕 > 𝒕−𝜶 left tailed test
 303

Example 8.4.1
An organization believes that the underemployment rate during first quarter is less than
during second quarter. Using the data below, can you conclude that the underemployment rate
during first quarter is lower than during second quarter? Use 𝛼 = 0.05 level of significance.
Year first second
Quarter Quarter
2008 14.8 17.5
2009 20.9 15.6
2010 14.8 15.8
2011 18.1 17.7
2012 13.6 22.3
2013 19.3 20.1
2014 18.0 14.2
Source: Labor Force Survey

Solution
Step 1. State the null hypothesis
𝐻0: The underemployment during first quarter and second quarter is the same. 𝜇1 = 𝜇2 𝐻1:
The underemployment during first quarter is lower than during second quarter. 𝜇1 ≠
𝜇2. Claim.
Step 2. Compute the critical value.
At 𝛼⁄2 = 0.05⁄2 = 0.025, with 𝑑𝑓 = 𝑛 − 1 = 7 − 1 = 6 from the t table the critical
values are ±2.4469.
Step 3. Compute the test value
a. Compute 𝑑𝑖, 𝑑𝑖2, ∑ 𝑑, and ∑ 𝑑2.

first second
𝑑 𝑑2
quarter quarter

b. Find the mean and standard deviation of 𝑑𝑖.

∑𝑑 2 7 + 5 3 + ( 1) + + 3 8 37
 304

c. Compute 𝒕,

Step 4. Interpret the result.

The computed test value −0.30596 is greater than the critical -1.895. This
implies that you must accept the null hypothesis. Thus, the organization claim must be
rejected. Therefore, the underemployment during first quarter is not significantly lower
than during second semester.

Technology Application: Using Microsoft Excel Software Step by Step

Example 8.4.2
Use excel to carry out test of differences (small sample) between the means of the monthly
inflation rate in Pangasinan and Ilocos Norte.

Solution

Step 1. In Cell A1 enter: Quarter 1. In cell B1 enter: Quarter 2. Enter the underemployment rate
data during first quarter and second quarter in the respective columns.

Step 2. Select Data→ Data Analysis →t-Test: Paired Two Sample for Means. Click OK.

Step 3. For Variable 1 Range enter: $A$2:$A$8. For Input Variable 2 Range enter: $B$2:$B$8.
For Alpha enter: 0.05. Lastly, for the output range enter $D$4. Click OK.

t-Test: Paired Two Sample for Means

Variable 1 Variable 2
17.07143 17.6
Mean
 305

Variance 7.319048 7.86

Observations 7 7
Pearson Correlation -0.37664
Hypothesized Mean
Difference 0
Df 6
t Stat -0.30596 t-Value
P(T<=t) one-tail 0.384988
t Critical one-tail 1.94318
P(T<=t) two-tail 0.769976
t Critical two-tail 2.446912 Critical-
Value

The computed t-value give the same result with the test of differences between two
sample means (dependent samples) using excel.

Exercises 8.4
Exercise 1 to 5. Test the claim using t-test of dependent about the mean of the difference of two
populations at the given level of significance. Assume that the data are normally distributed
1. Claim: 𝜇𝑑 ≠ 0; 𝛼 = 0.01. Statistics: 𝑑̅ = 2.4, 𝑠𝑑 = 1.2, 𝑛 = 38
2. Claim: 𝜇𝑑 < 0; 𝛼 = 0.1. Statistics: 𝑑̅ = 11.4, 𝑠𝑑 = 17.5, 𝑛 = 26
3. Claim: 𝜇𝑑 ≥ 0; 𝛼 = 0.05. Statistics: 𝑑̅ = 0.75, 𝑠𝑑 = 2.2, 𝑛 = 10
4. Claim: 𝜇𝑑 = 0; 𝛼 = 0.01. Statistics: 𝑑̅ = 5.62, 𝑠𝑑 = 13.2, 𝑛 = 18
5. Claim: 𝜇𝑑 < 0; 𝛼 = 0.05. Statistics: 𝑑̅ = −1.5, 𝑠𝑑 = 5.13, 𝑛 = 12
Exercises 6 to 9, at 𝛼 = 0.05 determine whether each sample is comparable with each other. The tables
shows the average and standard deviation of the production per semester of livestock in Ilocos Region
from 2002 to 2014.
Source: Philippine Statistics Authority
6. Carabao
Mean Standard
Deviation
Semester 1 4,898 604
Semester 2 4,873 380.62

7. Cattle
Mean Standard
Deviation
Semester 1 14821 1472
Semester 2 14,781 1915.84
 306

8. Hog
Mean Standard
Deviation
Semester 1 38052 1843
Semester 2 38,463 2128.27
9. Goat

Mean Standard
Deviation
Semester 1 5384 223
Semester 2 4623 546.66 Exercises 10 to 11, the data in each problem is
the volume of production of irrigated palay during
first quarter and third quarter from 2000 to 2013 in Ilocos Norte and Ilocos Sur. For each exercises
determine if there is a significant difference between the volume of production of palay during first
quarter and second quarter. Source: Philippine Statistics Authority 10. Ilocos Norte

Quarter 1 Quarter 3
2000 37902 29412
2001 37000 28283
2002 39456 31898
2003 41515 32357
2004 47548 30301
2005 38502 31025
2006 42456 30304
2007 42452 35304
2008 45065 36214
2009 44766 38389
2010 49060 37764
2011 51,789 39,075
2012 49988 39555
2013 54143 53780
2014 53780 33940

11. Ilocos sur

Quarter 1 Quarter 2
2000 14,800 5,082
2001 16,297 9,739
2002 14,968 5,730
2003 7,047 6,279
2004 13,511 9,849
2005 12,525 13,930
2006 14,550 16,454
2007 15,000 10,006
2008 16,328 11,811
2009 17,062 23,757
 307

2010 17,297 12,712

2011 17,528 11,568
2012 19,279 15,450
2013 19,305 11,284
2014 21,092 11,958
Exercise 12 to 15. The data below shows the report of Labor Force Survey about labor and employment
in Ilocos Region from 2003 to 2014. In each event determine if there is significant difference between the
first quarter and second quarter data. Use 𝛼 = 0.01. 12. Number of Labor Force
Quarter 1 3524 3,449 - - - - 3225 3117 2952 1831 2785 2718
Quarter 2 3526 3443 - - - - 3246 3152 2984 2905 2803 2736

13. Labor force participation rate

Q1 61.8 62.1 61 61.7 60.9 62.3 61 61.5 63.1 64.7 62.7 63.9
Q2 61.9 60.8 61.4 61.6 61.2 61 58.7 60.7 63.1 65 63.2 67.4

14. Employment Rate9

Q1 90.7 92.4 91.1 90 92.1 91.5 91.2 90.3 86.5 88.1 88.1 88.8
Q2 90.8 90.6 92 90.92 90.7 91.7 90.8 90.7 86 84.1 88 84.8

15. Unemployment rate

Q1 9.3 7.6 8.9 10 7.9 8.5 8.8 9.7 13.5 11.9 11.9 11.2
Q2 9.2 9.4 8 9.8 9.3 8.3 9.2 9.3 14 15.9 12 15.2

8.5 Testing difference between two proportions

This section explains how to conduct a hypothesis test to determine whether the difference between two
proportions large samples is significant. The two proportion z-test is appropriate when the following
conditions are met:
a. Each datum for each sample is randomly selected.
b. The samples are independent
c. 𝑛1𝑝1 ≥ 5, 𝑛1𝑞1 ≥ 5, 𝑛2𝑝2 ≥ 5, and 𝑛2𝑞2 ≥ 5
Hence, for two large and independent samples of sizes 𝑛1 and 𝑛2, their sample proportions are
(approximately) normally distributed with means 𝑝1 and 𝑝2 and standard deviation and standard deviation
respectively. By means of theorem 5.2, the sampling distribution for 𝑝̂1 − 𝑝̂2
the difference between the sample proportions, is normal distribution with mean
𝜇𝑝̂1−𝑝̂2 = 𝑝1 − 𝑝2
and by means of theorem 5.7, the standard deviation of the sampling distribution for is,

√
Definition 8.5.1 Mean, Standard deviation of sampling distribution 𝒑𝟏 − 𝒑𝟐.
 308

For two large and independent samples, the sampling distribution of the difference between two
proportions 𝒑𝟏 − 𝒑𝟐 is approximately normal, with its mean and standard deviation given as
𝝁𝒑̂𝟏−𝒑̂𝟐 = 𝒑𝟏 − 𝒑𝟐 and
respectively.

Hypothesis testing about 𝑝1 − 𝑝2

A hypothesis test for 𝑝1 − 𝑝2 is based on the assumption that 𝑝1 = 𝑝2. To calculate the test statistics for 𝑝1
− 𝑝2 we can use the formula

(̂ ̂) ( )

When you performed test hypothesis about 𝑝1 − 𝑝2, usually the value of 𝑝1 and 𝑝2 are unknown.
Assuming the hypothesis is true and 𝑝1 = 𝑝2 a common value of 𝑝1 and 𝑝2 denoted by 𝑝̅, is calculated by
using the formula

1 2

Where 𝑥1 = 𝑛1𝑝̂1 and 𝑥2 = 𝑛2𝑝̂2. The value of 𝑝̅ is called the pooled sample proportion. Using the value of
the pooled sample proportion, you can compute an estimate of the standard deviation of as
follows:

Where 𝑞̅ = 𝑝̅ − 1

Definition 8.5.2 Test Statistic 𝒛 for 𝒑𝟏 − 𝒑𝟐

The test statistics z for 𝒑𝟏 − 𝒑𝟐, is calculated as follows:
When are given we have,

When 𝒑𝟏 𝐚𝐧𝐝 𝒑𝟐 are not known we have,

( ) ( )

If the null hypothesis states that 𝑝1 = 𝑝2, 𝑝1 ≤ 𝑝2, or 𝑝1 ≥ 𝑝2, then 𝑝1 = 𝑝2 is assumed and the expression
𝑝1 − 𝑝2 is equal to 0 in the preceding test.

Hypothesis Testing procedure for testing the difference between two proportions
 309

Method 1. Traditional method

1. State the hypotheses.
State null (𝑯𝟎) and alternative (𝑯𝟏) hypotheses.
2. Determine the critical value.
Two tailed
Divide the alpha value by 2 and then locate the z score value
that corresponds to the quotient.
One tailed
Locate the z score value corresponds to the alpha value
3. Find test statistics. Compute,

or

4. Interpret the result. Two tailed test

If 𝒛−𝜶⁄𝟐 < 𝒛 < 𝒛𝜶⁄𝟐 accept the null hypothesis, otherwise reject.
One – tailed test. Accept the null hypothesis if,
𝒛 < 𝒛𝜶 right tailed test
𝒛 > 𝒛−𝜶 left tailed test

Example 8.5.1
The researcher claims that the proportions of teachers in the higher education institution with
M.A/M.S. degree in Region I and II are the same. You randomly select 10 HEI’s in both regions. In
Region
I the selected schools have a total of 303 faculty, 122 of them are M.A./M.S. degree holder. However, in
Region II the selected schools have a total of 454 faculty, 171 of them are M.A./M.S. degree holder
(Based on the report of Private Higher Education Institution). At 𝑎 = 0.05, can you support the claim
that proportions of teacher in HEI’s with M.A./M.S. degree in Region I and II are the same.

Solution
Step 1. State the hypotheses
𝐻0: 𝑝1 = 𝑝2. Claim 𝐻1: 𝑝1 ≠ 𝑝2 Step 2. Determine
the critical value
The test is two-tailed test and the alpha value is 0.05, the critical values are 𝑧 = ±1.96.
Thus the null hypothesis is rejected when 𝑧 does not fall between -1.96 and 1.96. Step 3. Find
the test statistics.
a. Compute 𝑝̅ and 𝑞̅
+ 122 + 171 293
 310

b. Solve

c. Find 𝑧.
122 171

Step 4. Interpret the result

The computed z value was not falls between -1.96 and 1.96. This implies that the null
hypothesis must be rejected.

Therefore, there is enough evidence at the 5% level of significance to support the claim that the
proportion of HEI’s faculty in the regions is the same.
Exercise 8.5
Exercises 1 to 5. Decide whether the normal sampling distribution can be used. If it can be used, test the
claim about the difference between two population proportions 𝑝1 and 𝑝2 at the given level of significance
using the given sample statistics. Assume the sample statistics are from independent, random samples.
1. Claim: 𝑝1 = 𝑝2
Sample statistics: 𝑥1 = 15, 𝑛1 = 108, 𝑥2 = 17, 𝑛2 = 80 𝛼 =
0.01
2. Claim: 𝑝1 > 𝑝2
Sample statistics: 𝑥1 = 37, 𝑛1 = 56, 𝑥2 = 33, 𝑛2 = 60 𝛼 =
0.05
3. Claim: 𝑝1 ≤ 𝑝2
Sample statistics: 𝑥1 = 124, 𝑛1 = 1056, 𝑥2 = 115, 𝑛2 = 800 𝛼 =
0.5
4. Claim: 𝑝1 ≠ 𝑝2
Sample statistics: 𝑥1 = 9, 𝑛1 = 75, 𝑥2 = 15, 𝑛2 = 70 𝛼 =
0.01
5. Claim: 𝑝1 ≥ 𝑝2
Sample statistics: 𝑥1 = 135, 𝑛1 = 276, 𝑥2 = 117, 𝑛2 = 280
𝛼 = 0.1
 311

Exercise 6 to 8. A researcher believes that the proportion of teachers on the year 2012 in SUC’s in Ilocos
Region with Ph.D. degree is the same on the following pair of school. Verify the claim of the researcher
in each pair of schools.
6. Use 𝛼 = 0.05
Baccalaureate M.A./M.S. Ph.D.
Mariano Marcos State University 167 223 71
University of Northern Philippines 182 218 71

7. Use 𝛼 = 0.01
Baccalaureate M.A./M.S. Ph.D.
Ilocos Sur Poytechnic State College 95 42 13
Pagasinan State University 183 279 102

8. Use 𝛼 = 0.01
Baccalaureate M.A./M.S. Ph.D.
Don Mariano Marcos Memorial State 58 95 442
University
North Luzon Philippines State 48 29 10
College

9. The percentage of persons with disabilities in La-Union with 741,906 total populations is 1.8, while
in
Ilocos Sur with 658,587 total population the percentage of persons with disability is 2 as of 2010
(Source: 2007 Census of Population). Is there enough evidence to conclude that the proportion of
adults with disability in the provinces is the same? Use 𝛼 = 0.05
10. According to 2010 Census in Population and Housing, the number of overseas Filipino workers 10
years old and over in Ilocos Norte and Pangasinan is 15,409 and 64,106 respectively. The percentage
of overseas Filipino workers who are high school graduate in Ilocos Norte and Pangasinan is 26.57
and 29.96 respectively. At 𝛼 = 0.05, is there enough evidence to conclude that the proportion of high
school graduate Filipino workers in the provinces is the same?
11. The Private Higher Education Institution reported that the percentage of faculty who are M.A./M.S.
degree hoder in Colegio de Dagupan and Lyceum Northwestern University in the year 2012 is 55.08
and 40.84. This year, there are 118 faculty in Colegio de Dagupan and 191 faculty in Lyceum
Northwestern University. Determine if the percentage of faculty who are M.A./M.S. degree hoder in
these schools is the same. Use 𝛼 = 0.01.
12. Philippine Regulation Commission reported that the percentage of passers at Mariano Marcos State
University in Nursing and LET-Elementary in the year 2012 is 92.22 and 88.65 respectively. In these
disciplines the school score higher than the national percentage of passers. There are 270 and 229
Nursing and Bachelor of Elementary Education takers respectively in this school. Determine if the
percentage of passers in the nursing examination and LET-Elementary Education is the same. Use 𝛼
=
0.05.
13. A researcher claims that the percentage of male and female Pilipino (5 years old and over) in
Pangasinan with difficulty in remembering or concentrating is the same. The percentage of male and
female Pilipino (5 years old and over) in this province with difficulty in remembering or
concentrating is 13.35 and 15.19 respectively. There are 36,806 and 44,830 male and female Pilipinos
 312

in this province with at least one type of functional disability. Use 𝛼 = 0.05, is there enough evidence
to accept the claim of the researcher? Source: Census of Population and Housing 2010
14. An area supervisor believes that the proportion of elderly (60 years old and over) female Pilipino
people in La Union in the year 2010 and 2007 is the same. The proportion of female Pilipino people
in this province in the year 2010 and 2007 is 0.5839 and 0.5781 respectively. There are 38,827 and
34,762 elderly female Pilipino people in this province on the year 2010 and 2007 respectively. At 𝛼 =
0.01, is there enough evidence to accept the claim of the area supervisor? Source: 2010 Census of
Population and Housing and 2007 Population Census.
15. An enumerator claims that the unemployment rate in the first quarter of the year 2013 is lower than in
the first quarter of 2014. The unemployment rate in the first quarter of 2013 and 2014 is 7.6 and 9.3
respectively. There are 3,449 and 3,524 labor forces on the first quarter of the year 2013 and 2014
respectively. At 𝛼 = 0.01, is there enough evidence to accept the claim of the enumerator? Source:
Labor Force Survey

Chapter 8 Exercises
Exercise 1 to 2. Test the claim about the variance of the two populations at the given level of significance.
Assume that the data are normally distributed.
1. Claim: 𝑠1 = 𝑠2; 𝛼 = 0.05. Statistics: 𝑠1 = 3.43, 𝑛1 = 14, 𝑠2 = 2.54, 𝑛2 = 10.
2. Claim: 𝑠1 ≠ 𝑠2; 𝛼 = 0.01. Statistics: 𝑠1 = 17.23, 𝑛1 = 8, 𝑠2 = 19.14, 𝑛2 = 12.
Exercise 3 to 10. Test the claim about the difference of two populations at the given level of significance.
Assume that the data are normally distributed.
3. Claim: 𝜇𝑑 ≥ 0; 𝛼 = 0.05. Statistics: 𝑑̅ = 12.68, 𝑠𝑑 = 2.27, 𝑛 = 24.
4. Claim: 𝜇𝑑 ≠ 0; 𝛼 = 0.01. Statistics: 𝑑̅ = 56.78, 𝑠𝑑 = 18.96, 𝑛 = 16.
5. Claim: 𝑝1 = 𝑝2; 𝛼 = 0.1. Sample statistics: 𝑥1 = 23, 𝑛1 = 68, 𝑥2 = 28, 𝑛2 = 82.
6. Claim: 𝑝1 < 𝑝2; 𝛼 = 0.05. Sample statistics: 𝑥1 = 12, 𝑛1 = 45, 𝑥2 = 53, 𝑛2 = 123.
7. Claim: 𝜇1 ≠ 𝜇2, 𝛼 = 0.01. Sample statistics: 𝑥1̅ = 46.67, 𝑠1 = 7.53, 𝑛1 = 18 and 𝑥̅2 = 52.74, 𝑠2 =
6.32, 𝑛2 = 23. Assume 𝜎12 ≠ 𝜎22.
8. Claim: 𝜇1 ≤ 𝜇2, 𝛼 = 0.10. Sample statistics: 𝑥1̅ = 0.956, 𝑠1 = 0.231, 𝑛1 = 12 and 𝑥̅2 = 1.28, 𝑠2 =
0.271, 𝑛2 = 10. Assume 𝜎12 = 𝜎22.
9. Claim: 𝜇1 = 𝜇2, 𝛼 = 0.05. Sample statistics: 𝑥1̅ = 117.83, 𝑠1 = 31.29, 𝑛1 = 46 and 𝑥̅2 =
136.76, 𝑠2 = 18.79, 𝑛2 = 54.
10. Claim: 𝜇1 ≥ 𝜇2, 𝛼 = 0.05. Sample statistics: 𝑥1̅ = 65.72, 𝑠1 = 13.81, 𝑛1 = 54 and 𝑥̅2 = 57.63, 𝑠2 =
18.92, 𝑛2 = 38.
Exercise 11 to 24, Perform the hypothesis testing procedure for each problem.
11. Paul believes that the variance of the real wage rate of male palay workers and corn workers is the
same. The data below is the real wage rate male palay and corn workers from year 2000 to
2011. At 𝛼 = 0.05, verify the claim of Paul. Source: Bureau of Agriculture
Palay Workers Corn Workers
168.72 157.71 154 155.23 132.42 124.08 116.62 121.18
150.22 144.4 138.37 148.45 121.87 123.76 119.31 122.58
158.53 158.38 160.9 165.37 121.6 126.76 132.8 133.97
 313

12. Raphael claims that the variance of the number of Filipino emigrants in Bantay and Candon
Ilocos Sur are not the same. The data below shows the number of Filipino emigrants in the said
municipalities from 1988 to 2012. Is there enough evidence to accept the claim of Raphael? Use
Source: http://www.cfo.gov.ph/
13. A field supervisor claims
Bantaythat the distribution of the monthly purchasing
Candon power of peso in La-
Union and
7 2 9 Ilocos Sur is the
6 4 3 0 5 9 same. You
9 8 8 7 0 4 0 8 randomly
5 4 4 3 5 0 2 8 selected 15
samples for
6 6 6 4 3 1 1 6 2 3 5 6 each province.
7 5 7 2 4 5 6 7 At 𝛼 = 0.05, is
8 8 2 7 there enough
9 9 2 3 6 6 evidence to
9 10 1 support the
supervisors
9 11
claim. Source:
National Statistics office.
La-Union Ilocos Sur
0.73 0.73
0.74 0.74
0.74 0.74
0.76 0.78
0.76 0.78
0.76 0.78
0.76 0.77
0.75 0.76
0.75 0.76
0.75 0.76
0.74 0.75
0.78 0.79
0.77 0.8
0.77 0.8
0.75 0.78

14. Cassava is a perennial plant that grows best under tropical, moist, fertile, and well-drained soils.
Cassava has more protein than that of other tropical foods sources like yam, potato, etc. The mean
and standard deviation of the price of cassava per kilogram in Pangasinan during summer (March
to May) and rainy season (June to October) from 2000 to 2010 is shown below. At 𝛼 = 0.05, is
the price of cassava per kilogram during summer differ from price during rainy season.
Summer Rainy Season
Mean 5.93 5.62
Standard Deviation 2.32 1.03
Source: Philippine Statistics Authority
 314

Exercise 15 through 18. An enumerator claims that the monthly retail price per kilogram of Camote tops
[Sweet potato] and Kangkong is the same. The data in each table is the average monthly retail price of
Camote tops [Sweet potato] and Kangkong from January 2010 to May 2015. At 𝛼 = 0.05, is there enough
evidence to support the enumerators claim in each provinces?
Source: Philippine Statistics Authority
15. Ilocos Norte 16.
Mean Standard deviation
Camote tops 32.17 4.09
Kangkong 26.20 3.31

17. Ilocos Sur

Mean Standard deviation
Camote tops 22 5.47
Kangkong 21.96 4.73

La-Union 18.
Mean Standard deviation
Camote tops 43.35 5.77
Kangkong 41.19 6.62

19. Pangasinan Matthew claims that the yearly number

Mean Standard deviation Filipino emigrants in San Juan (Lapog)
Camote tops 20.35 4.59 Ilocos Sur is greater than in Sto. Domingo
Kangkong 23.56 6.87 Ilocos Sur. The data below shows the yearly
number of Filipino emigrants in the said
municipalities from 1988 to 2012. At 𝛼 = 0.05, verify the claims of Matthew.
San Juan (Lapog) Sto. Domingo

4 2 6 8

9 7 5 1 0 3 0 5 9 9

9 6 6 5 2 1 1 4 0 1 1 2 2 3 3 3 3 5 5
9 7 6 6 4 2 1 5 3 6 9

2 2 0 6 0 1

8 3 7 0 4

8 1

Source: http://www.cfo.gov.ph/

Exercise 20 through 22. A group of students believe that the level of implementation of the Pantawid
Pamilyang Pilipino program along the given areas as perceived by the respondents according to their sex
is the same. The data below shows the level of implementation of the Pantawid Pamilyang Pilipino
program as perceived by the members of the 4P’s enrolled in the municipality of Bayambang. The data is
 315

gathered using a 3-point likert scale questionnaire. Verify the claim of the students at the given level of
significance?
Source: Kimberly V. Cacabilos, Orlando F. Tagum Jr. Beverly Q. Miranda, Marife B. Padilla.
Implementationof Pantawid Pamilyang Pilipino Program In Bayambang.
20. Social Development. Use 𝛼 = 0.05.
Male 2.82
Female

21. Emergency Assistant Program. Use 𝛼 = 0.01.

Male 2.2 2.8 2.8 2.8 2.8 2.4 3 3 2.8 2.8 2.8
Female 3 3 3 3 2.4 2.6 2.8 2.6 2.8 2.4 2.6 2.6 2.4 2.6
22. Nutrition Program. Use 𝛼 = 0.10.
Male 3.00 3.00 3.00 2.33 3.00 2.67 2.33 2.67 2.67 2.67
Female 3.00 3.00 3.00 3.00 3.00 3.00 2.67 2.67 3.00 3.00 3.00 3.00
23. Luke claims that the average retail price per semester of ripe mango carabao in Ilocos Region in
1st semester is less than in the 2 nd semester. The data below is the average retail price per semester
of ripe mango carabao from year 2002 to 2014. At 0.01, is there enough evidence to support the
claim of Luke?

1st Semester 2nd Semester

39.57 76.95
38.34 77.87
40.55 70.44
46.75 67.91
37.97 87.4
51.92 101.07
54.88 108.67
52.87 113.13
57.09 96.49
58.28 77.82
49.21 87.81
55.49 88.26
Source: Philippine Statistics Authority